For decades, the Supreme Court has expanded the Federal Arbitration Act (FAA), and companies have inserted arbitration clauses in hundreds of millions of consumer and employment contracts. But in an article forthcoming in the University of Pennsylvania Law Review, I explore a different way in which arbitration’s shadow is growing.

Traditionally, drafters only attempted to mandate arbitration of disputes that were connected to the contract that includes the arbitration provision (the “container contract”). Thus, even the broadest arbitration clause merely applied to “any controversy or claim . . . arising out of or relating to this agreement.” To enforce this limit, courts often refused to compel arbitration of claims that either (1) arose from shocking and unforeseeable wrongdoing, or (2) were brought by or against non-signatories, or (3) stemmed from events that occurred after the container contract had expired.

Now, however, businesses have started to experiment with what I call “infinite” arbitration clauses. These provisions exhibit at least one of the following characteristics. First, they insist that they govern all claims between the parties—even those that do not stem from the container contract. For instance, Sprint’s arbitration clause applies to “ANY (we really mean ANY) disagreements about our relationship.” Accordingly, at least on paper, infinite clauses cover behavior that has nothing to do with the original transaction, such as sexual harassment after the sale of household goods or “a punch in the nose during a dispute over medical billing.”

Second, infinite arbitration clauses extend beyond the original contractual partners. Consider AT&T Mobility’s Customer Agreement, which announces that it binds and benefits a rainbow of non-signatories, including the parties’ “subsidiaries, affiliates, agents, employees, predecessors in interest, successors, and assigns.” To put the reach of this language in perspective, AT&T Mobility is a subsidiary of AT&T Corporation, which is the parent of ninety-eight separate firms. Therefore, if a plaintiff sues a member of the AT&T Corporation family—say, DIRECTV, HBO, or Yellowpages.com—the defendant will argue that the plaintiff had agreed to arbitrate against them simply by signing up for cell phone service with Mobility.

Third, infinite provisions have no sunset date. Although the common law condemns perpetual contracts, infinite clauses—such as the one in Wells Fargo’s checking account agreement—purport to “survive[] the closing of [an] account or termination of any service.”

In sum, infinite clauses stretch to the horizon and last forever. They are less a contractual provision and more a kind of arbitration servitude.

Courts are divided about whether to take infinite provisions literally. Several have refused to allow companies to compel arbitration in such broad strokes. For example, as one judge explained, enforcing AT&T Mobility’s arbitration clause could spawn absurd results:

If [the plaintiff] were hit by Mobility’s delivery van or tripped over a dangerous condition in Mobility’s store, her tort claim would have to go to arbitration. If she bought shares of stock in Mobility and later claimed a decrease in share price was the result of corporate malfeasance, her securities-fraud claim would have to go to arbitration. And since the arbitration clause purports to survive termination of the underlying service agreement, this obligation to arbitrate any claim whatsoever against Mobility would last forever.

Similarly, other judges have applied state law to hold that boundless arbitration clauses are unconscionable or must be construed in accordance with a consumer or employee’s reasonable expectations.

But this jaundiced view of infinite clauses is hard to square with the Supreme Court’s recent FAA decisions. These opinions admonish lower courts to “enforce arbitration agreements according to their terms” and to ignore any state rule that discriminates against arbitration. Taking these commands to heart, other courts have opined that there is nothing troubling about a “clause that requires arbitration of all disputes between the parties.” For example, a federal district court in Florida recently ordered arbitration of a lawsuit brought by an employee against her employer for allegedly failing to prevent a co-worker from raping her. The plaintiff asserted that her complaint was unrelated to her employment contract because the attack occurred while she was off the clock and far from where she performed her job. Yet because the arbitration provision expressly extended beyond the employment relationship, the court held that this argument was a non-sequitur.

Against this backdrop, I argue that judges should recognize two checks on infinite arbitration clauses. First, I contend that some of these provisions fail because they attempt to impose arbitration on parties who never actually agreed to the process.

The clearest example of this phenomenon involves non-signatory plaintiffs. In general, only the parties to a contract are bound by it. However, the FAA also absorbs third-party beneficiary principles, which require non-signatories to arbitrate if an agreement unambiguously showcases the signatories’ intent to benefit the non-signatory.

Infinite provisions try to satisfy this test simply by declaring the drafter’s wish to bind vast, open-ended groups of individuals. For example, wireless, cable, and internet service providers mandate arbitration not just for the account holder, but for “all authorized and unauthorized users.”

A similar phenomenon occurs in nursing-home admission contracts. When a nursing home causes the death of a resident, an arbitration clause in the contract between the institution and the resident only goes so far. To be sure, because the decedent has signed the contract, the arbitration agreement applies to a survival action brought by the decedent’s personal representative on behalf of the estate. But several jurisdictions also allow a decedent’s relatives to seek redress for their own emotional and pecuniary injuries through a wrongful death lawsuit. Because these plaintiffs have never signed the container contract, nursing homes have expanded the ambit of their arbitration provisions to include “all persons whose claim is derived through or on behalf of the [r]esident, including any parent, spouse, sibling, child, guardian, executor, legal representative, administrator, or heir of the [r]esident.”

The flaw here is simple: No one can create contractual consent out of whole cloth. A random person who “uses” a cell phone or the spouse of an elderly negligence victim has never assented to the container contract. Indeed, it was someone else who signed on the dotted line or clicked “I agree.” Moreover, the oblique references to binding “users” or other non-signatories are insufficient to transform them into third party beneficiaries. Thus, the infinite portions of these arbitration clauses are just words on a page. No matter what a contract says, “federal law does not force arbitration upon a party that never agreed to arbitrate in the first place.”

Second, my article argues that another check on infinite clauses is hiding in plain sight. Section 2, the FAA’s centerpiece, makes arbitration provisions specifically enforceable under federal law. It is the source of the vigorous federal policy in favor of arbitration, including the statute’s merciless preemption of state law.

However, section 2 instructs courts to enforce a term “in any . . . contract . . . to settle by arbitration a controversy thereafter arising out of such contract.” As the italicized words reveal, the statute only governs agreements to arbitrate “controvers[ies] . . . arising out of [the container] contract.” In turn, “to arise” has long been understood as “[t]o originate; to stem (from) . . . [or] [t]o result (from).”

However, infinite arbitration clauses purport to cover allegations without regard to whether they flow from the container contract. Indeed, they attempt to suck into their maw all causes of action—no matter their source. Because section 2 requires that a claim “arise[s] out of” the container contract, it does not apply to the infinite portions of these clauses.

This boundary was no accident. Congress modeled the FAA on New York’s pioneering 1920 arbitration statute.  Yet New York’s version of section 2 applied across the board to any lawsuit “arising between the parties to the contract.”  In contrast, federal lawmakers chose to narrow the FAA to cases that were tethered to the container contract.

Likewise, one draft of the FAA applied to complaints that were merely grounded in the parties’ general “transaction[s].” The word “transaction” is broader than “contact.” Thus, if Congress had passed this version of the FAA, section 2 would have covered agreements to arbitrate claims that emerged from the sprawling universe of the parties’ relationship, rather than the narrower subject matter of the container contract. Ultimately, though, Congress deleted this phrase, reinforcing the necessity of a nexus between the plaintiff’s claim and the container contract.

For these reasons, section 2 often does not apply to ultra-broad arbitration clauses. And as section 2 falls away, state courts and lawmakers can freely regulate these provisions by finding them to be unconscionable or outside the scope of an individual’s reasonable expectations.

This post comes to us from Professor David Horton at the University of California, Davis – School of Law. It is based on his recent article, “Infinite Arbitration Clauses,” available here.