On Friday March 16, 2018 the D.C. Circuit issued its long awaited decision on various challenges to the Federal Communication Commissions (FCC) 2015 Omnibus Declaratory Ruling and Order (Omnibus Order) on the Telephone Consumer Protection Act (TCPA), based on claims the FCC exceeded its authority and was capricious and arbitrary when issuing its rulings. The decision was a mixed bag, vacating the FCC’s expansive approach to the “capacity” of a device to store or generate and dial numbers in the context of defining an autodialer under the statute, and the FCC’s “one free call” safe harbor for reaching the “called party” at a reassigned telephone number. On the other hand, it upholds a call recipient’s right to revoke previously granted consent by any reasonable means, and the scope of the FCC’s exemption from TCPA liability for calls and text messages regarding urgent healthcare matters. The court did not establish definitive alternative standards for autodialers or reassigned numbers, so the vacated rulings on those points effectively return matters to pre-2015 uncertainty.
In vacating the FCC’s approach to clarifying what constitutes an autodialer, the court starts, as it must, with the statutory definition (paraphrased above), which “naturally raises two questions: (i) when does a device have the ‘capacity’ to perform the two enumerated functions; and (ii) what precisely are those functions?” After exploring those questions, the court concludes the FCC’s approach cannot be sustained, at least given its assumption that a call made with a device having the capacity to function as an autodialer can violate the statute even if autodialer features are not used to make the call.
The court downplays the “present capacity” vs. “potential capacity” question, in favor of examining “capacity to do what?” and how much effort is required to enable that capacity. Ultimately, the court spent much ink addressing an issue that many pointed out in the Omnibus Order proceeding, and which the FCC recognized, that the FCC’s expansive definition means that smartphones can be autodialers (or maybe even are autodialers, if all that is needed under the statute/rules is to install an app). The FCC had declined to resolve that question and instead said it would revisit it if unwanted calls from smartphones became an issue.
But the court wasn’t persuaded to wait. It speaks at great length about how the FCC’s position makes smartphones autodialers, and how Congress never intended to regulate devices that the vast majority of the public uses on a daily basis as equipment that could create liability for statutory damages. This vast disconnect, between number of industry users of “autodialers” that Congress believed itself to be regulating, and the hundreds of millions of “everyday callers” the FCC brought within the rules, indicated the FCC had strayed far beyond its statutory authority.
The FCC’s ruling was also arbitrary and capricious, the court held, because it “offered no meaningful guidance to affected parties” as to what is allowed and what is prohibited. Among other things, it left completely unclear not only the above-noted point about how much is required to enable capacity to function as an autodialer, but also the extent to which dialing from a list, as opposed to numbers the device itself generates, factors into the analysis. The FCC’s ruling also was unclear, the court said, because it refused to answer whether human intervention (in dialing) is dispositive, and if not, to what degree it must be present, while also holding (based on “potential capacity”) that a device can still be an autodialer even with some human intervention.
Despite thoroughly rejecting the FCC’s approach, the court did not decide what the autodialer definition encompasses (or excludes). The closest it gets to providing guidance is what might be viewed as endorsement of a passage in Commissioner O’Rielly’s dissent from the Omnibus Order that, rather than concentrating on the autodialer definition, the focus should be on the autodialer prohibition, where it restricts the use of an autodialer to “make any call.” This, Commissioner O’Rielly urged, means “that the equipment must, in fact, be used as an autodialer to make the calls” before a TCPA violation can be found. As the court observed, this interpretation “would substantially diminish the practical significance of the  expansive understanding of ‘capacity,’” because “[e]ven if the definition encompasses any device capable of gaining autodialer functionality through the downloading of software, the mere possibility of adding those features would not matter unless they were downloaded.” This in turn would mean that “everyday calls made with a smartphone would not infringe the statute” because the ability to configure a smartphone to be an autodialer would not matter “unless the relevant software were in fact loaded onto the phone and were used to initiate calls or send messages.”
Nonetheless, the issue is returned to the FCC, should it wish to revisit it.
So Where Does This Leave Us?
The D.C. Circuit leaves matters, for the most part, in no better position on the autodialer issue than before the Omnibus Order. The FCC had stated on several occasions that “predictive dialers” are autodialers, but that determining whether other dialing technologies, such as “preview dialers” (where agents are presented with an onscreen file and must take an action to initiate the call), are autodialers should be left to courts on a case-by-case and device-by-device basis. Those cases had produced mixed results. As a practical matter, this lack of definitive guidance presented a challenge for defendants hoping to quickly escape lawsuits by arguing they did not use an autodialer, forcing them to go through discovery and summary judgment and, if facts regarding the dialer remained in dispute, to trial.
Although the D.C. Circuit’s ruling vacating the FCC’s attempt to provide clarity on the autodialer definition did not articulate clear parameters, it gave some degree of guidance to future courts that the FCC’s approach was overly expansive, which may lead courts to take a narrower view than some perhaps had in the past (and certainly where they applied the FCC’s ruling). It also reinstates the ability to argue that the statutory provision means equipment must, in fact, actually be used as an autodialer, regardless of its other capabilities. And for now it is still the case that the hallmarks remain the extent to which a given device can both store/produce numbers and dial them, without human intervention, and what it means for the device to have the “capacity” to do so.
The TCPA prohibits certain telephone calls to be made without the “called party’s” consent, where the level of and/or need for consent varies depending on whether the call is or is not for marketing purposes, and whether it is made to a cellphone or landline. While seemingly straightforward, this definition begs the question: Who is the “called party?” The two options are the intended recipient of the call (from whom consent may have been previously obtained), or the actual recipient, who may be someone to whom a phone number has been reassigned and thus never consented. If the former, the consent obtained from that person renders calls requiring consent permissible; but if the latter, callers/texters may proceed thinking they have consent, only to find they’ve reached a non-consenting new subscriber and thus face liability.
In the Omnibus Order the FCC determined that the “called party” is the person actually reached, even if the number has been reassigned and regardless whether the caller/texter knows that. The Order also created a “safe harbor” for the first call/text to the reassigned party – but for only that one call, regardless whether it connects and/or leads to actual knowledge of the reassignment. The FCC took this approach because it noted there must be some level of “reasonable reliance” on consents obtained, and so that there would not be “strict liability” for callers, who have no chance to learn of the reassignment. But this left the possibility – likelihood, in reality – that that first call would not necessarily alert the caller/texter to the reassignment.
Although the court found the FCC’s actual-party approach to be a permissible interpretation of the statute, it nevertheless vacated the ruling in full because “one free call” is only an illusory “safe harbor,” and runs counter to the avoidance of strict liability with which the FCC justified it. The court found the one-free-call safe harbor arbitrary and capricious since it serves no purpose in ensuring the caller/texter learns of the reassignment, and as a result, does not avoid strict liability.
The court does not, however, impose any rule of its own, or suggest what the right answer is under an actual-party regime. Instead, it vacates the ruling in its entirety. It did so because leaving the actual-party rule in place, without any safe harbor, would impose strict liability for any call to a reassigned number, and the Commission expressly avoided adopting such a rule. So this, too, sends us back to the drawing board, though the court does note the reassigned number problem may ultimately be addressed by a mechanism such as a reassigned number database, which is currently the subject of an ongoing FCC rulemaking.
So Where Does This Leave Us?
Even with this decision, the reassigned number issue remains a problem. Before the Omnibus Order, the debate was over whether “consent of the called party” meant the party who had given the consent/number to the caller/texter, i.e., the “intended recipient,” or the party who actually receives a particular call/text, i.e., the “actual recipient.” Courts tended to lean toward actual-recipient, consistent with the FCC’s ensuing ruling in the Omnibus Order.
Other than holding that the actual-recipient approach is defensible, the Court did not endeavor to say what the right answer is on potential liability for calls/texts to reassigned numbers in a consent-based environment. In the meantime, and especially inasmuch as it seems few relied on “one free call” as anything other than a defense on the rare occasions it applied, the industry remains in search of an effective solution for avoiding potential liability for calling reassigned numbers. Holding the most promise at this time appears to be development of a central reassigned number database that while not mandatory, would provide a useful safe harbor for callers.
CONSENT REVOCATION The court’s affirmance of the FCC on the issue of consent revocation means the status quo over the last three years since the Omnibus Order remains intact. In the absence of the TCPA expressly allowing call recipients to revoke previously granted consent, the FCC held in the Omnibus Order that consent can be revoked by any reasonable means, written or oral, that clearly expresses the consumer’s desire and that is reasonably expected to communicate that desire to the party to whom consent had previously been given. Noting there was no dispute (even among the FCC and those who appealed) that consent can be revoked for calls that are otherwise forbidden by the TCPA or FCC rules, the court held the FCC’s ruling was reasonable, especially inasmuch as it relieves callers/texters from having to shoulder any “undue burden” to adopt systems of receiving/accepting/implementing consent revocations. Significantly, decisions by courts on this point in the wake of the FCC’s Omnibus Order have been fairly reasonable, holding, for example, that simply saying something to a run-of-the-mill retail employee will not suffice, that equivocation in expressing the intent or extent of withdrawal can render it ineffective, and that trying to revoke consent for texts by replying in ways that eschew clear instructions from the sender on how to opt-out are not “reasonable” withdrawals of consent.
This challenge related solely to the FCC’s exemption for certain urgent/exigent healthcare messages as it relates to HIPAA, the FCC’s 2012 allowance for autodialed/prerecorded calls/texts by HIPAA “covered entities,” and the TCPA’s “emergency purposes” exception for autodialed/prerecorded calls/texts. Included among the exempted calls were appointment and exam confirmations and reminders, wellness checkups, hospital preregistration instructions, preoperative instructions, lab results, post-discharge follow-up intended to prevent readmission, prescription notifications, and home healthcare instructions. Expressly carved out from the exemption were calls “that include telemarketing, solicitation, or advertising content, or which include accounting, billing, debt-collection, or other financial content.” To take advantage of the exemption the call/text must satisfy various criteria, including being free to end users and not exceeding particular frequency and content limitations.
Petitioners argued the exemption’s inclusions and exclusions were arbitrary and capricious or in the alternative, conflict with HIPAA. The court rejected these positions, noting that “[t]here is no obstacle to complying with both the TCPA and HIPAA” because “the two statutes provide separate protections.” Also rejected was petitioners’ argument that a narrower exemption for healthcare calls made to wireless numbers was “inherently contradictory” because the TCPA treats wireless numbers differently than residential numbers and that “[t]he statute itself contemplates that calls to wireless numbers tread [more] heavily upon ... consumer privacy interests.” As with consent revocation, this ruling simply leaves in place the state of affairs that has existed on these points since the Omnibus Order issued.
The FCC now has some work ahead of it, and we are left with the kind of uncertainty we had before the Omnibus Order regarding what constitutes an autodialer and what to do about reassigned numbers. We are aided, however, by a repudiation of an overly broad approach to autodialers, reinstatement of the ability to argue that “present capacity” should be the rule, and the prospect that the FCC (or courts) may look to other statutory interpretations that remove the focus from the “capacity” question. And the FCC’s rulemaking on a potential reassigned numbers database takes on added importance. Meanwhile, consent-revocation remains exactly as it has been for the nearly last 3 years, as does operation of the health-care exemption.