Analyzing unconscionability in arbitration agreements

Does that arbitration clause read like a deal with the devil? A lighthearted and meaningful presentation of unconscionability

2022 September

Surprisingly, one can have fun analyzing unconscionability in arbitration agreements. Unconscionability jurisprudence evokes Faustian bargains, magnifying glasses, and shocks to the conscience. In the first part of this article, we’ll review the fundamentals of unconscionable arbitration contracts. In the second, we’ll explore the magic numbers of procedural unconscionability. In the last section, we’ll look at the interplay of law and facts in substantive unconscionability.

Unconscionability fundamentals

Let’s start with the basics. Under federal and California law, arbitration contracts are valid unless they are unenforceable or unconscionable. (Armendariz v. Foundation Health Psychcare Services, Inc. (2000) 24 Cal.4th 83, 98 (Armendariz).) The unconscionability standard is the same in federal and state courts. (Poublon v. CH Robinson Company (2017) 846 F.3d 1251, 1260.)

Unconscionability has two components: procedural and substantive. (Armendariz, supra, 24 Cal.App.4th at p. 114.) An arbitration contract must be both procedurally and substantively unconscionable. (Ibid.)

To evaluate procedural unconscionability, the court will ask three questions:

Was there adhesion (unequal bargaining position and a form contract)?

Was there surprise (undisclosed terms)?

Was there oppression (pressure)?

Each “yes” answer above increases the level of procedural unconscionability. A single yes usually means a low level. Two yeses are a moderate level. All three constitute a high degree of procedural unconscionability.

Establishing substantive unconscionability is more complex. Not all one-sided contract provisions are unconscionable. (Sanchez v. Valencia Holding (2015) 61 Cal.4th 899, 911.) The terms must be overly harsh, unduly oppressive, or unreasonably favorable. (Id. at p. 911.) Those descriptions “all mean the same thing.” (Ibid.) The terms must be so one-sided that they shock the conscience. (Id. at pp. 910-911.)

Procedural and substantive unconscionability complement each other. Cases reference a sliding scale: If you have a lot of one, you need less of the other. (Armendariz, supra, 24 Cal.4th 114.) However, “minimal” procedural unconscionability requires a “high” degree of substantive unconscionability. (Davis v. Kozak (2020) 53 Cal.App.5th 897, 917 (Davis), citing Peng v. First Republic Bank (2013) 219 Cal.App.4th 1462, 1470.)

If a contract is procedurally and substantively unconscionable, the court next looks to severance. The court must sever the unconscionable parts of the contract and enforce the rest unless unconscionability permeates. (Armendariz, supra, 24 Cal.4th at p. 122.)

Permeation depends on the number and purpose of the substantively unconscionable provisions. Where there is one unconscionable provision, that provision is usually severable. (See Cisneros Alvarez v. Altamed Health Services (2021) 60 Cal.App.5th 572, 596 (Cisneros); Nguyen v. Applied Medical Resources Corp. (2016) 4 Cal.App.5th 232, 255-256; Poublon, supra, 846 F.3d at pp. 1273-1274.) It may be an abuse of discretion not to sever. (Farrar v. Direct Commerce, Inc. (2017) 9 Cal.App.5th 1257, 1275.)

Where there are two unconscionable terms, cases have held that those permeate the contract with unconscionability. (Armendariz, supra, 24 Cal.4th at p. 124, Penilla v. Westmont Corp. (2016) 3 Cal.App.5th 205, 223; De Leon v. Pinnacle Property Management Services, LLC (2021) 72 Cal.App.5th 476, 492-493.) Three or more illegal provisions almost guarantee permeation. (See Ali v. Daylight Transport, LLC (2020) 59 Cal.App.5th 462, 481-482 (Ali); Ramirez v. Charter Communications, Inc. (2022) 75 Cal.App.5th 365, 386–387 (Ramirez); Magno v. The College Network, Inc. (2016) 1 Cal.App.5th 277, 292 (Magno); Baxter v. Genworth North America Corp. (2017) 16 Cal.App.5th 713, 737-738 (Baxter).)

In sum, an unconscionable contract is too hard for PaSUP: Procedural and Substantive Unconscionability Permeates. Enough background. Let’s go down to the crossroads.

A deal with the devil

You represent blues great Robert Johnson. At the crossroads, legend has it he traded his soul with the devil for blues guitar mastery. Mr. Johnson seeks to invalidate his contract because it is unconscionable.

The devil claims the contract is not adhesive, so it is conscionable. (I have found no case using “conscionable,” but it means what you think it means. (Black’s Law Dict. (11th ed. 2019) p. 379 col. 1).) The devil proves that he drafted a bespoke contract for Mr. Johnson, signed in blood. Believe it or not, there is a case on point.

Faust, depending on which version of the story you read, might or might not end up going to hell because he freely – and without duress – sold his soul to the devil; he was pretty well off by the standards of his time. But someone who makes a Faustian pact to become someone else’s slave in our system is going to find that, among other things, the deal is unconscionable, period, end of discussion.(Harper v. Ultimo (2003) 113 Cal.App.4th 1402, 1409 (Harper).)

In real life, the Harpers didn’t sell their souls; they contracted for some work in their backyard. The contractor, Ultimo (sounds villainous enough), allegedly damaged the yard, the sewer system, the soil, and ultimately (pun intended) the Harper’s plumbing. (Harper, at p. 1405.) The Harpers sued for negligence, fraud, and breach of contract, seeking punitive damages. (Ibid.)

The Court of Appeal held, “both procedural and substantive unconscionability are so present that it is almost impossible to keep from tripping over them.” (Id. at p. 1406.) The Harper court found both surprise and oppression:

Here is the surprise: The customer must inevitably receive a nasty shock when he or she discovers that no relief is available, even if out and out fraud has been perpetrated….Here is the oppression: The inability to receive full relief is artfully hidden by merely referencing the Better Business Bureau arbitration rules, and not attaching those rules to the contract for the customer to review.(Harper at p. 1406, emphasis added.)

Keep in mind that the court issued the Harper opinion in December 2003. Google was a nascent company, Gmail wouldn’t appear until the next year, and the iPhone and Android phone were four and five years out, respectively. Although the Internet Archive shows that the Better Business Bureau had its rules online then, the U.S. Census Bureau reported that only 54.7% of houses had internet access that year. [https://web.archive.org/web/20031204112805/http://www.dr.bbb.org/programs/ids.asp] [https://www.census.gov/content/dam/Census/library/publications/2005/demo/p23-208.pdf]

The Harpers alleged that the Better Business Bureau’s arbitration rules were substantively unconscionable. That allegation was vital. The California Supreme Court has explicitly held that failure to attach arbitration rules is only significant when a party alleges those unattached rules are substantively unconscionable. (Baltazar v. Forever 21, Inc. (2016) 62 Cal.4th 1237, 1246.) Mere failure to attach the applicable rules does not establish procedural unconscionability. (Ibid.) But I digress, back to adhesion.

Very clever readers will have noticed in our discussion of procedural unconscionability above that we did not make any reference to the traditional criteria of adhesion. There was no need…. [T]his case certainly involves procedural unconscionability regardless of whether the contract is adhesive.(Harper, supra, 113 Cal.App.4th 1410, emphasis in original.)

Harper’s takeaway is that adhesion, surprise, and oppression all contribute to procedural unconscionability, but none is a prerequisite. A greater number shows a greater degree of procedural unconscionability.

Adhesion is about relationships, not labels

While Harper holds that adhesion isn’t necessary to establish procedural unconscionability, adhesion is still relevant. When evaluating adhesion, the critical inquiry is whether the party with superior negotiating position presented a take-it-or-leave-it contract. Classic examples are employment (Farrar, supra, 9 Cal.App.5th at p. 1266), consumer (Fisher, supra, 66 Cal.App.5th at p. 1095), and housing rental contracts (Penilla, supra, 3 Cal.App.5th at p. 215.)

However, courts have held that some contracts are adhesive, even when the parties alleging unconscionability were high-level executives (Stirlen v. Supercuts, Inc. (1997) 51 Cal.App.4th 1519, 1534) and law firm partners (Ramos v. Superior Court (2018) 28 Cal.App.5th 1042, 1058-1059 (Ramos)). It makes no difference if the contracting worker is an employee or an independent contractor. (Subcontracting Concepts (CT), LLC v. De Melo (2019) 34 Cal.App.5th 201, 214-215 (Subcontracting Concepts).) The question is the parties’ relative bargaining strength and ability to negotiate the arbitration terms, not labels like employee and independent contractor. (Ibid. See also Ali, supra, 59 Cal.App.5th at pp. 473-474; Ramos, supra, 28 Cal.App.5th at p. 1058.)

Surprise!

Surprise covers deceptive practices like using incomprehensible language or hiding things in the fine print. For the finest print, let’s squint at Fisher v. MoneyGram (2021) 66 Cal.App.5th 1084 (Fisher).

In Fisher, MoneyGram printed its arbitration clause in six-point type on the back of a transfer form. (Fisher, supra, 66 Cal.App.5th at p. 1092.) According to Wikipedia, six-point type is 2.117 mm high, or about .083 inches. Mr. Fisher was 63 years old with poor eyesight. He needed trifocals and a magnifying glass to read the contract. (Id. at p. 1092.)

In response to the motion to compel arbitration, Mr. Fisher called a typography expert with 20 years of experience. (Fisher, at p. 1097.) The expert’s opinion? “[T]he Terms & Conditions may be the most challenging-to-read document I have ever encountered.” (Ibid.) The trial court and appellate court agreed. This almost impossible-to-read term on the back of a form was a surprise. (Id. at 1103-1104.) But tiny type on the back of a document seems a straightforward call. Let’s look at some more nuanced surprise issues.

Language barriers

According to the Judicial Council, Californians speak over 200 languages and dialects. (https://www.courts.ca.gov/languageaccess.htm) California courts have scrutinized sharp practices affecting people with limited English skills. Decisions finding unconscionable surprise have two common features:

Low English skills, and

Time constraints.

For instance, when a contractor presented a take-it-or-leave-it contract to Portuguese-speaking contract laborers, demanded the laborers sign the contract on the spot, and failed to explain what rules would govern, that was a surprise. (Subcontracting Concepts, supra, 34 Cal.App.5th at p. 211.) When mobile home managers knew their potential residents had limited English skills but presented English arbitration agreements with insufficient time for review, that was also a surprise. (Penilla, supra, 3 Cal.App.5th at pp. 216-217.)

On the other hand, the court found no surprise when an employee had sufficient English skills and time to review the arbitration agreement, even though English was her second language. (Cisneros, supra, 60 Cal.App.5th at p 589.) In another case, the court found no surprise based on language when the employee had earned a college degree in English, attended four years of college in English, and verified on his job application that he understood English. (Nguyen, supra, 4 Cal.App.5th at p. 250.)

If it isn’t deceptive, hidden, or sudden, it isn’t a surprise.

Oppression

Oppression is the third procedural unconscionability factor. It means a lack of negotiation and meaningful choice. (OTO L.L.C. v. Kho (2019) 8 Cal.5th 111, 126 (Kho).) Circumstances to consider when evaluating oppression are:

The time given to consider the contract

The amount and type of pressure to sign

The length and complexity of the contract and the arbitration term

The education and experience of the signer

The signer’s access to an attorney

(Id. at pp. 126-127.)

Some of these considerations overlap with surprise factors. Courts frequently lump oppression and surprise together, leaving which was which as an exercise for the reader.

Courts have recognized that adhesive employment contracts are inherently oppressive. (Kho, supra, 8 Cal.5th at p. 127.) However, a standard take-it-or-leave-it employment contract alone establishes a low level of oppression. (Nguyen, supra, 4 Cal.App.5th at pp. 247-248.)

We learn the difference between the low-level oppression in all employment contracts and the high-level oppression that impresses a court in OTO L.L.C. v. Kho (2019) 8 Cal.5th 111 (Kho). In Kho, an employer presented to an existing employee an agreement with an arbitration clause. This happened during work hours in his workspace while he tried to work. Because the employee was paid by the piece, the time he spent reviewing the agreement diminished his pay. No one explained the contract, and the employee did not receive a copy. (Kho, at pp. 127-128.) The employee did not have access to an attorney, and the low-level “porter” who presented the contract did not appear to have the authority to negotiate. (Kho, supra at pp. 127-128.)

Although the contract was one page, the arbitration paragraph was 51 lines long, incorporating more than a dozen statutes and regulations. The Supreme Court described it this way: “A layperson trying to navigate this block text, printed in tiny font, would not have an easy journey.” (Kho, at p. 128.)

The court observed that Mr. Kho had little time to review the contract, and what time he used directly impacted his income. The court noted that keeping one’s job puts additional economic pressure on an employee. (Id. at p. 127.) There was “significant” oppression. (Id. at p. 127.)

Courts have found oppression in other transactions. There was oppression when an unsophisticated buyer of solar panels had little time to review an “arcane” arbitration clause. (Aanderud v. Superior Court (2017) 13 Cal.App.5th 880, 895.) There was oppression when consumers had limited time to review documents for admission to elder care or mental health facilities. (Dougherty v. Roseville Heritage Partners (2020) 47 Cal.App.5th 93, 103-104, [elder], Nelson v. Dual Diagnostics Treatment Center, Inc. (2022) 77 Cal.App.5th 643, 661-662 [mental health].) There was a high degree of procedural unconscionability (and implicitly oppression) when an online nursing school rushed prospective students through the signing process. (Magno, supra, 1 Cal.App.5th at pp. 285-286.)

The thread is clear: The opportunity to bargain depends on the negotiating power, time, and comprehension – oppression results when any of those are diminished.

Wrapping up procedural unconscionability

Adhesion, surprise, and oppression can each establish some procedural unconscionability. (Fisher, supra, 66 Cal.App.5th at p. 1097.) As these factors accumulate, the level of unconscionability grows. (Id. at p. 1107.)

Substantive unconscionability

While procedural unconscionability focuses on the circumstances surrounding the contract’s execution, substantive unconscionability concentrates on the contract’s terms: Are they so one-sided that they shock the conscience? (Sanchez v. Valencia Holding, supra, 61 Cal.4th at pp. 910-911.) A bad bargain alone is not substantively unconscionable. (Id. at p. 911.) Not all one-sided contracts are unconscionable. (Id. at p. 911, citing Pinnacle Museum Tower Association v. Pinnacle Market Development (US), LLC (2012) 55 Cal.4th 223, 246.) A one-sided contract term can be permissible if a business can show a legitimate commercial need. (Id. at p. 912.)

In this section, we’ll look at some frequently challenged arbitration terms and see what distinguishes one-sided bad bargains from substantively unconscionable agreements. It’s a representative, but not an exhaustive list.

Carve-outs and mutuality

When an arbitration clause applies to all disputes between the parties, courts are unlikely to find substantive unconscionability. (See, e.g., Torrecillas v. Fitness International, LLC (2020) 52 Cal.App.5th 485, 491.) A clause covering “any claim, dispute, or controversy” is mutual, not one-sided. (Sanchez v. CarMax (2014) 224 Cal.App.4th 398, 403-404 (CarMax).) Simply listing possible claims subject to arbitration is not an impermissible carve-out if the list does not limit the claims one can bring. (Baltazar, supra, 62 Cal.4th at p 149.) And merely using first-person pronouns, e.g., “I agree,” does not make an otherwise mutual agreement one-sided. (Nguyen, supra, 4 Cal.App.5th at 252; see also Davis, supra, 53 Cal.App.5th 897, 915.)

But carving out specific claims from arbitration is substantively unconscionable if the carve-outs unfairly favor the party with superior bargaining power. (Ramirez, supra, 75 Cal.App.5th at pp. 383-384.) For instance, mandating arbitration for workplace and wage-related claims like wrongful termination, discrimination, harassment, retaliation, medical leave, disability discrimination, and workplace safety while exempting claims like unfair competition, trade secrets, non-compete agreements, and intellectual property rights unfairly favors the employer. (Ibid. See also Davis, supra, 53 Cal.App.5th at 917.) Similarly, when a confidentiality provision protects employer secrets but not employee secrets, it is unconscionable. (Davis at p. 917.)

Carve-outs for injunctive relief are more complex. Bilateral injunctive carve-outs are permissible, even if an employer is more likely to use them. (Baltazar, supra, 62 Cal.4th at pp. 1247-1248.) Such carve-outs restate existing law. (Ibid.)

In contrast, unilateral injunctive carve-outs are unfairly one-sided. (Ramirez, supra, 75 Cal.App.5th at p. 383; Ali, supra, 59 Cal.App.5th at p. 480.) But one-sided terms for provisional remedies can be conscionable if the business can show a legitimate business need, e.g., allowing repossession of a car after a default. (See, e.g., Sanchez v. Valencia Holding, supra, 61 Cal.4th at p. 940.)

What companies cannot do is preclude relief that a consumer or employee would enjoy under the law. For instance, in California, barring Private Attorney General Act (PAGA) claims or precluding a worker from seeking relief from the Labor Commissioner is unconscionable. (Subcontracting Concepts, supra, 34 Cal.App.5th at pp. 214-215 (Subcontracting Concepts).) Limiting remedies such as equitable relief, punitive damages, and attorney fees is unconscionable. (Ibid.; see also Dougherty, supra, 47 Cal.App.5th at p. 107.)

However, restating the law or explicitly limiting damages to what the law would otherwise allow is conscionable. (See Ramirez, supra, 75 Cal.App.5th at p. 376 [prevailing party attorney fees found conscionable because the remedy was consistent with statute]; see also Poublon, supra, 846 F.3d at pp. 1267-1268 [limiting attorney fees is conscionable if consistent with statutory limits].) The key question is whether the limits take away what the law would otherwise give.

Statute of limitations

Courts have almost uniformly struck down arbitration clauses that limit a claimant’s time to bring a claim. (Magno, supra, 1 Cal.App.5th at pp. 290-291; Baxter, supra, 16 Cal.App.5th at p. 731; Ramirez, supra, 75 Cal.App.5th at p. 374; De Leon, supra, 72 Cal.App.5th at pp. 492-494; Penilla, supra, 3 Cal.App.5th at p. 222 [one year]; Ali, supra, 59 Cal.App.5th at p. 477 [120 days].) Similarly, requiring a party to commence an arbitration within 120 days of an arbitrator’s appointment is moderately unconscionable. (Baxter, at p. 735.)

However, some shortened time limits are permissible. At least one court has held that a 30-day limit to seek arbitrable review of an informal proceeding is not unconscionable on its face. (Epstein v. Vision Service Plan (2020) 56 Cal.App.5th 223, 247-248.) In this situation, the court reasoned, the time limit for arbitral review is more like the 60-day deadline to file a notice of appeal under California Rules of Court, rule 8.104(a). (Ibid.) A party alleging unconscionability would have to show why 30 days would be insufficient. (Ibid.) That party must use the four Fs: facts, facts, facts, and more facts. In the following topics, we’ll see that courts commonly require an evidentiary showing to establish substantive unconscionability.

Discovery

Limited discovery is a hallmark of arbitration. (Ramirez, supra, 75 Cal.App.5th at p. 385.) However, discovery must allow a party to vindicate their statutory rights in consumer and employment arbitrations. (Baxter, supra, 16 Cal.App.5th at p. 727.) If a party cannot do so, the discovery limit is unconscionable. (Ibid.)

Context is important. Where parties allege statutory claims like elder abuse, workplace harassment and retaliation, wage and hour violations, or consumer claims, a court is more likely to find that commercial arbitration discovery limits are insufficient to vindicate those claims. (See, e.g., Dougherty, supra, 47 Cal.App.5th at p. 106.) Proof standards are also important. A claim that requires proof of abuse or neglect by clear and convincing evidence will need more discovery. (Id. at pp. 105-106.)

Many arbitration agreements set explicit limits on depositions and other forms of discovery and set ill-defined guidelines for an arbitrator’s discretion to order discovery beyond the defaults. Where the arbitration clause does not expressly give the arbitrator discretion to order additional discovery, courts are unlikely to read it in. (Ramirez, supra, 75 Cal.App.5th at p. 385-386.) Moreover, courts have struggled to interpret discovery clauses with novel standards.

What’s more stringent, a “showing of need” or “good and sufficient cause?” One court has declared that it’s the latter. (Baxter, 16 Cal.App.5th at p. 727.) Is “sufficient cause” equivalent to “good cause?” The Court of Appeal has its doubts. (Davis, supra, 53 Cal.App.5th at pp. 911-912.) We know that “good and sufficient” cause is less stringent than “compelling need.” (Baxter, at p. 729.) These non-standard phrases have challenged the courts. (Ibid. See also Davis, supra, at p. 912.)

However, language alone cannot render discovery limits substantively unconscionable. Parties seeking to invalidate discovery limits must present evidence that the discovery limits prevent vindicating their rights. (Torrecillas, supra, 52 Cal.App.5th at p. 497).

A factual showing is essential. In one case (CarMax), a discovery limit of three depositions was permissible, but in another (Ramirez), a four-deposition limit was not. The difference? In CarMax, the party made no showing that three was insufficient. (CarMax, supra, 224 Cal.App.4th at pp. 405-406.) In Ramirez, the party proved that she needed seven depositions. (Ramirez, supra, 75 Cal.App.5th at pp. 385-386.)

Remember your four Fs: facts, facts, facts, and more facts.

Conclusion

Let’s recap. First, an unconscionable arbitration contract is a deal that’s too hard to PaSUP. Second, procedural unconscionability grows with the accumulation of adhesion, surprise, and oppression. Third, low procedural unconscionability requires high substantive unconscionability. Fourth, establishing substantive unconscionability usually requires the four Fs. See, that wasn’t so bad.