Welcome to the Antitrust Antidote—a quarterly publication analyzing significant U.S. antitrust decisions from legal and economic perspectives. Authored by former Federal Trade Commission (FTC) enforcer Koren W. Wong-Ervin with former FTC economist co-authors Jeremy Sandford and Nathan Wilson, alternating each quarter. The title of this series, “Antitrust Antidote,” while mostly meant to be humorous (perhaps limited to those who have heard Koren’s “let’s talk economics” as a cure for a bad day), also refers to the practical guidance we aim to provide throughout the series. We hope you enjoy it!
***
There were a number of decisions from June through September 2025 including: (1) a fifth recent decision (U.S. et al. v. Apple) refusing to apply the more defense-friendly refusal-to-deal standard; (2) yet another lower court decision (involving the “famed” Hermès Birkin bag) declining to apply a per se illegal standard to a tying claim brought under both Sections 1 and 2 of the Sherman Act (despite the fact that tying by a firm with market power is still technically per se illegal under Supreme Court precedent); and (3) a number of collusion cases, including one relating to alleged algorithmic collusion.
(refusal-to-deal, market definition, specific intent to monopolize)
On June 30, 2025, a New Jersey federal court denied Apple’s motion to dismiss the DOJ’s and State Attorneys General’s (Plaintiffs’) complaint alleging that Apple illegally maintains a monopoly over smartphones by selectively imposing contractual restrictions on, and withholding critical access points from, developers, which serves to undermine apps, products, and services that would otherwise make users less reliant on the iPhone, promote interoperability, and lower costs for consumers and developers. According to Plaintiffs, Apple’s course of conduct (described as a “series of ‘Whac-A-Mole’ contractual rules and restrictions”) allows it to extract more money from customers, developers, content creators, artists, publishers, small businesses, and merchants, among others.
Anticompetitive Conduct: Rejection of Refusal-to-Deal Framework
The court rejected Apple’s argument that its allegedly exclusionary conduct (e.g., blocking innovative super apps, suppressing mobile cloud streaming services, excluding cross-platform messaging apps, limiting the functionality of non-Apple smartwatches, and limiting third-party digital wallets) amounted to no more than lawful refusals to deal. The court reasoned that “Plaintiffs do not allege Apple is failing to deal with its smartphone rivals but rather that Apple’s conduct is imposing restrictions on developers and smartphone users.” The court further stated that, “to the extent Apple argues it can limit access to its ‘own proprietary technology available to third parties,’ this contention is a factual dispute that must be resolved through discovery, not this stage of the litigation where factual disputes are construed in favor of Plaintiffs” (citing a decision relating to legitimate business justifications). For a discussion of the other recent decisions refusing to apply the refusal-to-deal standard, see Wilson & Wong-Ervin (2025).
The court concluded that Plaintiffs sufficiently alleged anticompetitive conduct in that the various types of conduct alleged amounted to “technological barriers,” stating that “[c]ourts have found technological barriers to constitute anticompetitive conduct,” citing the D.C. Circuit’s 2001 Microsoft decision and the Ninth Circuit’s June 2025 decision in CoStar Group v. Commercial Real Estate Exchange (another recent decision refusing to apply the refusal-to-deal framework).
The court credited the following allegations of “technological barricades that constitute anticompetitive conduct”:
- With respect to super apps, Apple imposes exclusionary requirements, such as “requiring apps in the United States to display mini programs using a flat, text-only list of mini programs” and banning apps from “categorizing mini programs,” which discourages developers from creating apps.
- For cloud streaming apps, Apple “[effectively prevents] third-party developers from offering cloud gaming subscription services as a native app on the iPhone,” as it requires any cloud streaming game or update thereto to “be submitted as a stand-alone app for approval by Apple,” which has increased the cost of releasing games on the iPhone and limited the number of games developers can make available to users.
- Apple imposes technological limitations to prevent developers from offering key features to iPhone users that are accessible on iMessage, Apple Watch, and Apple Wallet.
- With respect to iMessage, “Apple designates the APIs needed to implement SMS as ‘private/’ which prevents third-party developers from combining the ‘text-to-anyone’ functionality of SMS with the advanced features of OTT messaging.” In contrast, iMessage allows users to send messages to anyone by entering their phone number because it “incorporates SMS and OTT messaging.”
- Apple degrades the functionality of third-party cross-platform smartwatches through “its technical and contractual control of critical APIs” by preventing iPhone users from being able to respond to notifications on their third-party smartwatches, obstructing iPhone users from maintaining a reliable connection with third-party smartwatches, and undermining the performance of third-party smartwatches that connect directly with a cellular network.
- Apple Wallet is the only app that can use NFC hardware to provide tap-to-pay, which prevents other digital wallets from offering the ability to authenticate digital payment options during online checkouts, thereby limiting third-party wallets’ ability “to provide a simple, fast, and comprehensive solution to purchasing.”
The court also declined, at least at this early stage of the case, to dismiss allegations related to additional conduct not alleged as standalone violations but rather as part of an unlawful course of conduct. The court made clear that it would “look at the anticompetitive nature of Apple’s alleged conduct in total in its analysis.”
Market Definition: Smartphones & Performance Smartphones Limited to the U.S.
The court found that, for motion to dismiss purposes, “the performance smartphone constitutes a ‘distinct submarket[] [of the broader “smartphone” market] for antitrust purposes,’ which, in construing the factual allegations in a light most favorable to Plaintiffs, varies in price and quality from entry-level smartphones.”
The court rejected Apple’s argument that limiting the geographic market to the United States “mask[s] the worldwide primacy of Android smartphones.” The court concluded that Plaintiffs’ allegations are “supported by factual allegations of consumer behavior,” pointing to allegations that U.S. users “demand services offered by U.S. retailers when they purchase a smartphone,” such as “activating their new device, setting it up, and transferring important content like apps, messages, photos, and video to their new smartphone.” In contrast, if a customer purchases a smartphone abroad, the smartphone may (i) face compatibility issues with the domestic carrier, (ii) lack “the necessary radio technology to take advantage of the carrier’s highest speed connections,” (iii) limit the customer from accessing support during setup, or (iv) not have a valid warranty. The court also pointed to allegations that, if customers purchase the smartphones through a U.S. retailer, they will have access to “valuable promotions,” whereas the “same promotions and free financing are unavailable to U.S. consumers who purchase their phones in other countries.” In addition, numerous “regulatory requirements” must be met to enter the smartphone market, which effectively bars some smartphone makers from offering smartphones to United States consumers. Plus, Apple sets different prices “for the same smartphone in the United States separately from those in other countries.”
Specific Intent to Monopolize
The court rejected Apple’s argument that the attempted monopolization claims fail because Plaintiffs did not plead any facts evidencing a specific intent to monopolize. The court found that the complaint “includes numerous statements allegedly made by Apple executives regarding the barriers set in place to maintain its monopoly,” including:
- “Apple blocked cloud gaming apps that would have given users access to desirable apps and content without needing to pay for expensive Apple hardware” as it feared a world where “all that matters is who has the cheapest hardware” and consumers could “buy[] a [expletive] Android for 25 bux at a garage sale and . . . have a solid cloud computing device that ‘works fine.’”
- “Apple’s Senior Vice President of Software Engineering explained that supporting [over the top] messaging in [iMessage] ‘would simply serve to remove [an] obstacle to IPhone families giving their kids Android phones.’”
- Apple’s Senior Vice President of Worldwide Marketing “forwarded an email to CEO Tim Cook making the same point: ‘moving iMessage to Android will hurt us more than help us.’”
Tina Cavalleri, et al., v. Hermès (N.D. Cal.)
(tying)
On September 17, 2025, a California district court dismissed with prejudice a claim that Hermès engaged in unlawful tying by refusing to sell a customer “a famed Birkin bag unless they are ‘deemed worthy’ by buying other Hermès products such as shoes, scarves, jewelry, clothing, and the like.” The court concluded that Plaintiffs failed to “plausibly allege relevant product markets, defendant’s market power within those markets, or an injury that the antitrust laws were intended to prevent.”
Of note is the court’s reasoning in rejecting Plaintiffs’ “theory that all tying claims are a per se violation” of the Sherman Act, both Sections 1 and 2. While tying by a firm with market power is still technically a per se unlawful Section 1 violation under the Supreme Court’s 1984 decision in Jefferson Parish Hosp. Dist. No. 2 v. Hyde, the court in the Hermès case relied on the Supreme Court’s subsequent 2006 decision in Illinois Tool Works, Inc. v. Independent Ink, in which it stated: “[T]his Court’s strong disapproval of tying arrangements has substantially diminished,” as well as other Supreme Court precedent instructing that “[i]t is only after considerable experience with certain business relationships that courts classify them as per se violations.” The Hermès court concluded that: “Experience with the luxury handbag industry is not such that a presumption of per se liability is obvious, as plaintiffs would have it,” and “[n]othing” in Plaintiffs’ second amended complaint “establishes that the alleged practices with respect to the Birkin bag come within the narrow band of per se violations.”
The court went on to state that, even when viewed “through the lens of per se liability,” the complaint still fails for two main reasons.
First, Plaintiffs’ allegations that “Birkin bags constitute between 60% and 70% of the Elitist Luxury Handbag Market” lacked factual support. Plaintiffs relied on two papers about luxury good consumption from 2010 and 2014, which the court found to “simply describe[] consumer perceptions about product quality,” rather than to adequately define a market under the hypothetical monopolist test. The court also noted that high shares alone are not sufficient.
Second, Plaintiffs failed to allege anticompetitive effects. The court reasoned that the “‘tied market’ alleged here is a kaleidoscope of products covering everything from ‘scarves and shawls, ready to wear clothing, footwear, watches, jewelry, fragrances, accessories (including hats, gloves, ties, and sunglasses)’ to ‘home goods such as tableware, furniture, blankets, and decorative objects like vases and trays,’” yet the amended complaint “is bereft of any facts that might support lumping such a hugely diverse array of non-substitutable products into a single market, and equally devoid of facts indicating that competition for these goods has been illegally restrained by Hermès.”
As a matter of economics, tying can harm competition to the extent it denies scale to rival producers of the tying good, causing them to exit, to scale back product development, or to not enter in the first place. In general, such effects are possible only if the act of tying shifts purchases toward the tied product—in the present case, towards Hermès’s scarves, clothing, footwear, and other products—and away from rival producers of the product at sufficient scale that the rival producers can no longer cover their fixed costs and must exit (or otherwise scale back). On the other hand, tying that merely transfers some sales from one firm to another without weakening a competitor (by depriving them of minimum-efficient scale) is unlikely to reduce consumer welfare. Plaintiffs in the present matter appear to have struggled to articulate why Hermès’s conduct would have lessened the competitive constraints on Hermès by causing a competitor to exit, or by otherwise weakening a competitor.
A recent Wall Street Journal article estimated that Hermès produces only 120,000 Birkin and Kelly bags per year, worldwide. Even if Birkin bags account for the majority of this quantity, and even if the sale of Birkin bags were conditioned on purchase of a significant number of additional Hermès items, this conduct would only have the potential to divert a limited volume of commerce away from Hermès’s competitors. By way of reference, a Statista estimate indicates that women’s apparel accounted for nearly $200 billion in U.S. sales alone. It seems unlikely that the tying of a small number of Birkin bags to other Hermès apparel shifts purchases among this vast volume of commerce to a degree sufficient to drive competitors out of business and to materially weaken the competitive constraints faced by Hermès.
Hanson Dai, et al. vs. SAS Institute Inc., et al. (N.D.Cal.)
(price fixing, algorithmic collusion)
On July 18, 2025, a California district court dismissed with leave to amend a price fixing suit against Hyatt and other hotels alleging that they had conspired to fix hotel room prices nationwide and in 17 distinct geographies through their use of revenue management and profit optimization software provided by IDeaS, a subsidiary of SAS Institute Inc.
Plaintiffs alleged that the hotel Defendants provided IDeaS with confidential information that could be directly accessed and examined by competitors so as to optimally set prices and potentially discipline participants not adhering to the scheme. In addition, Plaintiffs alleged that IDeaS has facilitated the indirect sharing of non-public information through its algorithmic provision of updated pricing recommendations in response to changes in the information it receives from the hotel firms.
The court notes that the allegations are not without precedent, and that Defendants do not dispute their use of the IDeaS software. However, the Court ultimately concluded that the Plaintiffs had not met their burden of establishing that Defendants’ common usage of the platform was sufficient to conclude that the Defendants’ behavior could only reasonably be explained as the result of an agreement as opposed to independent, self-motivated decision-making. Specifically, the Court explained that Plaintiffs have not provided facts about the timing of different Defendants’ decisions to use IDeaS. Nor did they lay out how and why subsequent decisions were unprecedented deviations from historical norms.
Beyond the lack of facts surrounding defendants’ decisions to use the software and what impacts this may have had on their pricing, the Court also notes that the plaintiffs do not reasonably identify “plus factors” that would support a conclusion that the defendants’ conduct could only be explained by an illegal agreement. For example, the plaintiffs were said not to have identified an invitation to collude. Moreover, while the Complaint emphasized that non-public information was exchanged, the Court found little persuasive evidence that this was in fact occurring.
As laid out above, the economics of price fixing are fairly clear. When rivals can reach agreement to deviate from self-interested behavior, it often harms consumers through reduced output and higher prices. However, in order to credibly conclude that shared behaviors represent an anticompetitive conspiracy, facts are needed to understand why self-interest alone cannot explain all parties’ decisions. Two neighbors need not have agreed in order to both decide to bring umbrellas with them on an overcast day.
Isabel Litovich, et al. vs. Bank of America, et. al. (S.D.N.Y.)
(price fixing)
On September 2, 2025, a New York district court dismissed with prejudice a conspiracy suit against Bank of America, Merrill Lynch, and other major financial institutions active in the corporate bond market. The suit alleged that Defendants conspired to suppress the secondary market for corporate bonds.
The plaintiffs are investors in the corporate bond market who had bought and sold “odd lots” of bonds in the secondary market directly to and from defendants. Odd lots consist of trades that involve less than 1,000 bonds or less than $1 million in par value. “Round lots” are larger trades, involving more than 1,000 bonds or greater than $1 million in par value. Individuals tend to trade in odd lots, while institutional investors trade in round lots. Defendants are dealers for both, making money by selling bonds at a price higher than they’re willing to buy them. This difference is known as the “spread.” Defendants are alleged to collectively account for nearly two-thirds of the bond underwriting market and roughly 90 percent of bond trading volume. Defendants are alleged to have higher (25 to 300 percent) spreads on odd lots than round lots.
Plaintiffs allege that Defendants conspired to preserve this gap in spreads between odd and round lots. Specifically, defendants are said to have (1) punished dealers whose behavior could have eroded the spread gap; (2) acquired control of trading platforms to ensure they did not act to close the spread gap; and (3) agreed to boycott retail-oriented electronic trading platforms which emphasized transparency.
The complaint provides two examples of behavior they see as indicative of the defendants’ focus on punishing dealers that offered retail investors lower spreads. One of the two took place in the 1990s and the other sometime between 2015 and 2019. The complaint also notes a number of different entities that defendants are alleged to have taken over or otherwise co-opted in order to prevent them from closing the spread gap.
The court concluded that the plaintiffs had failed to adequately substantiate the plausibility of their claims. Fundamentally, the Court’s logic was that similar behavior by similarly situated firms is not on its own sufficient to establish the likelihood of a conspiracy. For example, the Court notes the absence of facts to support the allegation that the defendants had acted to punish dealers whose behavior could have reduced the spread gap. Furthermore, plaintiffs are said to have failed to provide facts sufficient to show that the choices made by defendants could only plausibly be explained through coordination as opposed to unilateral self-interest. In particular, the Court notes the importance of recognizing that companies may rationally choose not to invest in or assist entities whose business models would have the effect of subverting their own.
Even if the plaintiffs had not lacked facts sufficient to support their allegations, the court concluded that their allegations were time-barred. The case was filed in 2020, and no specific price fixing conduct was precisely attributed to have taken place within the prior four years as required. The conduct that allegedly took place between 2015 and 2019 was insufficiently detailed to establish that it was within the necessary window. Furthermore, the Court rejected the argument that the four-year window should be ignored owing to defendants’ fraudulent concealment of their bad acts, once more citing an insufficient amount of factual support.
Overall, the decision emphasizes that it is not sufficient to note that common actions by competitors may have redounded to the detriment of other actors. In order to sufficiently allege antitrust harms, the counterfactual world wherein the competitors focus exclusively on their individual self-interest has to be described credibly, drawing on facts.
Portillo et al. v. CoStar Group, Inc. (W.D. Wash.)
(collusion)
On August 29, 2025, a Washington district court dismissed with leave to amend a price fixing suit brought by seven plaintiffs who had purchased hotel rooms against CoStar and a group of hotel operators. The suit alleges that benchmarking reports supplied to the hotels by Smith Travel Research (STR)—a subsidiary of CoStar—facilitated price fixing between the hotels, resulting in supra-competitive prices.
The benchmarking reports at issue contain aggregated information about the average price and quantity of hotel rooms. The nature of this information and the level of its aggregation were disputed: Defendants argued that STR’s benchmarking reports were “aggregated, averaged, and anonymized,” while Plaintiffs alleged that it was possible to reverse engineer the STR reports to identify the data provided to STR by a specific hotel, and that the STR reports thereby amounted to “price information” shared with competitors. However, the court stated that any dispute over whether STR’s data could be de-anonymized was immaterial, since Plaintiffs had not alleged that the de-anonymized data identified any hotel’s price for specific rooms on specific dates, as opposed to prices averaged across time periods and across all of a hotel’s rooms. As a matter of economics, average prices would not typically be adequate to support collusion, given that hotel prices typically change day by day.
As the court described, hotels receiving STR’s benchmarking reports agree to provide information to STR on: (1) backward-looking data on the quantity of rooms it sold and the quantity of rooms it had available; (2) forward-looking data on the quantity of rooms booked in the future and the number of available rooms; and (3) aggregated revenues and costs by “hotel operating department and undistributed cost centers.” While Plaintiffs characterized this information as “price information,” the court concluded that this term was a “linguistic stretch,” because the STR reports (even if de-anonymized) do not allow for the identification of actual prices for specific rooms on specific dates. Thus, the court concluded that the Plaintiffs’ allegations fell short of the standard set by cases Plaintiffs cited in their complaint: Container Corp. of Am., 393 U.S. 333 at 335 (1969), which found a Section 1 violation on the basis of defendants exchanging information on “the most recent price charged,” and In re Coordinated Pretrial Proc. in Petroleum Prods. Antitrust Litig., 906 F.2d 432, 445–47 (9th Cir. 1990), which described the term “price information” as consisting of actual prices charged.
The court also rejected Plaintiffs’ claimed parallel conduct. While Plaintiffs alleged that competitors exchange “real-time […] pricing and supply data,” they failed to explain why the aggregated information shared with STR amounts to information on actual prices, and thus doubted that “alleged parallel conduct of contracting with STR would be of any significance.” The court also rejected Plaintiffs’ economic analysis, which purported to show that Defendants “have been able to set higher prices compared to other luxury hotels.” This analysis appears to have been premised on the use of STR data as inputs to pricing algorithms, and on those algorithms allowing hotels to charge higher prices than they would otherwise. However, the court found that STR reports were one of many inputs used by hotels (and by pricing algorithms) in setting prices, and thus that any connection between STR reports and hotel pricing was “too speculative and attenuated” to demonstrate that the use of STR reports allowed hotels to charge higher prices.
The court additionally rejected Plaintiffs’ hub-and-spoke conspiracy theory on the grounds that Plaintiffs failed to support an inference of an agreement between the alleged hub (STR) and the spokes (hotels) given that neither the hub nor the spokes appeared to receive pricing information from each other.
The court granted leave to amend, stating that while Plaintiffs’ theory appeared speculative and conclusory, it was sufficiently complex that it might be clearer if alleged in a different way, and, in particular, if Plaintiffs were clearer as to the nature of the “price information” they alleged was exchanged between hotels via STR.
It is worth pausing to consider the economics of price fixing agreements. Economics predicts that when competing firms interact repeatedly, with no clear end date, the set of equilibrium outcomes expands to include both competitive outcomes (i.e., the outcome that would occur if the firms prioritized short-run profits) and collusive outcomes. The latter are possible because firms may be willing to sacrifice short-run profits (by competing less vigorously) in order to bolster future profits (by having rivals competing less vigorously). A necessary condition for collusion to emerge is that firms have the ability to monitor rivals’ prices, as otherwise each firm individually has an incentive to “cheat” by underpricing its rivals, without its rivals detecting the cheating. Because collusion is only possible if firms can monitor rivals’ prices, antitrust enforcers have scrutinized arrangements under which firms exchange information with each other, including those involving a third party, such as a trade association or data provider. However, as the CoStar court pointed out, to reasonably sustain collusion, the information shared must give each participant actionable information on the pricing of its rivals, and we agree that it is unclear that the aggregated and averaged information provided by STR reports clear that bar. One reason for this is that it is not clear why the aggregated information in STR reports would give hotels more granular information than could be obtained by monitoring rivals’ list prices, which are commonly posted online. Another is that the long time periods over which data in STR reports are aggregated make it hard for rivals to promptly detect cheating, which makes cheating more attractive.
Koren W. Wong-Ervin, Jeremy Sandford & Nathan Wilson [1]
[1] This issue was written by Koren W. Wong-Ervin (Partner at Jones Day & Senior Fellow at the George Washington University Competition & Innovation Lab); Jeremy Sandford (Partner at Econic Partners); and Nathan Wilson (Founding Partner at Econic Partners). The authors thank Ellen Geyer for her research assistance. The views and opinions set forth herein are the personal views or opinions of the authors; they do not necessarily reflect the views or opinions of the organizations with which they are affiliated, or those organizations’ management, affiliates, employees, or clients.
***
