Algorithmic “Collusion” Theories: Check Out Time or Extended Stay? Ninth Circuit Affirms Complaint Dismissal As Algorithm Scrutiny Intensifies

On August 15, 2025, the Ninth Circuit affirmed dismissal of a putative class action alleging Las Vegas Strip hotels used Cendyn Group’s revenue management software to artificially inflate prices in violation of Section 1 of the Sherman Act. In rejecting plaintiffs’ claims that the agreements between Cendyn and other defendants were anticompetitive, the Ninth Circuit reinforced that there must be a causal link between a contested agreement and a restraint of trade, which plaintiffs failed to allege. The court’s ruling—the first appellate case to address a challenge to pricing algorithms—should provide some guidance as scrutiny of pricing algorithms increases across numerous industries.

Independent Choice to License Software ≠ Anticompetitive Agreement

Although plaintiffs had originally alleged that the defendants were engaged in a hub-and-spoke conspiracy, plaintiffs narrowed their appeal to focus on Cendyn’s licensing agreements with the hotel operators as the basis for a Sherman Act violation. The court found these agreements were “ordinary sales contracts” that don’t typically restrain trade, distinguishing them from horizontal agreements (because Cendyn and the hotel defendants are not competitors) and vertical agreements (because Cendyn and the hotel defendants do not operate at different levels of distribution for the same end product).

The court concluded that the independent agreements between Cendyn and defendants could not form the basis of a Sherman Act violation, regardless of the number of agreements:

• While acknowledging that “an agreement” among hotel defendants “to follow Cendyn’s pricing recommendations would harm competition,” the court found that defendants’ “independent choices to use Cendyn’s software products [were not] anticompetitive.”[1]   
• Nothing in the Cendyn licensing agreement required a defendant to follow any recommendations or restricted any defendant’s ability to set its own rates.[2]
• There were no allegations that Cendyn shares or uses confidential information provided by a defendant for its pricing recommendations to any other defendant.[3]
• The allegation that the hotel defendants were “aware of their competitors’ adoption and use of Cendyn’s software products” was not sufficient to allege a violation.[4]

The court’s reasoning reflects other cases where plaintiffs’ allegations regarding pricing software were held to be insufficient—including a challenge to the use of Cendyn software in Atlantic City dismissed last year[5] and another case against national hotel chains over their use of software provided by IDeaS, dismissed in July 2025.[6] Those decisions likewise focused on the lack of connection between hotels’ choice to license software and increased prices, particularly where there were no allegations that the software used competitors’ confidential information when generating price recommendations.

Algorithmic Collusion Cases Show No Signs of Slowing Down

The Ninth Circuit’s decision comes at a time when algorithmic collusion cases from multiple industries are advancing in courts around the country. Below are some notable updates on these cases.

Greystar Settles RealPage DOJ Case.

In August 2025, the U.S. Department of Justice (DOJ) reached a settlement with Greystar, a property manager defendant in DOJ’s suit challenging the use of revenue management software by residential property managers. DOJ and several states allege that RealPage’s pricing algorithm relies on non-public data submitted by competing property managers to generate pricing recommendations and that property managers regularly follow these recommendations, purportedly aligning and inflating rental rates.  

Notably, the settlement prohibits Greystar from using any “Revenue Management Product” that uses nonpublic data from properties managed by others to generate rent recommendations for Greystar or that uses Greystar’s non-public data to generate recommendations for others. Nonpublic data (per the settlement) includes past, present, or prospective information that could be used to determine inventory, rents, or other rental terms. Greystar also may not use any Revenue Management Product that incorporates a rental price floor or other incentive to follow the tool’s recommended prices and may not agree with any other property owner or manager to use a particular Revenue Management Product.  

Insurers Face Hub & Spoke Claim Over Reimbursement Platform.

A case involving Multiplan (a platform that benchmarks payment rates for out-of-network medical providers) and several of the countries’ largest insurers will proceed to discovery this year.[7] In July 2025, the court denied defendants’ motions to dismiss. The court found the complaint sufficiently alleged a “hub-and-spoke” conspiracy among competing insurers because it alleged it was against each payor-defendant’s self-interest to use the tool absent an assurance that its competitors would follow. Plaintiffs also alleged that MultiPlan served as a go-between for insurer defendants, including allegations of specific communications about competitors’ rate-setting practices.  

Key Takeaways

The Ninth Circuit’s decision, coupled with recent decisions and settlements from lower courts, helps crystallize some of the key facts to consider when evaluating the use of pricing algorithms:

• Competitors can use the same pricing software, and the mere fact that they are aware of other competitors’ use of the software does not lead to an antitrust violation.
• Pricing software poses less risk when recommendations do not have to be followed and when the user retains the ability to make independent decisions about pricing.
• Pooling or sharing of competitively sensitive information (including past, current, or future, and individual or aggregated data regarding prices, inventory, customers, or other inputs) to generate recommendations creates significantly more risk.

Companies relying on revenue management software with pricing algorithms should continue to assess how their software stacks up to mitigate risk.

[1] Gibson Slip Op. at 18.

[2] Gibson Slip Op. at 23.

[3] Gibson Slip Op. at 9.

[4] Gibson Slip Op. at 19.

[5] Cornish-Adebiyi et al. v. Caesars Entertainment, Inc. et al., Case No. 23-cv-02536-KMW-EAP 2024, WL 4356188 (D.N.J. Sept. 30, 2024).

[6] Hanson Dai et al. v. SAS Institute Inc., et al., Case No. 24-cv-02537-JSW, 2025 WL 2078835 (N.D. Cal. July 18, 2025).

[7] In re. Multiplan Health Insurance Provider Litigation, 24-cv-07695 (N.D. Ill.).