McDonald’s and its ubiquitous restaurants face an algorithmic pricing antitrust test.
McDonald’s franchisees are independent businesses that can compete with one another and, according to McDonald’s, set their own menu prices. A new antitrust lawsuit asks whether those pricing decisions remain truly independent when franchisees supply sales data to a common machine-learning system that recommends what they should charge.
That question is at the center of a proposed class action filed Oct. 2, 2026, in federal court in Chicago, which accuses McDonald’s and its independently owned franchisees of using a centralized pricing system to coordinate menu prices in violation of Section 1 of the Sherman Act. McDonald’s disputes that its system displaces independent franchisee pricing (Thomas v. McDonald’s USA, LLC, No. 1:26-cv-12149, N.D. Ill.).
On Oct. 15, a second proposed class action, Rossi v. McDonald’s Corporation, No. 1:26-cv-12651 (N.D. Ill.), made similar claims but added Deloitte Consulting LLP as a defendant, which the suit alleges operated the franchisee-facing layer of the pricing system. This article addresses the Thomas case first then turns to what the Rossi case adds.
At first glance, the cases may sound like more testing of old antitrust law’s application to artificial intelligence, but the technology may be the least novel element.
The more fundamental questions are familiar ones:
- When are separately owned businesses acting independently, and when are they acting together?
- And can competitors cross that line by funneling confidential information through a common intermediary rather than exchanging it directly?
Are Franchised Restaurants One Economic Actor or Many?
Approximately 95% of McDonald’s U.S. restaurants are franchised. The franchisees are separately owned legal entities and are characterized in the McDonald’s franchise agreement as independent contractors. They aren’t guaranteed exclusive territories and may compete for customers with other McDonald’s franchisees as well as company-owned restaurants.
McDonald’s has described intramural competition as intentional. In a 2023 submission to the Federal Trade Commission (quoted in the complaint), it said it does not grant exclusive territories because competition from nearby locations gives franchisees an incentive to maintain quality. It has also repeatedly said franchisees control their own menu prices. In 2024, McDonald’s USA President Joe Erlinger said franchisees “set menu prices for their restaurants,” and at a September 2026 investor presentation, Global Chief Marketing Officer Morgan Flatley said they “retain control over independent pricing decisions.”
Independence is an important factor in antitrust law. In American Needle, Inc. v. NFL, 560 U.S. 183 (2010), the Supreme Court rejected the idea that participants in a common enterprise necessarily constitute a single entity for purposes of Section 1’s prohibitions against anticompetitive coordination and trade restraints. The inquiry focuses on the challenged conduct and whether it joins together separate economic decisionmakers who otherwise could compete.
The principle has been applied to burger franchises before. In Arrington v. Burger King Worldwide, Inc., 47 F.4th 1247 (11th Cir. 2022), the Eleventh Circuit concluded that Burger King and independently owned franchisees were capable of concerted action under Section 1 in a case challenging restrictions on hiring among restaurants in the chain.
These cases raise the question whether a simple “we’re all McDonald’s” answer is enough. The relevant inquiry is what decisions the restaurants make independently and whether the challenged arrangement interferes with that independence.
Franchise Coordination Is Not Automatically Unlawful
Other Supreme Court cases are worth examining. In Continental T.V. v. GTE Sylvania and later Leegin Creative Leather Products v. PSKS, the court recognized that restrictions imposed within legitimate manufacturer-dealer relationships can promote competition and generally should be evaluated under the rule of reason rather than condemned automatically. Leegin extended that principle to vertical minimum-price restraints (Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36 (1977) and Leegin Creative Leather Products, Inc. v. PSKS, Inc., 551 U.S. 877 (2007)).
Key questions in cases like this include:
- Are such pricing tools part of a legitimate vertical franchise relationship, providing nonbinding guidance intended to improve the performance of individual restaurants and the franchisor’s system as a whole?
- Does such a tool constitute vertical assistance from a franchisor to independently owned franchisee, or a horizontal mechanism through which competing franchisees coordinate prices?
The Data Flows Up to McDonald’s — and the Alleged “30% Rule”
McDonald’s calls franchisees independent price-setters, but substantial information about their businesses flows back to it. The Thomas complaint quotes McDonald’s Franchise Disclosure Document as saying that transaction-level point-of-sale data, including sales, transactions and product mix, is held on McDonald’s servers with “no contractual limits on its right to access such information,” and says franchisees also submit monthly profit-and-loss statements.
Collecting that data is not inherently anticompetitive; what plaintiffs say happens next is more consequential. They allege a machine-learning “pricing engine” generates restaurant- and menu-item-level recommendations “[i]nformed by insights from millions of transactions and a broad range of market data,” and, relying on public reporting, that it employs a so-called “30% rule”: the system allegedly considers recommending a price increase for an item when at least 30% of stores recently raised their prices for that item.
If established, that would describe independent restaurants providing nonpublic data to a common intermediary that pools it and sends pricing recommendations back to the businesses that supplied it. McDonald’s rejects that characterization and has not yet filed a formal answer. “AI does not set menu prices at McDonald’s restaurants – McDonald’s franchisees do,” a spokesperson told the Associated Press, adding that the optional tools “do not automate, coordinate or fix pricing in any way.” McDonald’s called the lawsuit “filled with inaccuracies” and said it would vigorously defend itself.
Where Is the Agreement?
Whether there was an agreement among rivals is central to cases of this kind. Section 1 requires concerted action, and sophisticated software doesn’t eliminate that requirement. Imagine that every franchisee separately sends its sales information to McDonald’s, which analyzes the data and gives each restaurant a recommendation designed to increase its revenue, and each franchisee remains free to accept or reject the advice. Even if virtually every restaurant uses the same system, that does not necessarily establish that the franchisees agreed with one another to coordinate prices.
But an agreement may not need to be solely an agreement on prices. One theory is that the relevant agreement is to participate in the system itself: each competitor knowingly contributes nonpublic data to a common intermediary that pools it and sends competitor-informed outputs back to the same competitors, whether or not they agreed on any particular price.
Some courts have been receptive to that framing when pleadings allege pooled confidential data. In Duffy v. Yardi Systems, Inc., 758 F. Supp. 3d 1283 (W.D. Wash. 2024), the court found a plausible agreement where competitors allegedly traded commercially sensitive information for pricing recommendations, and the Ninth Circuit in Gibson v. Cendyn Group, LLC, 122 F.4th 1058 (9th Cir. 2025), cert. denied, No. 25-1109 (U.S. Apr. 20, 2026) noted that its analysis might differ if the algorithm pooled competitors’ data. See sidebar for more details.
The question raised in algorithmic pricing cases is whether competitors, in this case McDonald’s franchisees, know competitors use the same engine, agree to use its output rather than price independently, and face pressure to adhere to it.
Are the Recommendations Optional?
McDonald’s says franchisees retain final pricing authority, but the complaint alleges it does more than make its pricing system available. Franchisees must be “constructively engaging with McDonald’s approved Pricing Consultant and Tools,” and McDonald’s records and follows up when franchises deviate from recommended pricing, the complaint says, alleging there are consequences for franchise owners who want to renew or expand.
Per Se or Rule of Reason?
The complaint characterizes the arrangement as horizontal price fixing that is per se unlawful, that it should be condemned without an in-depth rule-of-reason examination of any competitive effects. A quick-look analysis is all that’s needed, plaintiffs maintain. A naked agreement among unrelated restaurants to fix hamburger prices would pose an easy question, but a franchisor supplying pricing analytics inside an extensive and otherwise legitimate economic relationship is more complicated. Questions around whether the business owners agreed to a pricing system, fed it with confidential information, and followed the recommendations are at issue. If the system merely provides advice then a more detailed examination may be called for.
The question in algorithmic pricing cases is less about the algorithm and more about how people agreed to use these tools.
What Does the Rossi Complaint Add?
The Rossi complaint defines a class of anyone who bought menu items at a U.S. McDonald’s restaurant, company-owned or franchised, from Jan. 1, 2019, and alleges a conspiracy beginning no later than 2019. Its distinguishing feature is the role it assigns to Deloitte Consulting LLP (Rossi v. McDonald’s, et al, No. 1:26-cv-12651, N.D. Ill., Eastern Div.).
According to the complaint, Deloitte did not simply license software. It allegedly developed the “Strategic Pricing Analyzer” that is the “McDonald’s Pricing Analyzer” franchisees use to review recommendations, and franchisees were allegedly using it by April 2019. Deloitte’s “Pricing Advisors” serve restaurants within assigned regions, adjust recommendations for local market events, and present pricing trends at regular franchise meetings, the complaint states, adding that Deloitte is the “approved Pricing Consultant” McDonald’s has required franchisees to engage with since January 2026.
Plaintiffs characterize Deloitte as an independent co-conspirator, not merely McDonald’s agent, and allege that its compensation was tied to the incremental profit the tool generated. The complaint also describes Tiger Analytics: according to two former employees, it now runs the pricing engine, with McDonald’s supplying the rules and corporate targets. The complaint acknowledges that the public record does not disclose how the work of Deloitte, data analytics and AI consulting firm Tiger Analytics, and McDonald’s internal team has been divided.
Rossi asks the court to bar the use of one restaurant’s nonpublic data to generate recommendations for its competitors and to bar consideration of pricing compliance in renewal and growth decisions.
Conclusion
The central question is whether McDonald’s and its franchisees agreed to substitute a shared pricing mechanism for independent competitive decision-making. Other cases suggest that this is a different inquiry from whether competitors shared data and used the same software. The case is nominally about artificial intelligence, but the central antitrust question predates machine learning by generations: Are competitors determining prices independently, or are they determining them together? AI may simply provide a new way to ask that question.
/>i
