A California Federal Court Allows a Zero-Fee Antitrust Damages Model in Affinity Credit Union v. Apple Inc.

The article examines class certification and the proposed zero-fee benchmark in the Apple Pay issuer litigation. It considers the economic evidence needed to distinguish an observed pricing difference from harm caused by unlawful conduct.

A California Federal Court Allows a Zero-Fee Antitrust Damages Model in Affinity Credit Union v. Apple Inc.

The American Counsel, September 27, 2026

On September 23, 2026, Judge Jeffrey S. White of the Northern District of California certified a class of payment-card issuers challenging Apple's fees for Apple Pay. The court also denied Apple's motion to exclude an economist whose damages model uses a competitive issuer fee of zero. If the issuers establish liability and that benchmark, the model would count the fees Apple charged class members during the class period as damages. The order permits that theory to proceed without deciding that Apple violated the antitrust laws or that the competitive fee actually would have been zero.[1]

The case raises a consequential question for antitrust jurisprudence. A platform can provide a service to one group without charging that group directly while obtaining value elsewhere in its business. When a rival charges for a comparable service, the difference can provide evidence of an overcharge. Establishing the amount attributable to unlawful conduct requires an explanation of why the prices differ. The zero-fee model in Affinity Credit Union v. Apple Inc. brings that causal question into focus.

The plaintiffs are financial institutions that contracted with Apple to enable their cards for use in Apple Pay. They allege that Apple suppressed competition in tap-and-pay mobile wallets on its devices and used the resulting power to charge issuers supracompetitive transaction fees. The certified class comprises U.S. entities that issued an Apple Pay-enabled payment card and paid Apple a fee on transactions using that card. The order leaves market definition and monopoly power to be established through common proof.[2]

Apple's standardized pricing helps explain why the court permitted class treatment. The order describes one fee for credit transactions and another for debit transactions. If the issuers prove that those fees applied uniformly without exception and were supracompetitive, the resulting impact could be established across the class. The economist Christopher Vellturo supplies the proposed measure. His yardstick analysis compares Apple's fees with those of other mobile wallets that charge issuers zero. Under that model, damages equal the sum of the challenged fees.[3]

The arithmetic is simple once zero is adopted as the competitive benchmark. The justification for that price supplies the model's essential causal claim. Uniform charges can make an injury susceptible to common proof. Uniformity alone cannot establish that the charges exceed a competitive level.

There is public support for investigating the zero-fee comparison. In 2015 Visa announced a commercial framework for mobile payments that included Android Pay and later Samsung Pay. The announced framework provided for no pass-through fees between technology partners and participating financial institutions. In September 2023 the Consumer Financial Protection Bureau reported that card issuers paid zero transaction fees to Google Pay and Samsung Pay. These historical sources support the existence of the pricing difference. They do not establish the competitive price Apple would have charged throughout the damages period.[4]

The distinction between the particular fee and the economics of the service matters. A wallet-provider fee of zero says nothing by itself about the costs of processing a payment or the provider's return from the broader business. It also does not establish that network services, merchant acceptance or integration are costless. A sound comparison must identify the same category of charge while accounting for material differences in the services. Visa's announced framework creates a further question about how network arrangements shape the observed price and how those arrangements would operate in a competitive iOS wallet market.

The court's October 2025 notice of questions for hearing identifies the competing explanations. According to the notice, Apple's expert argued that a zero issuer fee would require Apple Pay to obtain revenue outside the proposed antitrust aftermarket. Vellturo responded that Apple receives benefits comparable to those Google and Samsung receive from their wallets, including increased product sales and customer loyalty. The court asked whether the proposed zero rate necessarily implicated other markets that had not been defined.[5]

Those were questions for the hearing and summaries of expert positions. They were not findings that the providers receive equivalent benefits.

The existence of costs cannot alone establish that a particular group would pay a positive price under competition. A service that makes another product more attractive can be commercially useful even when some participants use it without a direct charge. Conversely, the existence of indirect benefits does not establish that those benefits are equivalent across providers or sufficient to support the same price. A persuasive zero-fee benchmark must connect the observed price to the market Apple would have faced without the challenged conduct.

That is the discipline imposed by Comcast Corp. v. Behrend. The Supreme Court reversed class certification where a damages model failed to isolate the harm attributable to the liability theory accepted for class treatment. The Court required a fit between the alleged violation and the loss being measured. Its discussion of benchmark markets also required attention to economic conditions unrelated to the alleged wrongdoing.[6]

Applied here, the relevant inquiry includes whether differences in wallet functions, contractual arrangements or costs explain any part of the difference in issuer fees. The question is whether the comparison reliably isolates the effect of the challenged restraints. The public order does not describe Vellturo's adjustments in enough detail to permit an independent assessment of that work. The absence of those details from the order provides no basis to assume that the expert ignored them.

Comparators also need not be identical. In Alaska Rent-A-Car, Inc. v. Avis Budget Group, Inc., the Ninth Circuit upheld the admission of a damages analysis despite differences between the businesses and geographic markets being compared. The expert had explained the comparisons. The trial court had found sufficient data and a causal connection. Although that decision concerned contract damages, it illustrates why imperfections in an explained benchmark do not automatically make the analysis inadmissible.[7]

The threshold remains meaningful. Federal Rule of Evidence 702 requires the proponent to establish that the testimony satisfies the rule's reliability requirements by a more-likely-than-not standard. Those requirements encompass the factual basis and the application of the method. The 2023 Committee Note rejects treating every objection to those matters as merely a question of weight. The Ninth Circuit reinforced that point in Engilis v. Monsanto Co. while preserving the distinction between assessing reliability and determining whether an expert's conclusion is ultimately correct.[8]

Judge White concluded that Apple's objections to Vellturo concerned conclusions that could be tested through competing evidence. That disposition allows the model to be used. It does not make every zero-fee comparison reliable or settle the economic dispute. Assessing the ruling more critically would require the reports and objections underlying the court's abbreviated discussion.[9]

The broader significance concerns where platform power is exercised and where its financial consequences appear. In its hearing notice, the court expressly questioned how an aftermarket defined through iOS users' lock-in could support a claim by issuers. Consumers' position within the device ecosystem helps define the alleged restriction. Issuers pay the disputed charge. The plaintiffs' theory connects those two positions through the value of access to the same transactions.[10]

This connection extends the concern addressed in this publication's discussion of two-sided markets. A service's price to one participant can conceal the terms imposed on another. Examining the issuer fee makes that distribution visible. The analysis still must establish harm to competition and a causal relationship between the challenged conduct and the claimed overcharge.[11]

Timing matters as well. Apple now offers eligible developers in the United States a route to NFC and Secure Element access under commercial agreements and technical conditions. Those arrangements require separate examination of their practical competitive effects. Their existence also means that a description of historical exclusion should not become an unqualified assertion about current access.[12]

The zero-fee theory deserves consideration because a competitive service can generate value without charging every participant. Its force depends on proof that competition would have constrained Apple's issuer fees in the way the model predicts. The corresponding discipline for courts is to examine the economic basis for the allocation of charges. A company's chosen source of revenue cannot supply its own justification for the price that competition would permit.


[1]   Affinity Credit Union v. Apple Inc., No. 22-cv-04174-JSW, Order Granting Motion for Class Certification and Denying Motion to Exclude Testimony of Christopher Vellturo, PhD, at 1–2, 7–10 (N.D. Cal. Sept. 23, 2026), ECF No. 329 (certifying the issuer class and allowing the proposed expert model).

[2]   Affinity Credit Union, No. 22-cv-04174-JSW, ECF No. 329, at 1–2, 4–6, 8 (describing the allegations, identifying common questions and defining the certified class).

[3]   Id. at 6–7 (describing standardized credit and debit fees, conditioning classwide impact on proof at trial and explaining the zero-fee yardstick model).

[4]   Visa, Visa Announces New Commercial Standard for Mobile Payments (May 28, 2015) (announcing Google's participation in a framework without pass-through fees between technology partners and financial institutions). Visa, Samsung Joins the Visa Digital Enablement Program (Aug. 13, 2015) (describing Samsung's participation in the same commercial framework). Consumer Fin. Prot. Bureau, Big Tech's Role in Contactless Payments: Analysis of Mobile Device Operating Systems and Tap-to-Pay Practices, § “The Android ecosystem” & n.40 (Sept. 7, 2023) (reporting zero issuer transaction fees for Google Pay and Samsung Pay). The CFPB's description cites contemporaneous reporting. These sources do not constitute a review of each issuer's executed agreements.

[5]   Affinity Credit Union v. Apple Inc., No. 22-cv-04174-JSW, Notice of Questions for Hearing, at 2 (N.D. Cal. Oct. 6, 2025), ECF No. 265 (summarizing the competing expert positions and asking about the markets implicated by the proposed zero fee). The notice cites the defense expert's report paragraphs 241 and 245 and Vellturo's reply report paragraphs 209–222. The account here follows the court's notice. The complete reports were not available for this article's review.

[6]   Comcast Corp. v. Behrend, 569 U.S. 27, 35–38 (2013) (requiring the damages model to measure harm attributable to the accepted liability theory and addressing conditions unrelated to the challenged conduct). Id. at 32–33 n.4 (distinguishing the certification question from the expert-admissibility issue the Court did not decide).

[7]   Alaska Rent-A-Car, Inc. v. Avis Budget Group, Inc., 738 F.3d 960, 969–70 (9th Cir. 2013) (affirming admission of an explained damages comparison despite differences between the comparator businesses and markets). Id., amended slip op. at 15–16 (describing the trial court's assessment of the supporting data and causal connection).

[8]   Fed. R. Evid. 702 (requiring the proponent to establish the specified reliability conditions by a more-likely-than-not standard). Fed. R. Evid. 702 advisory committee's note to 2023 amendment (explaining the court's responsibility to assess sufficient factual support and reliable application). Engilis v. Monsanto Co., No. 23-4201, slip op. at 13–16 & 14 n.9 (9th Cir. Aug. 12, 2025) (rejecting a presumption of admissibility and explaining that the result would be the same under either version of Rule 702).

[9]   Affinity Credit Union, No. 22-cv-04174-JSW, ECF No. 329, at 9–10 (treating Apple's objections to Vellturo as challenges to the weight of his conclusions and denying exclusion).

[10]  Affinity Credit Union, No. 22-cv-04174-JSW, ECF No. 265, at 2 (questioning the relationship between user lock-in, market definition and issuer injury).

[11]  See Two-Sided Market Structure Creates an Antitrust Blind Spot for Non-Paying Users, The American Counsel (Sept. 23, 2026) (examining the consequences of allocating platform charges among different participants). Affinity Credit Union, No. 22-cv-04174-JSW, ECF No. 329, at 5–7 (identifying the competitive-harm and impact issues that remain for common proof).

[12]  Apple, NFC & SE Platform for Secure Contactless Transactions (last visited Sept. 27, 2026) (describing eligible territories, commercial agreements and requirements for access to the platform). The public page identifies U.S. eligibility beginning with iOS 18.1 and requires applicable commercial terms. It provides no numerical fee schedule and does not establish the extent of entry or the effect on competitive prices.

Subscribe to The American Counsel

Don’t miss out on the latest issues. Sign up now to get access to the library of members-only issues.
jamie@example.com
Subscribe