Fundamentally unfair: an empirical analysis of social media arbitration clauses.

Author:Koenig, Thomas H.
Position:Continuation of I. Empirical Study of Arbitration Clauses Employed By Social Media F. Do Arbitral Clauses Satisfy Consumer Due Process Principles? 2. Social Media TOU Are Not Reciprocal b. Remediless Social Media Wrongs through Conclusion, with footnotes, p. 375-411
  1. AAA Principle 2: Access to Information (122)

    1. Social Media's Affirmative Duty to Inform

      The AAA requires providers of goods and services such as social networks to "undertake reasonable measures to provide Consumers with full and accurate information regarding Consumer ADR Programs." (123) To comply with the AAA's Consumer Due Process Protocol, a SNS employing consumer arbitration must give the consumer "(1) clear and adequate notice regarding the ADR provisions, including a statement indicating whether participation in the ADR Program is mandatory or optional, and (2) reasonable means by which Consumers may obtain additional information regarding the ADR Program." (124)

      JAMS's parallel due process principle states that "[t]he consumer must be given notice of the arbitration clause. Its existence, terms, conditions and implications must be clear." (125) For 50 percent of the AAA clauses, a user would have to read at least 4,967 words just to reach the first word of the arbitration clause. Only one out of the forty social media providers specifying the AAA gave users an opportunity to opt out of mandatory arbitration. Even in this sole exception, the opt-out right is illusory because Instagram gives users only thirty days from the date the TOU was originally posted to opt out, which means that the window to exercise this option has long since passed. (126) For the nineteen JAMS social media arbitral clauses, no SNS offered an opportunity to opt out of arbitration. In contrast, a quarter of clauses in checking account and credit card contracts allowed consumers to opt out of arbitration. (127)

    2. AAA and JAMS Arbitral Clauses Fail to Inform Users

      SNS TOU are standard form contracts marketed to consumers with identical terms for all users and no likelihood of individual negotiation. (128) Standard-form licenses are broadly enforceable so long as the license satisfies three conditions: (1) the customer has an opportunity to review the terms of the license, (2) the user manifests assent after having an opportunity to review the terms, and (3) the actions are "attributable in law" to the customer. (129) More than three out of four social media providers (76%, N=45), however, predicate contract formation on browsewrap that purports to bind the user without any affirmative act. Forty out of fifty-nine TOU with AAA or JAMS clauses assert aggressive rolling contract provisions that give the SNS the right to change the terms after contract formation at will. Houzz, for example, states that it has the right to change its service agreement and the TOU at will:

      We reserve the right at any time, with or without cause, to:

      * change the terms and conditions of this Agreement;

      * change the Website, including eliminating or discontinuing any Information or Services or other feature of the Website; or

      * deny or terminate your use of and/or access to the Website.

      Any changes we make will be effective immediately upon our making such changes available on the Website or otherwise providing notice thereof. You agree that your continued use of the Website after such changes constitutes your acceptance of such changes. You hereby acknowledge that you have carefully read all of the terms and conditions of our Privacy Policy (which can be accessed at privacypolicy) and agree to all such terms and conditions. Be sure to return to this page periodically to ensure familiarity with the most current version of this Agreement. (130) Seventeen of the nineteen SNS choosing JAMS as the arbitral provider impose rolling contract terms that can be changed at will by the provider but not the consumer (89.5%). Sixty-six percent of the social networks choosing the AAA as the provider structure their TOU as a rolling contract (N=26). (131) Users must check periodically for current terms because their continued use constitutes acceptance of any new terms, even if the user is unaware of any changes. (132)

      Social media providers made little attempt to draw users' attention to the presence of arbitration clauses, burying them deep within the TOU or service agreement. Courts have rejected agreements where a licensor makes little effort to draw attention to pro-licensor terms. In Bragg v. Linden Research, Inc. (133) the developer of "Second Life," an online virtual world developed by the defendant, confiscated the plaintiffs virtual property and removed him from the website. (134) The plaintiff filed suit, and Linden Research responded by filing a motion to compel arbitration based on their arbitration agreement. The court found the arbitration agreement to be procedurally and substantively unconscionable because it was obscurely situated deep in take-it-or-leave-it TOU.

      The Bragg court found lack of mutuality and that Second Life failed to give Bragg sufficient information on the costs and rules of arbitration, as required by the rules of the International Chamber of Commerce. (135) The court reasoned that Second Life could have easily explained the arbitration procedure in either its TOU or through a hyperlink to another page. In our sample, only a handful of the providers explained the arbitration procedure or provided a hyperlink where they could learn about the arbitration rules.

      The mandatory predispute binding arbitration clauses in SNS' TOU typically were written in impenetrable legalese and placed where they were almost certain to go unnoticed. The first word in the arbitration clause of the fifty-nine AAA and JAMS clauses was buried 5,662 words deep in the TOU (median 5,360 words). Because of the obscure location of the arbitration clause, social media users are likely to be unaware that they have waived their right to go to court.

    3. Difficult-to-Read Arbitration Terms

      The duty to read is a long-standing principle in Anglo-American contract law, but there is no concomitant duty of providers to make terms readable. (136) Critics argue that the opportunity to read maxim is a legal fiction because consumers almost never review the boilerplate (137) and that courts and commentators need to abandon the "opportunity to read" requirement as signifying meaningful assent. (138) To assess the readability of the fifty-nine arbitration clauses selecting the AAA or JAMS, we tested each TOU text using the Flesch Reading Ease Formula, (139) which is a standard measure for determining comprehensibility. The readability tests did not rate a single social media TOU to be "easy" or "fairly easy" to comprehend. As Figure Two (below) indicates, the overwhelming conclusion was that TOU were difficult or fairly difficult to comprehend. Terms of use written at a standard level were by far the smallest category. The largest category of TOU was drafted to be fairly difficult to read, while difficult TOU were the second largest category.

      A score of sixty is considered the standard readability score under the Flesch Reading Ease Formula. (140) Scores higher than sixty are more readable than the standard. The mean readability of the social media TOU specifying the AAA or JAMS as providers is fifty-one (median=53), which is nine points more challenging to understand than what is considered a standard score. Under the Flesch-Kincaid Reading Ease Test, text scored as fifty-one is classified as "fairly difficult" to comprehend. Similarly, the Flesch-Kincaid grade level for the AAA and JAMS TOU was grade 11.3 (median=11), as Figure Three (below) reveals. When TOU are very hard to understand, the dominant party can impose one-sided terms that harm the unsophisticated party. (141) Because consumer arbitration clauses impose many restrictions on consumers' legal rights, they should, at a minimum, be understandable by the average user.

      The rights-foreclosure clauses in the social media TOU were written at a much higher-grade level than the TOU as a whole. The average readability of the fifty-nine AAA or JAMS arbitral clauses was at the grade fifteen level, the reading ability of a junior in college, which precludes the possibility that the typical user will understand them. (142) The average American comprehends at between an eighth and ninth grade reading level according to the largest existing study, (143) and there is no evidence that the reading levels have improved since this two-decades-old research. (144) The net effect of incomprehensible provisions, coupled with one-sided terms, is to produce an imbalanced agreement that lacks due process fundamental fairness.

      Twitter's TOU, for example, are drafted at a reading level between grade eighteen and nineteen. (145) Drafted for a person with almost twenty years of education, Twitter's dense warranty disclaimer provision is indecipherable to anyone without advanced training. (146) The use of unnecessarily opaque terms like "herein" and the failure of the SNS to explain what rights are erased seem calculated to obscure the implications of the terms of the "agreement" rather than to educate the consumer.

      Figure Four (above) reveals that the SNS arbitration clauses were generally more complex than the TOU as a whole, being rated as "difficult" or "very confusing" and requiring an average educational level of fourteen years of schooling to understand. The Flesch Reading Ease Readability score was 34.5, almost twenty points more difficult than the TOU as a whole and twenty-five points below the standard score of sixty. Only one of the fifty-nine clauses was "fairly easy" to read, with a Flesch Reading Ease score between seventy and seventy-nine.

      Five social media arbitration clauses were "fairly difficult." Fifty-three out of fifty-nine consumer arbitration clauses were either "difficult" (N=30) or "very confusing" (N=23). The mean Flesch Reading Ease score for the arbitration clauses was thirty-four, which is classified as "difficult" to read (median=34). The aggregate readability score for the arbitration clause was grade fourteen as compared to grade eleven for the TOU. Our findings are consistent...

To continue reading