Nevada Federal Court Denies Antitrust Conspiracy Based on Use of a Pricing Algorithm
See all Insights

Nevada Federal Court Denies Antitrust Conspiracy Based on Use of a Pricing Algorithm

Brownstein Client Alert, May 9, 2024

In what will be one of the most important decisions in antitrust since the proliferation of algorithm software, on May 8, 2024, Chief Judge Miranda Du of the United States District Court, District of Nevada, granted the hotel defendants’ motion to dismiss, with prejudice. Gibson v. MGM Resorts International, ECF No. 183, case no. 2:23-cv-00140-MMD-DJA (May 8, 2024) (“Order”). In this case, the plaintiffs, representing a class of hotel guests who stayed at various hotels on the Las Vegas Strip, alleged that the hotel defendants and a software company, Cendyn Group, LLC (“Cendyn”), conspired unlawfully to restrain trade in violation of Section 1 of the Sherman Antitrust Act, 15 U.S.C. Sect. 1, et seq, to fix prices.1 At issue was Cendyn’s software, which allegedly provides the hotel defendants two pricing recommendations. The plaintiffs alleged that the two products—GuestRev and GroupRev—incorporated a feature called RevCaster, which used a “rate shopper product” algorithm for collecting public rate information.

Chief Judge Du grounded her ruling on a finding that even if the algorithm resulted in pricing recommendations, nothing in the amended complaint alleged that the information shared from one competitor to the next was based on private information. More specifically, she found that since the plaintiffs alleged the defendants violated the Sherman Act by entering a hub-and-spoke conspiracy, consisting of a series of vertical agreements between Cendyn (the hub) and the hotel defendants (the spokes), it was critical that the plaintiffs alleged a tacit agreement between the hotel defendants to use Cendyn’s products in order to establish a “rim” to the conspiracy. Judge Du identified several deficiencies in the plaintiffs’ first amended complaint that undermined their allegation of a tacit agreement, and ultimately their case, including:

  • the failure to allege that the hotel defendants started to use the software around the same time, finding that while simultaneous adoption is not necessary, the fact that adoption by each defendant was spread out over the course of 10 years does not “raise the specter of collusion”;
  • the failure to allege that the hotel defendants agreed to be bound or to adopt the pricing recommendations from the software; and
  • the failure to allege the exchange of non-public information, finding that inference of exchanging nonpublic information—through machine learning techniques—is not sufficient. Indeed, this point is perhaps the most critical when viewed in the light of other antitrust algorithm cases in which there are allegations that the algorithm facilitated the exchange of non-public information. See e.g., In RealPage Inc., Rental Software Antitrust Litig., Case NO. 3:23-MD-03071 (M.D. Tenn. Dec. 28, 2023).

The plaintiffs also alleged a rule of reason claim, averring that the hotel defendants entered into vertical agreements with Cendyn to restrain trade. Yet again, Chief Judge Du found the plaintiffs failed to allege an agreement that bound the hotel defendants to the software’s pricing recommendations. Without an allegation that the hotel defendants are bound to accept the prices that GuestRev and GroupRev recommend, Chief Jude Du writes, it is “axiomatic” that there cannot be a vertical agreement. Order, 17.

This was the plaintiffs’ second opportunity to perfect their legal theory. Chief Judge Du previously dismissed the first complaint in October 2023, finding that it failed to allege facts sufficient to establish an agreement between the defendants. Gibson v. MGM Resorts International, 2023 WL 7025996 (D. Nev. Oct. 24, 2023). Notably, the court acknowledged the novelty of the case and some of the pleading challenges that go along with a new antitrust theory: “This case remains a relatively novel antitrust theory premised on algorithmic pricing going in search of factual allegations that could support it.” Order, 5. Nevertheless, Chief Judge Du still dismissed the amended complaint with prejudice, finding the allegations failing to present a plausible claim.

This opinion will be important for practitioners as they advise clients on the use of algorithms to gather pricing information. As Chief Judge Du clearly states, “consulting your competitors’ public rates to determine how to price your hotel room—without more—does not violate the Sherman Act.” The Order does not go so far as to alleviate all risk in use of machine learning as a tool to set prices. However, the Order does provide support to a contention that if the machine learning is simply used to improve the accuracy of an algorithm, and the algorithm continues to rely on public information, there is no Sherman Act violation. Hence, in order to assert a plausible claim, plaintiffs will need to establish that the machine learning is not only based on confidential, non-public information, but that it produces a recommendation that in itself is a biproduct of non-public information and/or that the algorithm pools non-public information together and bases its recommendation on the pooled information.

Chief Judge Du’s opinion starts to paint the picture of what plaintiffs must plead plausibly to allege an antitrust conspiracy based on an algorithm. Somewhat buried in Chief Judge Du’s opinion is an important footnote in which she disagreed with the Hotel Defendants’ argument that in order to allege a claim, plaintiffs must assert that all defendants used the same algorithm, in the same way, by selecting the same inputs. Judge Du opined that this overstates a plaintiff’s burden, noting that an algorithm can be customized but still delegate decision-making to a common source and if that common source relies on non-public information, that may be enough. A key takeaway is that even using a customizable algorithm with different inputs may not immunize users from an antitrust violation. As cases like this one proceed, practitioners and their clients will have a better understanding of how much delegation of decision-making to an algorithm can occur before they are at risk of an antitrust claim.

1 The undersigned authors represented Treasure Island, LLC, which was named as defendant in the litigation and dismissed with prejudice with the rest of the hotel defendants.

This document is intended to provide you with general information regarding Gibson v. MGM Resorts International. The contents of this document are not intended to provide specific legal advice. If you have any questions about the contents of this document or if you need legal advice as to an issue, please contact the attorneys listed or your regular Brownstein Hyatt Farber Schreck, LLP attorney. This communication may be considered advertising in some jurisdictions. The information in this article is accurate as of the publication date. Because the law in this area is changing rapidly, and insights are not automatically updated, continued accuracy cannot be guaranteed.

Recent Insights