Harrison (Buzz) Frahn, Thomas M. Cramer, Lily Cron (Summer Associate)
Simpson Thacher & Bartlett LLP
In an 5-4 opinion authored by Justice Clarence Thomas, the Supreme Court ruled on June 25, 2018, that American Express’s (“Amex”) practice of using anti-steering provisions, which prevent merchants from offering customers incentives to use credit cards with lower fees, is not an “unreasonable restraint on trade” in violation of Section 1 of the Sherman Act. The credit-card industry is a two-sided market, meaning that in a single transaction, credit-card companies serve both cardholders and merchants. The Supreme Court found that because of their interdependence, the two groups of customers make up a single market to be assessed together. In applying the “rule of reason” to evaluate the anti-steering provisions’ effects on the credit-card market as a whole, the Court found a lack of evidence of anticompetitive effects, instead determining that the provisions encourage competition and innovation.
The credit-card industry, predominantly controlled by Amex, Visa, MasterCard, and Discover, operates as a “two-sided platform,” meaning that it serves and intermediates two distinct groups: cardholders and merchants. Credit-card companies provide different services to each group, credit and rewards to cardholders and transaction processing and guaranteed payment to merchants. The two sides are mutually dependent because neither can receive these services without a single common transaction provided by the credit-card company. The link between the two groups causes “indirect network effects,” in that the value of the credit card to the cardholder depends on how many merchants accept the card and, in turn, the value of the credit card to the merchant depends on the number of cardholders. If merchant fees are too high, merchants will not accept the card. If merchants do not accept the card, or if the card offers suboptimal rewards, cardholders will choose not to use that card, which further incentivizes merchants to not accept it. Therefore, the credit-card platform must optimize its pricing on both sides, as to not create a “feedback loop of declining demand.” Ohio v. Am. Express Co., No. 16-1454, 2018 U.S. LEXIS 3845, at *1 (2018). This price optimization often results in disproportionate pricing between the two sides in order for credit-card platforms to offer good value on their services and remain competitive. In Amex’s case, this means high merchant fees and no interest fees for cardholders.
The Court explained how Amex’s business model differs from that of its main competitors, Visa and MasterCard. Whereas Visa and MasterCard offer limited rewards and charge merchants lower fees, appealing to a broad market of both sets of customers, Amex offers superior rewards for cardholders, which encourages more spending and attracts wealthier cardholders. Because those wielding an Amex are willing to spend more, Amex can charge merchants higher fees.
Recognizing Amex cardholders’ willingness to spend more, merchants attempted to avoid Amex’s higher fees by utilizing “steering” techniques, which incentivize cardholders to use other payment methods at the point of sale. “A merchant might tell the customer, for example, ‘American Express costs us more,’ or ‘please use Visa if you can,’ or ‘free shipping if you use Discover.’” Id. at *37. The merchant still gets the business of the Amex cardholder, but avoids the high merchant fee. To prevent this practice, Amex put anti-steering provisions in its contracts with merchants that disallow merchants from engaging in steering. Amex’s business model has shaped the credit-card industry, as competitors like Visa and MasterCard now offer similarly structured premium cards.
Government Asserts That AMEX’s Anti-steering Provisions Have Anticompetitive Effects
In October 2010, the United Stated Department of Justice, along with several individual states (collectively, “Plaintiffs”), brought a claim in the Eastern District of New York asserting that Amex’s anti-steering provisions violate Sherman Act Section 1 by imposing an “undue restraint on trade.” The district court determined that the cardholder and merchant markets were distinct, holding Amex’s anti-steering provisions to be an anticompetitive exercise of market power because they led to higher merchant fees. The Second Circuit reversed, holding that the two sides of the credit card market made up a singular market, in which they found no anticompetitive effects resulting from anti-steering provisions, reversing the district court’s determination.
To succeed on a Section 1 claim, Plaintiffs were required to prove that Amex’s anti-steering provisions are an “unreasonable restraint on trade” under State Oil Co. v. Kahn, 522 U.S. 3, 10 (1997). The Supreme Court did not find the provisions to be per se unreasonable, which is typically the case for agreements between competitors, known as horizontal restraints on trade. The Court determined instead that the anti-steering provisions are a vertical restraint on trade because they are agreements between businesses at various levels of distribution, making them not a per se violation.
The Court therefore applied the “rule of reason,” a “three-step, burden shifting framework” requiring a fact-specific assessment of the restraint’s effect on market competition “to distinguish between restraints with anticompetitive effect that are harmful to the consumer and restraints stimulating competition that are in the consumer’s best interest.” Id. at *17 (quoting Leegin Creative Leather Products, Inc. v. PSKS, Inc., 551 U.S. 877, 866 (2007)).
In assessing the first step, the Court found that Plaintiffs did not meet their burden of proving “that Amex’s anti-steering provisions have a substantial anticompetitive effect that harms consumers in the relevant market.” Id. at *3. If Plaintiffs had met the burden of the first step, the burden would shift “to the defendant to show a procompetitive rationale for the restraint,” after which the burden would shift back to the plaintiff to prove that “the procompetitive efficiencies could be reasonably achieved through less anticompetitive means.” Id.
Broad Market Definition Did Not Allow for a Finding of “Unreasonable Restraint on Trade”
The Court’s holding turned on the definition of the relevant market. The Court explained that a combination of products and services can together constitute a single market, the “arena within which significant substitution in consumption or production occurs.” Id. at *19 (quoting United States v. Grinnell Corp., 384 U.S. 563, 571 (1966)). The interdependent nature of pricing and resulting indirect network effects of Amex’s two-sided platform led the Court to determine that price increases on one side of the platform would only be anticompetitive if it “increased the overall cost of the platform’s services.” Id. at *20. Therefore, given the commercial reality that the credit-card industry operates as a single market, both the merchant and cardholder aspects of Amex’s business must be assessed together as a single market.
The Court distinguished the two-sided credit-card transaction platform from other two-sided platforms with less severe indirect network effects, where markets should be assessed separately. The Court offered the example of newspapers, which are a two-sided platform serving both readers and advertisers. The two customer groups are somewhat interdependent because as readership increases so does the price of advertising space. However, these effects only flow one way, as readership is not substantially affected by the amount of advertising, making the services provided to the two sides of the platform distinguishable. A newspaper lacks the single transaction seen with credit cards that simultaneously serves both sides of the market fueling the feedback loop. Unlike newspapers, the transactional nature of credit cards requires the network to “find the balance of pricing that encourages the greatest number of matches between cardholders and merchants,” establishing a single market. Id. at *21-22. The Court also emphasized that viewing the market as a whole is necessary to accurately assess competition, as only a credit-card company selling transactions to both merchants and cardholders could compete.
In a dissenting opinion, Justice Stephen Breyer argued that market definition is not always required and altogether disagreed with the majority’s determination that there is a singular two-sided credit-card market. In Justice Breyer’s view, the two sides are merely complementary, rather than cohesive. Further, Justice Breyer argued that, even accepting the majority’s market definition, anticompetitive effects were proved by the government.
The Government’s Evidence Was Deemed Insufficient to Overcome Rule of Reason Analysis
The government argued that Amex’s anti-steering provisions are anticompetitive because they increase merchant fees, asserting three theories, each of which the Court rejected. The Court found that the government’s direct evidence failed to demonstrate that Amex’s anti-steering provisions “increased cost of credit-card transactions above a competitive level, reduced the number of credit card transactions, or otherwise stifled competition in the credit-card market.” Id. at *24-25. Thus, the Court found that the government failed to meet the burden of the first step of the “rule of reason.”
The Court found that “the plaintiffs did not offer any evidence that the price of credit-card transactions was higher than the price one would expect to find in a competitive market.” Id. at *25. Beyond the fact that the government failed to provide reliable evidence of Amex’s transaction prices and profit margins compared to its competitors, the Court found that Amex’s increased merchant fees were indicative of the increased value of its services. Therefore, Amex was not charging a supra-competitive price. “That Amex allocates prices between merchants and cardholders differently from Visa and MasterCard is simply not evidence that it wields market power to achieve anticompetitive ends.” Id. at *26. The Court also found that evidence that Visa and MasterCard’s merchant fees have increased at locations where Amex is not accepted showed that the anti-steering provisions are not the cause of the increased fee. Instead, the Court held that the increase in merchant fees is due to “increased competition for cardholders and a corresponding market wide adjustment in the relative price charged to merchants.” Id.
The Court was unpersuaded that the government’s evidence that proceeds from Amex’s increased merchant fees were not entirely spent on cardholder rewards demonstrated that the anti-steering provisions gave Amex the power to charge anticompetitive prices. An increase in the number of credit card transactions during the relevant time period showed that output was not restricted, demonstrating a lack of anticompetitive effects. “Where . . . output is expanding at the same time prices are increasing, rising prices are equally consistent with growing product demand.” Id. at *27 (quoting Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 237 (1993)).
“The plaintiffs also failed to prove that Amex’s anti-steering provisions have stifled competition among credit-card companies.” Id. at *27. The Court pointed out that despite the presence of Amex’s anti-steering provisions, the credit-card industry has continued to flourish, increasing in both output and quality. Other leading credit-card providers have been incentivized to offer premium cards with better rewards. Merchant fee competition among credit card companies is robust, as evidenced by Amex’s history of at times lowering its fees to meet demand. Visa, MasterCard, and Discover’s increased market share resulting from their lower merchant fees was also indicative. In sum, contrary to the government’s position, the Court found that anti-steering provisions ultimately encourage competition.
The Court’s decision in Ohio v. American Express applied a broad understanding of competition in vertical markets. The decision established the required preliminary step of defining the market in vertical restraint cases. This holding heightens the burden for plaintiffs in Section 1 claims in the context of two-sided markets, requiring plaintiffs to prove unfair restraint on trade encompassing both sides of the market. In the short term, this decision may impact the policies of Visa and MasterCard, both of which eliminated anti-steering provisions in 2010 after settling similar antitrust litigation. More broadly, this Court’s decision may benefit growing technology companies that employ interdependent, multisided transactional markets.
Elizabeth C. Pritzker
Pritzker Levine LLP
On July 25, 2018, Judge Valerie Caproni issued opinions granting motions to dismiss brought by several “Non-Fixing Banks” in two multi-district cases alleging conspiracies to fix the prices of precious metals and futures. The two cases are: (1) In re Commodity Exchange Inc., Gold Futures and Options Trading Litigation, No. 14-MD-2548, 2018 WL 3585276 (SDNY July 25, 2018) (“Gold”), alleging a conspiracy to fix the price of physical gold and gold-denominated financial securities; and (2) In re London Silver Fixing, LTD., Antitrust Litigation, No. 14 MC-2573, 2018 WL 3585277 (SDNY, July 25, 2018) (“Silver), a benchmark-fixing case alleging a conspiracy to depress the price for physical silver and silver-denominated financial products, including silver futures. Although the complaints in both cases were supported by chat messages between investment traders of the “Fixing Banks” and the “Non-Fixing Banks,” demonstrating that chat participants in fact shared market positions and other proprietary information, Judge Caproni held that the evidence and other factual assertions in both cases failed to plausibly tie the “Non-Fixing Banks” either to the conspiracies or the price movements alleged in the respective complaints. In Silver, Judge Caproni additionally held that plaintiffs were not “efficient enforcers” as to the Non-Fixing Banks, and therefore lacked antitrust standing as required by Associated General Contractors v. California State Council of Carpenters, 459 U.S. 519 (“AGC”), and Second Circuit precedent applying AGC.
The Gold Ruling
Plaintiffs in these consolidated cases allege a conspiracy to fix the prices of physical gold and gold-denominated financial instruments (typically, gold futures or shares or options on gold exchange-traded funds) from 2004 to 2012. Gold, 2018 WL 3855276, at *1. During this period, the price of physical gold was wet daily through a private auction involving some of the largest banks in London. Plaintiffs allege that the afternoon “Gold Fixing” – also known as the “PM Fixing”—was a cover for a price-fixing conspiracy among the entity charged with operating the Gold Fixing, London Gold Market Fixing, Ltd, and the participant banks: The Bank of Nova Scotia, Barclays Bank plc, Deutsche Bank AG, HSBC Bank plc and Société Générale SA (collectively, the “Fixing Banks”). Id. Plaintiffs also named UBS AG and its affiliates which, while not itself a member of the Gold Fixing, alleging conspired with the Fixing Banks to suppress the price of gold as determined by the PM Fixing. Id.
Plaintiffs’ complaint alleges that the Gold Fix auction is conducted during a conference call among the Fixing Banks, with no third party participants, making it an almost perfect forum for collusion among competitors. Id. at *2. The market-clearing price in the auction (the “Fix Price”) is then published as a benchmark price for physical gold. This Fix Price, according to plaintiffs, effectively sets the market price for gold futures, options and forwards. Id. Plaintiffs point to trading data showing that the spot price of gold decreased during the PM Fixing on between 60 to 80 percent of the trading days during the class period even though, it is alleged, an efficient market would equally be likely to move upwards or downwards on a given day. Id. Plaintiffs’ analysis further shows that the downward movement in the price of physical gold consistently began in the minutes before the PM Fixing call began. Plaintiffs allege this points to coordinated trading based on foreknowledge of the Fix Price that would emerge from the auction. Id. Plaintiffs also allege patterns in Defendants’ price quotes which, Plaintiffs contend, link them to anomalous pricing behavior observed around the PM Fixing. Id.
In September 2016, Judge Caproni denied the Fixing Banks’ motion to dismiss, concluding that Plaintiffs plausibly alleged a conspiracy among the Fixing Banks to suppress the PM Fix Price. In re Commodity Exch., Inc., Gold Futures and Options Trading Litig., 213 F.Supp.3d 631, 642, 682 (SDNY 2016). The court granted UBS’s motion to dismiss, however, noting that UBS was not a member of the fixing panel during the class period, and Plaintiffs’ statistical analyses did not sufficiently connect UBS to the suppression of the PM Fixing. Plaintiffs’ barebones allegation that UBS quoted prices that were “lower than market averages,” the court found, were “inadequate to create a plausible inference of a conspiracy” as against UBS. Id at 663. In response to that finding, Plaintiffs filed an amended complaint that included, as additional support for their claims against UBS, 16 chat messages between a precious metals trade at UBS, and a precious metal trader at Deutsche Bank, one of the Fixing Banks. Gold, 2018 WL 3855276, at *3. “The chats describe brazen efforts to manipulate the gold markets through coordinated trading,” Judge Caproni writes in her July 2018 order. Id (setting forth the pertinent text of the chat messages). “Three of the chats reference Gold Fixing, but none references an agreement among UBS and the Fixing Banks to suppress gold prices.” Id. at *4. Plaintiffs’ amended complaint also contains additional statistical analyses allegedly showing that UBS, on average, quoted below-market prices beginning in the ten minutes before the PM Fixing and continued to quote below market prices until immediately before the London market closed, and that “UBS’s prices at the time of the PM Fixing fell in the bottom 5% and 10% for prices of the day far more often than they fell in the top 5% and 10%.” Id (quoting Plaintiff’s Third Amended Complaint).
UBS again moved to dismiss, arguing that the chat messages do not alter the court’s previous finding that Plaintiffs have not plausibly alleged UBS was involved in the alleged conspiracy. Id. at *4. Judge Caproni agreed. The chat messages, the court held, provide no direct evidence that UBS was involved in a scheme to suppress the PM Fix Price. Id. at *6. “Three messages reference ‘fixes,’ but none describe a scheme among UBS and the Fixing Banks persistently to suppress the Fix Price,” the court reasoned. Id. [O]ther references to the ‘fix’ describe proprietary trading that has no obvious connection to the fix-suppression conspiracy….” Id. Similarly, still “other chat messages describe coordinated trading and sharing of order flow information and proprietary pricing information, but they do not discuss manipulation of the PM Fixing.” Id. The messages also were late in the evening London time; therefore, the court held they “do not evidence sharing of information that would have been necessary to a fix-suppression scheme, or the sharing of that information close in time to the fixing itself. Id. at *8.
Judge Caproni also did not find Plaintiffs’ additional statistical analyses supportive of a plausible price-fixing conspiracy as against UBS. These analyses did not distinguish between the conduct of the individual defendants, the court found, but instead attempted to show that UBS and the Fixing Banks were collectively incentivized to suppress gold prices. Id. at *6. However, “[b]ecause the analysis does not specify the coefficient of variation [of gold positions] for any individual bank, the Court cannot determined whether UBS quoted ‘bunched’ prices or whether variance in UBS’s quotes is obscured by the group-wide analysis. Id. Plaintiffs’ attempts to bolster their allegations with UBS-specific statistical analyses – one that showed that UBS’s prices around the time of the PM Fixing were more frequently in the bottom 5% to 10% of the market that day than they were in the top 5% or 10% -- also was found to be insufficient. According to the court, “Plaintiffs do not specify why this fact supports their theory that UBS was involved in a fix-suppression conspiracy.” Id. at *7.
The court dismiss Plaintiffs’ claims against UBS holding that their Third Amended Complaint “fails to allege a plausible link between UBS and the price fixing scheme alleged against the Fixing Banks.” Id. at *9. In an unusual move, Judge Caproni opted to dismiss the claims against UBS with prejudice. “Plaintiffs are represented by competent, experienced counsel,” the court noted. Id. at *10. “If they had facts necessary to plug the obvious holes that exist in the TAC, the Court is confident those facts would have been included in the pleadings filed to date.” Id.
The Silver Ruling
In Silver, Judge Caproni also dismissed Plaintiffs’ antitrust claims against several Non-Fixing Banks which, as in Gold, were alleged to have participated in online chats with traders and representatives of alleged Fixing Banks.
According to Plaintiffs, the price of silver bullion during the class period was set in part through a daily auction among a group of small dealers (“the Silver Fixing”). Silver, 2017 WL 3585277, at *1. “Based on a sophisticated econometric analysis of thousands of price quotes from silver markets, Plaintiffs alleged that this daily private auction was a cover for a conspiracy among the participating banks, Deutsche Bank, HSBC, and the Bank of Nova Scotia (together, the ‘Fixing Banks.’), to suppress the price for physical silver.” Id.
In September 2016, the court denied motions to dismiss by the Fixing Banks, finding that Plaintiffs stated plausible antitrust claims against these entities. Id. at *1. Plaintiffs later settled with Deutsche Bank for $38 million and a cooperation agreement, pursuant to which Deutsche Bank produced to Plaintiffs a trove of preserved electronic chat messages among precious metal traders employed by Deutsche Bank and traders at Bank of America, Barclays, Standard Chartered, BNP Paribas and UBS (the “Non-Fixing Banks”). Id.
Plaintiffs used the text messages received from the Deutsche Bank cooperation agreement to amend their complaint and to allege that the Non-Fixing Banks conspired with the Fixing Banks and among themselves to manipulate the Silver Fixing, and the silver markets more generally. Id. at *1. Separately, the Commodity Futures Trading Commission (“CFTC”) initiated enforcement actions against UBS and Deutsche Bank, alleging that traders at these two Non-Fixing Banks “spoofed” the markets for precious metals and collaborated with traders at another financial institution to trigger stop-loss orders. Id. at *4. UBS and Deutsche Bank later settled their claims with the CFTC for $15 million and $30 million, respectively. Id. The U.S. Department of Justice also has charged two Bank of America / Merrill Lynch traders with commodities fraud (among other things) in connection with the alleged spoofing in the precious metals futures markets, including the Commodities Exchange or COMEX. Id.
Against this backdrop, the Non-Fixing Banks moved to dismiss on the grounds that the chat messages Plaintiffs obtained from Deutsche Bank did not connect them to any alleged conspiracy with the Fixing Banks and did not document any actionable manipulations of the silver markets (among other things). Id. at 1. Judge Caproni’s July 25 decision dismissed Plaintiffs’ claims against these entities, finding Plaintiffs’ allegations fail for three reasons.
First, the court found that Plaintiffs’ allegations of a “comprehensive” conspiracy to fix the price of physical silver and silver-based financial instruments did not meet the plausibility standard of Bell Atl. v. Twombly, 550 U.S. 544, 570 (2007). The court declined to infer the existence of a single, overarching conspiracy, reasoning that Plaintiffs’ allegations instead suggest merely “unrelated, internally inconsistent efforts to manipulate the sliver markets episodically.” Id. at *6. “Even though the TAC plausibly alleges that the Fixing Banks conspired to depress the Fix Price,” the court reasoned, “it does not explain why the Non-Fixing Banks, which are competitors and counterparties, would be in agreement.” Id. “The coordinated trading alleged in the TAC lacks a connection to suppression of the Fix Prices and, in fact, could have made it more difficult to profit from foreknowledge of the Fix Price,” the court held. Id.
Second, as in Gold, the court did not find the chat messages produced by Deutsche Bank provided a basis to link the Non-Fixing Banks to the alleged conspiracy. These messages took several forms. “Numerous chats between a trader at UBS and a trader at Deutsche Bank describe efforts to coordinate describe efforts to coordinate positions” (id. at *3), with one such chat also including traders at HSBC and Barclays. Id. Traders as Barclays also shared information with Deutsche Bank regarding “bid-ask spreads” and attempts to “spoof” the silver markets. Id.Id. Chats between Deutsche Bank and UBS referenced “collusion with traders at Barclays.” Id. The Deutsche Bank cooperation materials included “[s]everal chats describe[ing] real-time sharing of market positions and conditions, including bid-ask spreads quoted by BNP Paribas and Deutsche Bank’s positon heading into the Silver Fixing.” Id. “Two of the chats between Deutsche Bank and BNP Paribas reference collusive trading techniques” – including taking the “bulldozer” out on a prior occasion, a possible reference to triggering stop-loss orders. Id. Three chat messages reveal a trader at Standard Chartered (and formerly of HSBC) sharing current trading positions as well as Deutsche Bank’s position in the Silver Fixing. Id. “Finally, the TAC alleges conversations between Deutsche Bank and [Bank of America / Merrill Lynch]” that include “exchanges of information regarding bid-spreads”, “the price-level of stop-loss orders in the market,” and “their current positions in silver denominated derivatives.” Id.
Lacking econometric or statistical analyses linking these chat messages to specific trading times or trading behavior, however, the court held that Plaintiffs had not met their burden to plausibly plead a single, overarching conspiracy among the Fixing and Non-Fixing Banks, taken together. “The Court does not find Plaintiffs’ allegations of a single conspiracy among the Fixing Banks and Non-Fixing Banks to manipulate the Silver Fixing to be plausible.” Id. at *12. Instead, the court held, “the TAC plausibly alleges two conspiracies. Plaintiffs have plausibly alleged a conspiracy involving the Fixing Banks to suppress the Fix Price through the dialing fixing call. Plaintiffs have also plausibly alleged a conspiracy among the Non-Fixing banks to collude in the silver markets through market manipulation and information-sharing.” Id (emphasis added).
Having determined that Plaintiffs stated a separate, plausible conspiracy against the Non-Fixing Banks, Judge Caproni then faced the question of whether Plaintiffs had antitrust standing to assert such a claim under AGC. “The Second Circuit has identified four factors to consider whether a particular plaintiff has standing as an ‘efficient enforcer’ to seek damages under the antitrust laws: (1) whether the violation was a direct or remote cause of the injury; (2) whether there is an identifiable class of other persons whose self-interest would normally lead them to sue for the violation; (3) whether the injury was speculative; and (4) whether there is a risk that other plaintiffs would be entitled to recover duplicative damages or that damages would be difficult to apportion among possible victims of the antitrust injury…. Built into the analysis is an assessment of the ‘chain of causation’ between the violation and the injury.” Id. at * 12 (citing Gelboim v. Bank of Am. Corp., 823 F.3d 759, 772 (2d Cir. 2016) (citations omitted)).
The court found that Plaintiffs lacked antitrust standing as “efficient enforcers” against the Non-Fixing Banks for several reasons. The first reason has to do with the remoteness of the alleged injury. Id. at *12. Because Plaintiffs did not deal directly with any of the Non-Fixing Banks in their market purchases, and because Plaintiffs’ claims against these Non-Fixing Banks do not precisely depend on benchmark manipulation but, rather, “a comprehensive scheme of market manipulation, involving rigged bid-ask spreads and coordinated trading in unspecified silver markets,” the court held that Plaintiffs failed to plead facts or provide an econometric analysis that to plausibly show how the Non-Fixing Banks’ alleged coordinated trading impacted the markets for physical silver or silver-denominated instruments. Id. at *13. For similar reasons, the court also found “Plaintiffs’ injuries and the Non-Fixing Banks’ conduct is attenuated and inadequately alleged in the TAC.” Id.
The court found the other elements required for antitrust standing under AGC were lacking as well. “Class members who traded directly with the Non-Fixing Banks were more directly injured than Plaintiffs,” Judge Caproni held. Id. at *15. The duplicative recovery and apportionment of damages factor, in the court’s view, “also weights against Plaintiffs’ claims against the Non-Fixing Banks.” Id. at *16. Specifically, the court held that because “the CFTC and the Department of Justice have instituted enforcement actions and criminal cases against several of the defendants and their traders for the manipulative trading alleged in the TAC” there is a lessened “need for plaintiffs to function as private attorneys general and vindicators of the public interest.”
As in Gold, the court granted the Non-Fixing Banks’ motion to dismiss with prejudice. Id. at *28.
Judge Caproni’s motion to dismiss opinions in Gold and Silver demonstrate the difficulty of pleading antitrust and commodities fraud claims as against entities alleged to have engaged in market manipulation or other anticompetitive conduct but with whom no plaintiff has a contractual or direct purchase relationship. While the court appears to have required more specificity than other courts have to satisfy Twombly, the larger problem for plaintiffs asserting benchmarking or similar market-based claims, in the Second Circuit at least, may be pleading antitrust standing sufficiently to satisfy Circuit precedent interpreting AGC.
Harrison (Buzz) Frahn, Michael R. Morey, Lauren Ditty (Summer Associate)
Simpson Thacher & Bartlett LLP
On May 11, 2018, the Ninth Circuit reversed the U.S. District Court for the Western District of Washington’s dismissal of the Chamber of Commerce of the United States of America’s (the “Chamber’s”) federal antitrust claims against the City of Seattle in Chamber of Commerce of the United States of Am. v. City of Seattle, 890 F.3d 769 (9th Cir. 2018). The suit arose from a recently enacted Seattle ordinance that allows independent-contractor drivers to collectively bargain with rideshare app companies like Uber and Lyft for the rideshare fees that the companies impose on the drivers. The Chamber alleged, inter alia, that Seattle’s ordinance was preempted by Section 1 of the Sherman Act because the ordinance sanctioned price-fixing of the rideshare fees by private cartels of independent-contractor drivers. Seattle moved to dismiss on the basis that the state-action immunity doctrine protected the ordinance from preemption by Section 1 of the Sherman Act. The District Court agreed with Seattle, and the Chamber appealed.
The Ninth Circuit held that the state-action immunity doctrine did not protect the ordinance from preemption by Section 1 of the Sherman Act because: (1) the State of Washington had not “clearly articulated and affirmatively expressed” a state policy authorizing price-fixing of rideshare fees, and (2) the State of Washington did not “actively supervise” the anticompetitive policy imposed by the ordinance. Given that Seattle’s ordinance was the first municipal ordinance of its kind in the United States, the Ninth Circuit’s decision has the potential to impact future state and municipal attempts to unionize independent-contractor drivers in the rideshare industry. Seattle sought rehearing or en banc review on June 25, 2018.
Rideshare companies like Uber and Lyft provide smartphone apps that allow riders to request rides from independent-contractor drivers. Chamber of Commerce, 890 F.3d at 776. To receive ride requests through the app, drivers enter into non-exclusive contracts with various rideshare companies and pay licensing fees to the companies for the ride-referral services that the companies’ apps provide. Id. These licensing fees are a percentage of riders’ paid fares––after a driver completes a ride, the rideshare company that referred the ride to the driver takes a cut of the total ride fare and remits the remainder to the driver. Id.
On December 14, 2015, Seattle passed Ordinance 124968 (the “Ordinance”), which authorized collective bargaining between “driver coordinators”—like Uber and Lyft—and their independent-contractor drivers for the “nature and amount of payments to be made by, or withheld from, the driver coordinator to or by the drivers.” Id. at 778 (quoting Seattle, Wash., Municipal Code § 6.310.735(H)(1)). Under the Ordinance, a city-approved “qualified driver representative” can represent “qualifying drivers”—independent-contractor drivers who have driven at least 52 trips originating or ending within Seattle’s city limits—in negotiations with “driver coordinators” like Uber and Lyft. Chamber of Commerce, 890 F.3d at 777. Once the qualified driver representative reaches an agreement with a driver coordinator on the price of the service fees that the qualifying drivers must pay the driver coordinator, they submit the agreement to the director of Seattle’s Department of Finance and Administrative Services for approval. Id. at 778. The director then considers whether “the substance of the agreement promotes the provision of safe, reliable, and economical for-hire transportation services and otherwise advances the public policy goals set forth in [the Seattle Municipal Code] and in the Ordinance.” Id. The agreements are not enforceable unless the director approves them. Id. The Ordinance was the first municipal ordinance of its kind in the United States, and went into effect on January 22, 2016. Id.
After filing an initial complaint on March 3, 2016 that was dismissed as unripe because collective bargaining under the Ordinance had not yet begun, the Chamber filed an amended complaint on March 9, 2017 against the City of Seattle and the Seattle Department of Finance and Administrative Services (collectively, “Seattle”) for enacting and enforcing the Ordinance. The amended complaint asserted, inter alia, two federal antitrust claims for injunctive and declaratory relief alleging that the Ordinance “sanctioned price-fixing of ride-referral service fees by private cartels of independent-contractor drivers.” Id. at 775. The first claim alleged that Seattle violated Section 1 of the Sherman Act by enacting and enforcing the ordinance. The second claim alleged that the Ordinance conflicted with and was preempted by Section 1 of the Sherman Act. Seattle moved to dismiss the Chamber’s amended complaint and asserted, inter alia, that both of the Chamber’s federal antitrust claims should be dismissed because the doctrine of state-action immunity exempted the Ordinance from preemption by the Sherman Act.
District Court Opinion
In analyzing Seattle’s state-action immunity defense to the Chamber’s federal antitrust claims, the District Court began with the assumption that the Ordinance constituted a per se antitrust violation because it sanctioned “collusion between independent economic actors”––here, the independent-contractor drivers––“to set the prices they will accept for their services in the market”––here, the ridesharing fees imposed by the driver coordinators. Chamber of Commerce of United States v. City of Seattle, 274 F. Supp. 3d 1155, 1162 (W.D. Wash. 2017). After making this assumption, the District Court applied the Midcal state-action immunity test to determine whether the Ordinance was protected from preemption by Section 1 of the Sherman Act. Id. at 1163; see also Cal. Retail Liquor Dealers Ass’n v. Midcal Aluminum, Inc., 445 U.S. 97, 105 (1980). Under the Midcal test, anticompetitive acts by private parties are entitled to state-action immunity if: (1) the anticompetitive act is “clearly articulated and affirmatively expressed as state policy” and (2) the state “actively supervises” the policy. Midcal, 445 U.S. at 105. The District Court concluded that Seattle satisfied both parts of the Midcal test and held that Section 1 the Sherman Act did not preempt the Ordinance. Chamber of Commerce, 274 F. Supp. 3d at 1169.
First, the District Court reasoned that the State of Washington had “clearly articulated” a state policy authorizing anticompetitive regulation in the rideshare industry. Id. at 1163. In reaching this conclusion, the District Court relied on two Washington statutes that permitted local regulation of for-hire vehicles and taxicabs “without liability under federal antitrust laws.” Id. Although the statutes did not explicitly authorize collective bargaining between independent-contractor drivers and rideshare companies, the District Court concluded that the statutes authorized a “wide array of municipal regulation” in the for-hire transportation industry, including local regulation of license requirements, fees, routes, and “any other requirements adopted to ensure safe and reliable for-hire vehicle transportation or taxicab service.” Id. at 1163-65. To support its ruling, the District Court relied on the Supreme Court’s decision in S. Motor Carriers Rate Conference, Inc. v. United States, 471 U.S. 48 (1985), which held that the first prong of the Midcal test is satisfied when the State clearly intends to displace competition in a particular field with a regulatory structure. Id. at 1164-65. The District Court concluded that Seattle’s inability to point to a “specific, detailed legislative authorization” of collective negotiation of rideshare fees did not foreclose its state-action immunity defense. Id. Instead, the District Court ruled that the Washington legislature “clearly contemplated and authorized regulations with anticompetitive effects in the for-hire sphere.” Id. Therefore, the District Court concluded that Seattle satisfied the first prong of the Midcal test even though the statutes that Seattle relied on did not specifically authorize collective bargaining between rideshare companies and independent-contractor drivers. Id.
Second, the District Court determined that the ordinance satisfied the active-supervision prong of the Midcal state-action immunity test. Id. at 1169. Critically, the District Court drew no distinction between state oversight and municipal oversight in determining whether Seattle satisfied the supervisory component of the Midcal test. Id. at 1167. Relying again on S. Motor Carriers, the District Court reasoned that “[w]hen a municipal regulation allows private parties to engage in anticompetitive conduct . . . the municipality (not the state) must actively supervise the conduct.” Id. (citing S. Motor Carriers, 471 U.S. at 62). According to the District Court, Seattle satisfied the active-supervision prong of the Midcal test because the director of the Seattle Department of Finance and Administrative Services actively supervised the conduct of the independent-contractor drivers and rideshare companies under the Ordinance. Id. at 1168. The District Court saw no reason to require active state supervision of the Ordinance where there was adequate municipal supervision. Id. (emphasis added) (citing Town of Hallie v. Eau Claire, 471 U.S. 34, 46 (1985)). The District Court further stated that Seattle’s supervision of the Ordinance was not “merely a gauzy cloak of state involvement over what is essentially a private price-fixing arrangement” because the director of the Seattle Department of Finance and Administrative Services evaluated the collective-bargaining agreements between the drivers and rideshare companies before they went into effect, and the director was not permitted to “passively accept the agreement[s] proposed by the parties.” Id. at 1168-69. Because Seattle satisfied both prongs of the Midcal state-action immunity test, the District Court dismissed the Chamber’s federal antitrust claims. Id. at 1169. The Chamber appealed.
Ninth Circuit Opinion
The Ninth Circuit reversed the District Court’s dismissal of the Chamber’s federal antitrust claims. Like the District Court, the Ninth Circuit applied the Midcal state-action immunity test to determine whether Section 1 of the Sherman Act preempted the Ordinance. But the Ninth Circuit ultimately found both prongs lacking.
The Clear-Articulation Test
The Ninth Circuit first determined that the State of Washington had not “clearly articulated and affirmatively expressed” a state policy authorizing private parties to price-fix the fees that independent-contractor drivers pay to companies like Uber or Lyft in exchange for using the companies’ ride-referral services. Chamber of Commerce, 890 F.3d at 782. In teeing up its application of the clear-articulation test, the Ninth Circuit explained that the “inquiry …is a precise one. ‘The relevant question is whether the regulatory structure which has been adopted by the state has specifically authorized the conduct alleged to violate the Sherman Act.” Id. (emphasis in original) (quoting Cost Mgmt. Servs., Inc. v. Wash. Nat. Gas Co., 99 F.3d 937, 942 (9th Cir. 1996)). The Ninth Circuit also clarified that “[t]he relevant statutory provisions must plainly show that the state legislature contemplated the sort of activity that is challenged, which occurs where they confer express authority to take action that foreseeably will result in anticompetitive effects.” Chamber of Commerce, 890 F.3d at 782 (emphasis in original) (internal quotation marks omitted) (citing Hass v. Or. State Bar, 883 F.2d 1453, 1457 (9th Cir. 1989)).
In applying the clear-articulation test, the Ninth Circuit found that the statutes that Seattle relied on to support its state-action immunity defense––Wash. Rev. Code. §§ 46.72.001, 46.72.160, 81.72.200, and 81.72.210––did not “plainly show” that the State of Washington “contemplated” allowing independent-contractor drivers to price-fix their compensation, and that such an anticompetitive result was not foreseeable. Chamber of Commerce, 890 F.3d at 783–87.
The first statutes that the Ninth Circuit grouped together and analyzed (Wash. Rev. Code. §§ 46.72.001 and 81.72.200) provided, in relevant part, that “it is the intent of the legislature to permit political subdivisions of the state to regulate for hire transportation services without liability under federal antitrust laws.” Id. at 783. The Ninth Circuit held that these statutes were insufficient to invoke state-action immunity in light of Supreme Court precedent holding that “a State may not confer antitrust immunity on private persons by fiat,” id. (quoting F.T.C. v. Ticor Title Ins. Co., 504 U.S. 621, 633 (1992)), nor “validate a municipality’s anticompetitive conduct simply by declaring it to be lawful,” id. (quoting Town of Hallie, 471 U.S. at 39). Moreover, the Ninth Circuit reasoned that the plain language of these statutes merely focused on providing “privately operated for hire transportation services”––not regulating the fees that independent-contractor drivers pay driver coordinators for using their ride-referral app services. Id. In other words, the Ninth Circuit found that “[n]othing in the statute[s] evinces a clearly articulated state policy to displace competition in the market for ride-referral service fees charged by companies like Uber, Lyft, and Eastside.” Id. at 784.
The next two statutes (Wash. Rev. Code. §§ 46.72.160 and 81.72.210) fared no better. The Ninth Circuit found that these statutes focused on the regulation of for-hire vehicle services and provided in relevant part that “[c]ities … may license, control, and regulate all for hire vehicles operating within their respective jurisdictions.” Id. at 784 (emphasis added) (quoting Wash. Rev. Code. §§ 46.72.160). The Ninth Circuit concluded that the regulatory powers granted to Washington cities by these statues, such as “[r]egulating entry into the business of providing for hire vehicle transportation services” and “[r]egulating the routes and operations of for hire vehicles,” referred specifically to for-hire vehicles––not the fees that Uber and Lyft charge independent-contractor drivers for using their ride-referral app services. Id. at 784-85. Accordingly, the Ninth Circuit found these statutes to be inapposite to the specific anticompetitive practices sanctioned by the Ordinance. See id. at 785 (analogizing to Medic Air Corp. v. Air Ambulance Authority, 843 F.2d 1187, 1189-90 (9th Cir. 1988), which distinguished between the market for ambulance services and the market for dispatching ambulances in applying the clear-articulation test to find that state-action immunity did not shield the ambulance service market from preemption by Section 1 of the Sherman Act).
The Ninth Circuit rounded out its analysis on the clear-articulation prong of the Midcal test by noting that “Uber and Lyft did not exist when the Washington statutes were enacted,” and that the “very concept of digital ridesharing services was probably well beyond the imaginations of lawmakers two to three decades ago.” Chamber of Commerce, 890 F.3d at 787 (noting that Wash. Rev. Code. §§ 46.72.001 and 46.72.160 were enacted in 1996 and §§ 81.72.200 and 81.72.210 1984 were enacted in 1984).
For these reasons, the Ninth Circuit held that the Washington legislature had not “clearly articulated and affirmatively expressed” a state policy authorizing the anticompetitive practices sanctioned by the Ordinance. The Ninth Circuit then proceeded to analyze the second prong of the Midcal state-action immunity test: whether the state actively supervises the anticompetitive policy.
The Active-Supervision Requirement
The Ninth Circuit concluded that Seattle failed to satisfy the active-supervision prong of the Midcal test. The Ninth Circuit first explained that when a “state or municipal regulation [of] a private party is involved, …active state supervision must be shown, even where a clearly articulated state policy exists.” Id. at 788 (quoting Town of Hallie, 471 U.S. at 46). Moreover, the Ninth Circuit clarified that “[t]he Supreme Court has stated repeatedly that active supervision must be ‘by the State itself.’” Id. at 789 (emphasis added) (citing Midcal, 445 U.S. at 105, and N. Carolina State Bd. of Dental Examiners v. F.T.C., 135 S. Ct. 1101, 1112 (2015)).
In contrast, Seattle argued that it satisfied the active-supervision prong because, under the Ordinance, the director of the Seattle Department of Finance and Administrative Services evaluated the collective-bargaining agreements between the drivers and rideshare companies before they went into effect, and the enforceability of the agreements was subject to the director’s approval. Seattle further asserted that the Supreme Court used the term “State” in the opinions relied on by the Ninth Circuit “as shorthand for the State and all its agents, including municipalities,” and therefore the city director’s supervision of the Ordinance was sufficient. Chamber of Commerce, 890 F.3d at 789.
The Ninth Circuit disagreed and stated that “[t]he City cites no controlling authority to support its argument.” Id. In rejecting Seattle’s arguments, the Ninth Circuit focused on “the specific distinction the Supreme Court has drawn between cities, which are not sovereign entities, and states, which are,” and held that State––not municipal––supervision was required for Seattle to satisfy this prong of the Midcal test. Id. Accordingly, Seattle’s attempt to rely on municipal oversight of the anticompetitive acts sanctioned by the Ordinance was insufficient and incorrect as a matter of law. Chamber of Commerce, 890 F.3d at 789. And because “[i]t [was] undisputed that the State of Washington plays no role in supervising or enforcing the terms of the . . . Ordinance,” the Court concluded that Seattle failed to satisfy the active-supervision prong of the Midcal test. Id.
Having found that Seattle failed to satisfy the two requirements under the Midcal state-action immunity test, the Ninth Circuit remanded the case to the District Court for further proceedings.
The Ninth Circuit’s decision in Chamber of Commerce has the potential to impact future state and municipal attempts to unionize independent-contractor drivers in the rideshare industry, especially given that Seattle’s collective bargaining ordinance for independent-contractor drivers was the first municipal ordinance of its kind in the United States. Seattle filed a Petition for Rehearing or Rehearing En Banc on June 25, 2018, and the Chamber of Commerce’s response to the Petition is due in early August.