Defendants market online programs that offer discounts in exchange for a “membership fee.” Brian Schnabel was enrolled in one such program, Great Fun, after making a purchase on Priceline.com, and two years later his father Edward was enrolled in Great Fun after making a purchase on Beckett.com, a sports memorabilia site. Neither, they alleged, intentionally or knowingly enrolled, but they were both presented with “enrollment offer” pages and entered personal information on those pages:
Edward alleged that, at the time he thought Beckett.com was collecting his information, not a third party, and Brian alleged the same. Trilegiant has a record of Brian subscribing to their service under the username “SCHNABEL22.” The confirmation page allegedly similar to what Edward saw says “your Online Price Guide subscription has also been sent to [your email address]”; and features, below the hyperlink “Click here to claim up to $20.00 Cash Back on this purchase!”, a “button” titled “See Details” with a legend beneath reading: “Click above to learn how to get $20 Back from Great Fun.” But Great Fun is not further identified on the order confirmation page.
Trilegiant alleged that only by clicking on “See Details” or “Learn More” would Edward and Brian have been brought to the enrollment page. It wasn’t clear from the record whether they could have enrolled without ever seeing the enrollment pages by, for example, clicking instead on “Click here to claim up to $20.00 Cash Back on this purchase!” Because, the court concluded, even if they did see the enrollment pages no binding arbitration agreement was formed, the court didn’t need to resolve this question.
According to Trilegiant, neither could join Great Fun without affirmatively entering personal information including “city of birth” and a password. However, they weren’t required to reenter credit-card information when signing up for Great Fun; Beckett and Priceline passed that on. Indeed, the Beckett page said it was offering a “Special Award for Beckett Customers.” “Toward the bottom of the page, near an overview of some of the ‘Benefits’ of the program, though, there do appear the logos of several popular brands besides Beckett, suggesting that by accepting the offer, the purchaser will somehow be able to receive discounts when purchasing other goods or services.” The message also promised savings at participating restaurants and “top attractions and activities.” Small print said that “[t]here's no obligation to continue ... Great Fun benefits.... [The purchaser can] call us to cancel before the end of ... [the] FREE trial and owe us nothing[.]”
In addition, Trilegiant argued that it customarily emailed each newly enrolled member its terms and conditions; if the email bounced, it would send a paper version to the member’s billing addressed. Edward acknowledged receiving several emails from Great Fun, but Brian denied it. The court determined that this didn’t matter because even if they did receive the emails, the terms didn’t form part of a binding agreement between the parties.
The arbitration provision provided that any dispute could be brought in small claims court or by binding arbitration, with a class arbitration waiver and Connecticut choice-of-law terms.
Edward eventually discovered the credit card charges; he’d never made any discounted purchases using the program. He asked for a full refund, but Trilegiant offered to refund only four of six months. Brian also discovered thirty months of charges, and Trilegiant again offered to refund only four months. They sued on behalf of a class for allegedly deceptive “data pass” practices. Among other claims, they alleged the usual California claims. Defendants moved to compel arbitration and the trial court ruled that plaintiffs had never agreed to arbitrate.
The FAA reflects a national policy favoring arbitration as long as the parties have agreed to arbitrate. Whether they have done so is a matter for the court under state contract law. The choice-of-law provision in the agreement (not shown on the enrollment screen) wasn’t determinative unless and until the court found that the parties agreed to it. Fortunately it didn’t matter, since both relevant jursidictions (California and Connecticut) use substantially similar rules for determining whether the parties mutually assented to a contract term.
The key is the parties’ outward manifestations of assent, which can be words or silence, action or inaction, but the party must intend to engage in the conduct and know or have reason to know that the other party may infer assent from the conduct. Trilegiant argued that plaintiffs assented to the arbitration provision by receiving the emailed terms and then not cancelling their memberships during the trial period. Acceptance of a benefit may constitute assent, but only where the offeree decides to take the benefit with actual or constructive knowledge of the terms.
Where the purported assent is largely passive, contract formation often turns on whether a reasonably prudent offeree would be on notice of the term at issue. Clarity and conspicuousness is important to this question. There was no actual notice, so the court turned to inquiry notice and to whether plaintiffs’ conduct in enrolling in Great Fun, and then not cancelling their memberships before the free trial period expired, constituted an objective manifestation of their assent to the arbitration provision.
Assent is generally impossible without knowledge. Thus, an offer and its terms must generally precede acceptance. But there are some exceptions: “The conventional chronology of contract-making has become unsettled over recent years by courts' increased acceptance of this so-called ‘terms-later’ contracting. Assent by receiving later terms and not cancelling is similar to the theory of shrinkwrap licensing. Courts have found shrinkwrap licenses enforceable if consumers fail to return a product after reading, “or at least having a realistic opportunity to read,” the terms and conditions. Here, failure to cancel the Great Fun membership after receipt of the email arguably took the place of failing to return the product. In the alternative, the contract may have been formed at initial enrollment, then its terms changed by the email, which was accepted by plaintiffs’ failure to cancel. (Some authorities require consideration for contract amendments, but some of those hold that mutual arbitration supplies that consideration.) These approaches differ in the timing of contract formation, but the court didn’t need to distinguish them here because the later-emailed terms were never accepted by the plaintiffs. (The court noted that the initial enrollment page didn’t include a clause incorporating subsequent terms delivered by email, so it didn’t need to decide whether such a clause could bind the offeree to unknown “and effectively unknowable” terms.)
What is notice, in an age of infinite form contracts? One can assent to terms one doesn’t actually read, but the offer must still make clear to a reasonable consumer both that terms are being presented and that they can be adopted through the conduct that the offerer wants to constitute assent. An offeree isn’t bound by inconspicuous contractual provisions of which she’s actually unaware when contained in a document whose contractual nature isn’t obvious. Here, an unsolicited email from an online business didn’t put its recipients on inquiry notice of the terms in that email and those terms’ relationship to a service in which they’d already enrolled, nor of the fact that a failure to act affirmatively by cancelling the membership would constitute assent.
In a footnote, the court also deemed the email unclear, whether deliberately or not. The subject line, “Important information about your membership privileges” didn’t mention the contract or the terms. The body began with a welcome message and other details, including extensive discusison of “your great benefits.” The 13th paragraph began the “Terms & Conditions,” with the arbitration provision following seven paragraphs later. “But even had the email more clearly indicated that it contained an arbitration clause, the fact that it was delivered after enrollment and did not require any affirmative acknowledgment of receipt, undermines Trilegiant's assertion that the plaintiffs received sufficient notice to bind them to the additional terms through their inaction.”
It’s true that in the “modern commercial context, there are reasons to allow parties to contract without consideration of, and the possibility to negotiate, every term.” But duty-to-read cases involving terms delivered after a contract begins don’t “nullify the requirement that a consumer be on notice of the existence of a term before he or she can be legally held to have assented to it.” What’s sufficient inquiry notice depends on various factors including the conspicuousness of the term, the parties’ course of dealing, and industry practices, but the ultimate question is “whether reasonable people in the position of the parties would have known about the terms and the conduct that would be required to assent to them.” In a shrinkwrap case, when a purchaser opens the package and discovers the additional provisions, she will understand that, unless she returns the goods, she’ll be bound. (She will? Love the empiricism here.) So, she can’t begin to use the product until she’s been presented with the terms, whether she reads them or not. Other examples of after-arriving terms are likewise rooted in the reasonable expectations of the parties. In many such cases, the language of the original agreement contemplates modifications. Unilateral modification terms aren’t necessarily effective, but their inclusion at least bolsters the argument that the offeree is on inquiry notice when later terms arrive, “particularly where the modification (or amendment) is itself submitted in such a manner that a reasonable offeree would be likely to see it,” such as when it arrives with a bill. (The court noted, however, that legislation in some states was needed to make this work for credit card terms, and then rather cryptically suggested that even without specific legislation such a practice “may” support a conclusion that a reasonable person would be on actual notice.)
Trilegiant argued that plaintiffs’ receipt of the email was enough to establish that they were on inquiry notice. “But that someone has received an email does not without more establish that he or she should know that the terms disclosed in the email relate to a service in which he or she had previously enrolled and that a failure affirmatively to opt out of the service amounts to assent to those terms.” Register v. Verio, by contrast, involved a prior/ongoing relationship between the parties. “Nor would a reasonable person likely understand in some other way that disputes arising between him or her and Trilegiant were to be resolved by an alternative dispute resolution procedure.”
Also, email was different from shrinkwrap because “the recipient of the terms in this case would not have been confronted with the existence of additional terms before being able to benefit from Great Fun.” Other cases approving amendments involved their presentation “during the course of maintaining and using the service to which the terms apply.” Here, though, the arbitration provision “was both temporally and spatially decoupled from the plaintiffs' enrollment in and use of Great Fun.” The critical connection between the terms and the goods/services was missing.
A reasonable person “may understand that terms physically attached to a product may effect a change in the legal relationship between him or her and the offeror when the product is used.” (So, I guess servitudes on chattels are okay now?) “But a reasonable person would not be expected to connect an email that the recipient may not actually see until long after enrolling in a service (if ever) with the contractual relationship he or she may have with the service provider, especially where the enrollment required as little effort as it did for the plaintiffs here.” On these facts, the email wouldn’t have raised a red flag vivid enough to make a reasonable person anticipate a legally significant alteration in the contract.
“To be sure, the ‘duty to read’ rule combined with the ‘standardized form’ contract makes it unlikely in many contexts that a consumer will actually read such a agreement beyond a quick scan, if that.” The offeror doesn’t really expect customers to read, much less understand, the standard terms. But formalities must be observed! At the very least, consumers should be confronted with the terms they won’t read or understand “at a place and time that the consumer will associate with the initial purchase or enrollment, or the use of, the goods or services from which the recipient benefits,” since at least the consumer will be vaguely aware that there are terms “that may one day affect him or her.” Here, however, “Trilegiant effectively obscured the details of the terms and conditions and the passive manner in which they could be accepted.” The solicitation and enrollment pages, combined with the credit card passthrough, “made joining Great Fun fast and simple and made it appear—falsely—that being a member imposed virtually no burdens on the consumer besides payment.”
The court of appeals pointed out that shrinkwrap-approval cases often refer to the efficiency benefits of such contracting. “Here, however, there is no policy rationale supporting Trilegiant's approach inasmuch as there are a plethora of other ways—such as requiring express acknowledgment of receipt of the terms—through which Trilegiant could have met the minimum requirements of notice.” No case cited had found a contract formed/modified with a sequence and terms like this, and this court wasn’t going to be the first.
Lack of notice led to the next issue. “A requirement that the plaintiffs expressly manifest assent to the arbitration provision together with such assent would likely have overcome the email's defects in providing notice.” But the passive conduct of failing to cancel wasn’t enough, even if passive conduct may in other situations be enough to constitute assent. “In order to constitute acceptance, the failure to act affirmatively must carry a significance that reasonable people in the parties' positions would understand to be assent. A party cannot require an evidentiary trial before a trier of fact simply by asserting that the other party assented through a failure to respond to proffered contractual terms.” There has to be evidence that the offeree knew or should have known of the terms, and understood that accepting the benefit would be construed by the offeror as agreement. Here, there was no inquiry notice, and auto-debited payments were too passive for any reasonable finder of fact to find a manifestation of an understanding of the existence of the arbitration terms and an intent to be bound in exchange for continued benefits from Great Fun.
What about the hyperlink on the enrollment screen to the terms and conditions? Maybe it could have created a substantial question as to whether arbitration was part of the parties’ contract, but Trilegiant forfeited the argument by not raising it in the district court. In a footnote, the court of appeals expressed a bit of skepticism, noting that previously it had concluded that a browsewrap provision only readable after scrolling down multiple screens wasn’t enforceable because it didn’t provide inquiry or constructive notice.
But this wasn’t a browsewrap or clickwrap case. There was some indication next to the “click to subscribe” button that there were additional terms. But by contrast to a typical clickwrap, the button didn’t explicitly refer to the terms by asking the user to assent to them. Instead, it only suggested that the user would get benefits by clicking “yes.” “[I]t seems likely that the district court not only did not mention the hyperlink, but pointed out the peculiarity of the fact that the enrollment screen did not seem to indicate to the user that he or she would be bound by additional terms, precisely because the issue was not raised.”
Query: how much does hostility to the credit card passthrough practice—so abusive that Congress was moved to try to stop it—drive the court’s conclusion? The plaintiffs, and others who signed up for these “benefits,” often didn’t understand that they were signing up for a service at all, much less that the service would have arbitration provisions. But, as the court acknowledges, in a world of unknown contract terms, what’s special about this one? Is it possible for legitimate businesses to make arbitration so standard that a consumer has to expect that she’s signing up for one?