TCPA Emergency Purposes Exception: Which Alerts Qualify

“The term emergency purposes means calls made necessary in any situation affecting the health and safety of consumers,” according to the FCC’s rule at 47 C.F.R. § 64.1200(f)(4). A call or text that fits needs no consent under the TCPA’s robocall provisions. Safety recalls and school closures can fit. Sales messages don’t, and alerts that keep coming after a stop request are vulnerable. A company that sent a genuine safety alert to the right people often has a strong defense.

What does the emergency purposes exception cover?

It covers robocalls and automated texts made necessary by a situation affecting health and safety. Whether the recipient also has to be connected to that situation is contested, as the court decisions below show.

The statute bars any call “using any automatic telephone dialing system or an artificial or prerecorded voice” to a cell phone, with a carve-out for any “call made for emergency purposes,” under 47 U.S.C. § 227(b)(1)(A). Prerecorded calls to home landlines get the same carve-out, since the statute bars them “unless the call is initiated for emergency purposes,” per § 227(b)(1)(B). Congress never defined the phrase. The FCC did, in the one sentence quoted above.

An autodialer is equipment that has the capacity to store or produce numbers using a random or sequential number generator, and to dial them. A prerecorded call is one that plays a recorded or artificial voice instead of a live person. The exception only matters when one of those is in play. Whether a texting platform even counts as an autodialer is its own fight, which I covered in our post on autodialer claims after Duguid.

The exception doesn’t reach the Do Not Call rules, but it rarely needs to. Those rules govern telephone solicitations, and a pure safety alert isn’t one.

What did the FCC say about school alerts and utility outages in 2016?

It said safety alerts from schools are emergencies and ordinary school news isn’t. It put utility outage calls on a consent footing without deciding whether they’re emergencies.

The ruling is the FCC’s Blackboard and Edison declaratory ruling, FCC 16-88, 31 FCC Rcd 9054, adopted July 8 and released August 4, 2016. The Commission wrote: “We confirm that school callers may lawfully make autodialed calls and send automated texts to student family wireless phones without consent for emergencies including weather closures, fire, health risks, threats, and unexcused absences.”

In the same ruling, the FCC agreed that “the mere fact that an informational message comes from a school caller does not make it an emergency.” Those calls need consent. The FCC clarified that a parent or student who gives a school only a wireless number as a contact has consented to calls closely related to the school’s educational mission or official activities, absent instructions to the contrary.

Footnote 76 of the ruling adds two limits that plaintiffs now cite well outside the school setting. Third parties sending alerts with a school are covered only “as long as the messages are limited to the emergency at issue and do not include any marketing.” And the FCC wrote, “These calls must be about a bona fide emergency that is relevant to the called party.”

Utilities got relief on a different footing. The utility trade groups narrowed their request to consent, and the FCC said in the 2016 ruling that it did “not reach the question” of whether utility calls fall within the exception. It did repeat its 1992 view that “[s]ervice outages and interruptions in the supply of water, gas or electricity could in many instances pose significant risks to public health and safety.” It then clarified that customers who give a utility their wireless number consent to calls “closely related to the utility service,” including outage warnings and restoration updates, unless they’ve given instructions to the contrary.

That consent has a catch. The FCC wrote that “the utility company will bear the burden of showing it obtained the necessary prior express consent,” according to the ruling. A utility that can’t produce the record of how it got a customer’s number is back to arguing emergency, which is a harder case for anything short of an actual outage.

Does the 2020 COVID-19 ruling still protect pandemic messages?

Only narrowly, and I wouldn’t build a 2025 calling program on it.

On March 20, 2020, the FCC’s Consumer and Governmental Affairs Bureau issued a declaratory ruling, DA 20-318, confirming that the pandemic was an emergency under the TCPA. It set a two-part test. First, the caller had to be “from a hospital, or be a health care provider, state or local health official, or other government official as well as a person under the express direction of such an organization and acting on its behalf.” Second, “the content of the call must be solely informational, made necessary because of the COVID-19 outbreak, and directly related to the imminent health or safety risk arising out of the COVID-19 outbreak.”

The Bureau named what fails. Its ruling excluded calls “advertising a commercial grocery delivery service, or selling or promoting health insurance, cleaning services, or home test kits,” and it excluded debt collection, “even if such debt arises from related health care treatment.” A July 28, 2020 public notice, DA 20-793, added plasma-donation outreach to people who had tested positive, by health care entities acting “pursuant to guidance from federal, state, or local government officials.” In between, the Bureau’s June 2020 ruling on Anthem’s petition, DA 20-669, refused to create an “urgent circumstances” exemption and doubted that calls to “educate members about available services and benefits” were urgent at all.

Three things limit those rulings today. They came from a bureau, not the full Commission. The test leaves most private businesses out, because a retailer or employer isn’t on the caller list. And the national emergency the March ruling cited is over: Congress ended it in Public Law 118-3, which states that the March 13, 2020 national emergency “is hereby terminated,” approved April 10, 2023.

Which alerts have courts accepted, and which have they rejected?

Courts look at what the message was for and who received it. At the pleading stage, several have refused to apply the exception to messages that allegedly continued after a stop request.

Recalls. A federal court in San Diego dismissed a proposed class action over an automated recall call about ground beef. In Brooks v. Kroger Co., the court found the complaint’s marketing allegations conclusory, and in denying leave to amend it wrote that “Kroger had a bona fide emergency in its tainted and potentially life-threatening beef, and thus called potential consumers of that beef to warn them.” The order rejected the argument that a recipient must face direct harm, writing that “there is no statutory or legal justification to read the exception so narrowly.” The same order also read the FCC’s footnote 76 narrowly, concluding that “that language was not meant to be applied outside of schools,” and it treated the Ninth Circuit’s contrary use of that footnote in Duguid, discussed next, as dicta. That reading is contestable inside the Ninth Circuit. Still, I think Brooks is the most useful recall ruling a defendant can cite, as long as the recall went to people who plausibly bought or used the product.

Fraud and security alerts. The Ninth Circuit rejected Facebook’s emergency defense at the pleading stage in a suit over login-alert texts sent to a plaintiff who alleged he had no Facebook account. In Duguid v. Facebook, Inc., 926 F.3d 1146 (9th Cir. 2019), the court reasoned that on those allegations the plaintiff’s “account could not have faced a security issue,” so the texts fell outside the exception, relying on the FCC’s footnote 76. The Supreme Court reversed that decision in 2021 on the autodialer definition and didn’t address the emergency point. Financial institutions have a separate fraud-alert exemption for calls and texts to wireless numbers in § 64.1200(a)(9)(iii). It applies only if the message is free to the recipient, and it requires that alert texts and calls “must not include any telemarketing, cross-marketing, solicitation, debt collection, or advertising content.” The FCC handled health care calls the same way, through a separate conditioned exemption, which is a different analysis from this one. Our post on the financial institution exemption for fraud alerts covers its conditions in detail.

Public-health messages. A federal court in San Francisco rejected a pharmacy’s argument that prescription calls are always emergencies. Denying CVS’s motion to dismiss in St. Clair v. CVS Pharmacy, Inc., 222 F. Supp. 3d 779 (N.D. Cal. 2016), the court reasoned that “Continuing to call a customer (or former customer) about a prescription, when the customer has made clear that he does not want or need the calls,” falls outside the exception. In Florida, a court let a case proceed over texts about free COVID-19 testing that, according to the complaint, kept arriving after the recipient replied STOP. The order in Farhat v. Unique Healthcare Systems, LLC (M.D. Fla. Jan. 11, 2022) agreed with courts holding that a business can’t invoke the exception “if the defendant continues to send messages after the plaintiff has instructed the defendant to stop.” That was a pleading-stage ruling, and the parties later stipulated to dismissal with prejudice. St. Clair distinguished an earlier Missouri case, without endorsing it, in which five prescription calls “fell within the exemption at the summary judgment stage” after a review of the evidence.

Why does a marketing line cost you the exception?

Because the exception protects messages that are necessary for safety, and a sales pitch isn’t necessary for anyone’s safety.

The FCC’s school ruling limited third-party alerts to those that “do not include any marketing,” and its COVID ruling put advertising and telemarketing calls outside the exception by name. Once a message promotes something, it is also telemarketing, so the business needs signed written consent for a robocall and has to clear the Do Not Call rules for the number. One coupon in a recall text can turn a defensible alert into an unconsented telemarketing message, at $500 per message, or up to $1,500 if a court finds the violation willful or knowing.

The usual overreach is easy to picture. A safety notice goes out cleanly to a list of affected customers, and then a follow-up adds a “shop now” link or a discount on the replacement product. In my view, that follow-up is the riskiest thing a compliance team can approve. Expect a plaintiff to argue it colors the first message too, even though none of the decisions above adopts that theory. Keep the alert boring. Send the promotion later, through a separate lane, to people who consented to marketing.

Exposure for alerts at a glance

One more caution about every row that leans on an FCC ruling. The Supreme Court heard argument on January 21, 2025 in McLaughlin Chiropractic Associates v. McKesson Corp., and its question presented is “Whether the Hobbs Act required the district court in this case to accept the FCC’s legal interpretation of the Telephone Consumer Protection Act.” A decision is pending. Whatever the Court does, I’d want a defense that fits both the statute and the FCC’s definition over one resting on a bureau-level ruling alone.

When an alert triggers a demand letter

1. Put a litigation hold on the alert campaign, including the message files and the time-stamped send logs for the list the alert went to.
2. Pull the documents showing the emergency itself, such as the recall notice or the outage report, dated before the send.
3. Pull every opt-out and stop reply tied to the claimant’s number and the dates your system honored them.
4. Collect the vendor contracts and platform settings for the alert, and any Registry scrub records if the message touched sales.
5. Talk to a lawyer before replying to the letter or offering the claimant a credit or refund to close it out.

The firm’s page on TCPA litigation defense explains how claims involving automated safety alerts and recall notices move from a demand letter to a lawsuit.

Questions businesses ask about emergency alerts

Do we need consent to send a recall notice by text?

Probably not, if the recall involves a real safety risk and the text reaches people connected to the product, such as registered owners or customers who bought it. The Brooks court treated a contaminated-food recall call as an emergency. A recall for a cosmetic defect is different, since it doesn’t affect health or safety.

Can a customer opt out of emergency messages?

The statute doesn’t require consent for emergency calls, so there’s technically nothing to revoke. Some courts treat a stop request as a sign that further messages weren’t necessary, though. St. Clair and Farhat both refused to dismiss claims over messages allegedly sent after a stop request. I’d honor stop requests for routine alerts and save repeat sends for true, immediate hazards.

Are weather or outage alerts from a property manager emergencies?

They can be, if the alert warns residents about a hazard to them, like a gas leak or a building-wide power failure. The FCC has said outages can put public health and safety at risk. A notice about a scheduled amenity closure isn’t an emergency. I’d still collect each resident’s number with a disclosure that alerts will come by text, so consent backs you up either way.

Does the emergency exception apply to calls made by our vendor?

It can. The statute asks what the call was for, and the FCC’s school ruling covered certain third parties working with schools. The COVID ruling was stricter about who could call, so I wouldn’t assume every vendor qualifies. The vendor’s message still has to be limited to the emergency and sent to the right people.

If your company is being sued over an alert

If your company is facing a TCPA claim over a recall, outage, or safety message, the dated record of the emergency and the list the message went to will be at the center of the defense. Rosenberger + Kawabata represents businesses defending TCPA lawsuits and class actions through its business litigation practice, including claims involving automated safety alerts and recall notices. Contact Rosenberger + Kawabata online for a free and confidential consultation, or call (310) 894-6921.

Sources

The statute is the Telephone Consumer Protection Act at 47 U.S.C. § 227. The FCC’s rules are at 47 C.F.R. § 64.1200, as published in the eCFR through May 1, 2025. The 2016 school and utility ruling is the Blackboard and Edison declaratory ruling, FCC 16-88, 31 FCC Rcd 9054 (2016). The pandemic rulings are the COVID-19 declaratory ruling, DA 20-318, 35 FCC Rcd 2840 (CGB Mar. 20, 2020), the Anthem ruling, DA 20-669 (CGB June 25, 2020), and the plasma-donation public notice, DA 20-793 (CGB July 28, 2020). The end of the COVID-19 national emergency is in Public Law 118-3 (2023).

The court decisions are Duguid v. Facebook, Inc., 926 F.3d 1146 (9th Cir. 2019), rev’d on other grounds, 592 U.S. 395 (2021); St. Clair v. CVS Pharmacy, Inc., 222 F. Supp. 3d 779 (N.D. Cal. 2016); Brooks v. Kroger Co., No. 3:19-cv-00106 (S.D. Cal. Aug. 12, 2019); and Farhat v. Unique Healthcare Systems, LLC, No. 8:21-cv-1319 (M.D. Fla. Jan. 11, 2022). The pending Supreme Court case is McLaughlin Chiropractic Associates v. McKesson Corp., No. 23-1226.

This post is general information about the Telephone Consumer Protection Act and related laws as of May 1, 2025. It is not legal advice, and reading it does not create an attorney-client relationship. Courts and the FCC change the rules in this area often, and the outcome of any claim depends on its facts. Past results do not guarantee future outcomes.