On June 30, 2015, the same day as the launch of Apple’s new streaming music service, the Second Circuit Court of Appeals coincidentally affirmed a district court ruling that Apple conspired with five of the country’s largest book publishers to fix prices for ebooks and coerce Amazon to change its pricing model to accommodate those higher, fixed prices. United States v. Apple, Inc., No. 13-3741, Slip Op., (2nd Cir. June 30, 2015). The Second Circuit held that while Apple’s agreements with the publishers were vertical, they were a component of a horizontal conspiracy among the publishers that was a per se violation of § 1 of the Sherman Act.
Factual Background 1
A few short months before its anticipated January 27, 2010 iPad launch, Apple commenced negotiations with the six largest publishers in the U.S. In doing so, Apple recognized the publishers’ collective loathing2 for Amazon’s $9.99 price point for new releases and New York Times bestsellers – a practice that was not only a loss-leader for Amazon to encourage adoption of its electronic reader, the Kindle, but it was also cannibalizing the publishers’ hardcover sales, driving down hardcover prices, and cutting into their profitability. Slip Op. at 9.3 Throughout the negotiations and subsequent iBookstore launch, Apple took advantage of the publishers’ desire to move off that price point toward more economically favorable conditions.
In response, Apple developed three novel contract terms: move ebooks to an agency model, develop price caps for equivalent categories of ebooks and printed books, and require a most-favored nations clause. Each of these clauses, according to the Court, were innocuous on their own by effectively anticompetitive collectively.
First, Apple noted that the wholesale model, in which publishers’ sold ebooks to retailers at a wholesale price and the retailers made profit on a small markup, was unfavorable and sought to switch to an agency model, in which the publisher sets the price and Apple receives a fixed commission of 30 % for each sale. Slip Op. at 21-25. The Court found that this ensured each sale net Apple a profit, but created the added risk that Apple could open the iBookstore with higher prices than Amazon and be “brimming with titles, but devoid of customers.” Slip Op. at 24. To combat this risk, Apple knew it needed to cap prices for selected categories of titles and it needed to ensure the publishers’ had sufficient incentive to switch all other ebook retailers, particularly Amazon, to an agency model. Slip Op. at 24-25.
Second, Apple negotiated for price caps that would set maximum price limits for various categories of titles based on the equivalent title’s hardcover list price. Under the new caps, New York Times bestsellers were capped at $14.99 if the hardcover price was over $30 and $12.99 if under $30. For new releases, there were caps of (1) $12.99 for hardcovers between $25 and $27.50, (2) $14.99 for hardcovers between $27.50 and $30, and (3) $16.99 to $19.99 for hardcovers over $30. The caps incentivized publishers to raise their hardcover prices to shift books between the cap brackets. Prices on new releases rose 24.2% and bestsellers jumped 40.4% in the year following the iBookstore launch.
Third, Apple realized that its successful entry into ebooks hinged on its ability to competitively price with Amazon at the higher prices the publishers wanted. To do so, Apple needed a way to induce the publishers to collectively compel Amazon to sell its ebooks on the agency model. The most-favored nation clause (“MFN”) was precisely the elegant solution Apple needed. The MFN required the publishers “to offer any ebook in Apple’s iBookstore for no more than what the same ebook was offered elsewhere, such as from Amazon.” Slip. Op. at 26. With the 30 % Apple commission, the MFN actually resulted in a lower per-book profit for the publishers in the iBookstore, meaning that profitability hinged on the publishers’ ability to regain pricing control with Amazon via the agency model and set higher prices on new releases and bestsellers.
Critically, Apple’s efforts to reach these agreements also included constant communications with each of the publishers to encourage tougher negotiations with Amazon that would require agency model adoption. The Court listed several instances in the weeks leading up to and the weeks following the iBookstore announcement on January 27, 2010 where Apple met with or communicated with publishers urging them to move retailers to the agency model in an effort to meet the price caps and increase ebook prices. Slip Op. at 30-40. Moreover, the publisher CEOs were in constant contact during their own, apparently separate negotiations with Apple – including a three-day span exchanging 34 phone calls to discuss their individual progress. Slip Op. at 35.
Court’s Analysis: To Agree or Not To Agree? Whether 'Tis Horizontal or Vertical
The Court addressed two threshold questions critical to its analysis: (1) whether an agreement existed between Apple and the publishers to fix retail prices of ebooks, and (2) whether the nature of the restraint was horizontal, warranting per se treatment, or vertical, requiring full-blown rule of reason analysis. Ultimately, the Court found Apple had masterminded, in practical effect, a price-fixing conspiracy among and with the publishers and that the price-fixing restraint was thusly horizontal and so to be adjudicated under the stricter per se rule of liability.
Section 1 of the Sherman Act requires proof that an agreement exists that restrains trade. Here, the preliminary question was whether an agreement existed between Apple and the publishers to fix the retail prices of ebooks. The Court, relying heavily on its factual discussion and the findings of the district court, held that the proper “agreement” to consider was not the individual vertical agreements between Apple and the publishers, but the unwritten agreement “to raise consumer-facing ebook prices by eliminating retail price competition.” Slip Op. at 66. Were the vertical agreements the end of Apple’s pursuits, the Court reasoned, it may have been sufficient to show Apple had not agreed to the improper activity. But, because “Apple’s involvement in the conspiracy continued even past the signing of its agency agreements,” it signaled that Apple had “consciously played a key role in organizing their express collusion.” Slip Op. at 64-65.
Given its finding that an agreement existed, the Court was left to wrestle with whether that agreement could properly be categorized as horizontal or vertical. Settled antitrust law dictates that horizontal price-fixing arrangements, as one of the supreme evils of antitrust, remain per se illegal. Nearly a decade ago, the Supreme Court reclassified vertical price-fixing arrangements as potentially procompetive and therefore requiring analysis under the rule of reason. Nevertheless, the Court found it to be well-established that vertical agreements can often be useful evidence of the existence of a horizontal cartel. Slip Op. at 66. And here was no exception – the vertical agreements created the framework that enabled the publishers to collectively challenge Amazon and fix higher retail prices for new releases and bestsellers. As a result, per se treatment must apply.4
Apple raised two arguments disputing the application of the per se rule in this case: (1) independent vertical agreements with publishers cannot show that Apple consciously organized a conspiracy, even if the effect of the agreements was to raise prices, and (2) even if Apple did orchestrate a horizontal price-fixing conspiracy, its conduct was not subject to per se condemnation. The Court quickly and decisively dismissed Apple’s arguments against the application of the per se rule.
First, Apple’s conduct both incentivizing the collective action and pursuing it after executing the vertical agreements demonstrated to the Court that Apple’s vertical agreements were not independent. Parallel conduct on its own, as the Court admits, is not sufficient for antitrust liability, but instead requires plus-factors proving that the conduct was connected to such an extent as to infer an underlying agreement. Here, the Court found that Apple was not just capitalizing on its knowledge of the publishers’ incentives, but that Apple knew that the offered vertical contracts were only attractive to the publishers if all the publishers acted collectively to pressure Amazon into the agency model. Slip Op. at 59-62. The Court held that Apple knew that each publisher signing Apple’s agreement was signaling to the others that it was committed to forcing Amazon to the agency model. “In fact, this was the very purpose of the MFN.” Slip Op. at 60.
Second, the fact that Apple’s written agreements with the publishers were vertical was not dispositive because the focus of the analysis is on the anticompetitive effects of the restraint, which in this case was the horizontal agreement Apple organized among the publishers to raise and fix ebook prices. Slip Op. at 75. On this point, the Court read Leegin Creative Leather Products, Inc. v. PSKS, Inc., 551 U.S. 877 (2007), in support, stating that Leegin’s holding that enforcing minimum resale prices through vertical agreements was subject to the rule of reason turned on the lawfulness of the vertical agreements and not on allegations that the defendant in Leegin was party to an unlawful horizontal cartel. Essentially, the Court finds the reliance on Leegin by Apple and the dissent for the proposition that vertical agreements are subject to rule of reason analysis to be misplaced because it addresses the written vertical agreements, not the horizontal cartel that those agreements helped to create. Slip Op. at 75.5
The Court stated that the Sherman Act “requires evaluating the nature of the restraint, rather than the identity of each party who joins in to impose it, in determining whether the per se rule is properly invoked.” Slip Op. at 8. Importantly, the focus, then, is on the restraint, not necessarily how that restraint came to be. A horizontal price-fixing conspiracy remains precisely that, regardless of whether it was created by an express agreement or an unwritten agreement through a series of vertical agreements. The Court dusted off and cited Klor’s, Inc. v. Broadway-Hale Stores, Inc., 359 U.S. 207 (1959), for this proposition – that a retailer is per se liable for facilitating a distributors’ horizontal group boycott because group boycotts are per se illegal, not because the vertical arrangements between the retailer and distributors created and organized the horizontal cartel. This reasoning was critical for the court: “the question is whether the vertical organizer of a horizontal conspiracy designed to raise prices has agreed to a restraint that is any less anticompetitive than its co-conspirators, and can therefore escape per se liability. We think not.” Slip Op. at 79-80. Once the true restraint was identified and categorized, per se liability quickly followed.
Apple also pursued the oft-made, rarely-embraced “enterprise and productivity” argument. That is, Apple argued that the vertical agreements and the horizontal conspiracy were essential to create a new product and to ensure that this new product was available.6 Here, the Court stated, there was no resulting joint venture or otherwise similarly productive relationship between the conspirators. Absent procompetitive benefits to a new joint venture, the argument was swiftly erased.
In the end, the Court held that “the Publisher Defendants took by collusion what they could not win by competition.” Slip Op. at 86. To the majority: “In short, Apple and the dissent err first in equating a symptom (a single-retailer market) with a disease (a lack of competition), and then err again by prescribing the disease itself as the cure.” Slip Op. at 99.
Key Takeaways: When Does Per Se Analysis Actually Become The Rule of Reason?
The parties involved, particularly Apple, and the facts guarantee the case a lot of attention. For future application of the antitrust law, the Court’s opinion yields a few important takeaways: (1) per se analysis hinges, to some extent, on the characterization of an agreement as horizontal or vertical – a characterization that time and complexity have blurred, (2) MFNs are not, on their own, unlawful, but often contribute toward economic incentives that create antitrust liability, and (3) the per se analysis is increasingly blended with traditional, in-depth rule of reason analysis.
Greater market complexity has blurred the line between horizontal and vertical agreements, but the antitrust implications of each remain largely the same: horizontal agreements are treated with much harsher scrutiny than their vertical counterparts. The same is true here. Critically, too, a written vertical agreement is not sufficient to divert the antitrust court from finding the unwritten horizontal price-fixing conspiracy those vertical agreements created. And the antitrust laws are not so limited as to only apply to written agreements. Antitrust defendants should focus attention on couching their defense as one of vertical agreements with significant procompetitive benefits subject to the rule of reason and challenge any characterization that the alleged agreements were horizontal.
As Courts have recently and repeatedly reiterated, MFN clauses, on their own, are not necessarily violative of the antitrust laws; however, the effect or the incentives they create could induce antitrust liability. Nonetheless, the antitrust agencies have focused a lot of attention on putting the antitrust validity of MFNs in certain circumstances into question.7 And here, they obtained some important judicial support for their views—the Court held that the addition of the MFN had the intended purpose of requiring or encouraging the publishers’ collective action to change their sales model with Amazon. Firms looking to implement MFNs in their sales, supply, or other contracts should seek an antitrust evaluation of the potential anticompetitive effects.
Interestingly, for all its insistence on the per se rule as the appropriate benchmark for Apple’s conduct, the Court conducted a remarkably thorough and intricate factual and economic analysis en route to reaching its conclusion that the vertical agreements that created a horizontal conspiracy was a per se violation of Section 1.8 This type of analysis is the hallmark of a rule of reason analysis and is typically given short shrift in price fixing cases, but the Court (and the district court before it) spilled plenty of ink to detail the intricate incentive scheme that Apple’s price caps, most-favored nations clauses, and agency model created. Though the Court ultimately ignored Apple’s procompetitive justifications in favor of the per se rule, it signals that a routine application of the per se rule is becoming much less straight-forward as economics and the law evolve.
1 Mintz Levin professionals have addressed these facts in advisories drafted and published discussing the original Department of Justice lawsuit on April 11, 2012 and the district court opinion on July 10, 2013. Bruce D. Sokler, Robert G. Kidwell, and Farrah Short, “The Next Chapter: DOJ Sues Apple and Publishing Companies for Alleged e-Book Conspiracy,” Apr. 12, 2012, available at http://www.mintz.com/newsletter/2012/Advisories/1815-0412-NAT-AFR/; Bruce Sokler, Charles Samuels, Robert Kidwell, Christi Braun, Farrah Short, Helen Kim, and Shoshana Speiser, “A Book on Books: SDNY Issues 160-Page Opinion Declaring that Apple Violated Section 1 of the Sherman Act by Conspiring to Raise E-Book Prices,” July 15, 2013, available at http://www.mintz.com/newsletter/2013/Advisories/3220-0713-NAT-AFR/index.html.
2 The Court noted the district court’s finding that “roughly once a quarter, the CEOs of the [top six publishers] held dinners in the private dining rooms of New York restaurants, without counsel or assistants present, in order to discuss the common challenges they faced.” Slip Op. at 15. The CEOs reportedly freely discussed Amazon’s pricing because the publishers did not believe they competed on price, only on authors and titles. Id.
3 The Court sets forth, in minute detail across over 40 pages, the events that led to the iBookstore launch, the publishers’ successful efforts to convert Amazon to an agency model, and to effectuate higher ebook prices on new releases and New York Times bestsellers. Slip Op. at 1-44. The findings largely echo those of the district court’s 160-page opinion. United States v. Apple, Inc., 952 F. Supp. 2d 638 (S.D.N.Y. 2013).
4 There was much dispute among the judges as to whether the rule of reason applied. The dissent believed it did (and that it came out in Apple’s favor) and the concurrence held that per se illegality was the appropriate outcome and declined to address the rule of reason. The opinion also conducted a “quick look” rule of reason analysis, but it is entirely dicta as the Judge Raymond Lohier wrote separately to say that the per se rule was applicable and that the rule of reason was superfluous here, so he did not join in the majority’s rule of reason consideration. Slip Op. (Lohier, J., concurring in part and concurring in judgment). In dicta, the majority addressed two Apple arguments: (1) that the agreements allowed Apple to enter, and (2) prices for ebooks dropped since the iBookstore launch. Slip Op. (majority opinion) at 94-106. For the former, the majority opinion found that market entry was not a sufficient justification for a horizontal price-fixing conspiracy and whether Apple could profitably enter is irrelevant. Slip Op. at 94-104. For the latter, the majority stated that Apple failed to establish a connection proving that the conspiracy was necessary to create the procompetitive benefits of declining market prices and better competition between retailers. Slip Op. at 104-106.
5 The Court specifically notes: “But the relevant ‘agreement in restraint of trade’ in this case is not Apple’s vertical Contracts with the Publisher Defendants (which might well, if challenged, have to be evaluated under the rule of reason); it is the horizontal agreement that Apple organized among the Publisher Defendants to raise ebook prices.” Slip Op. at 75. The Court also points out that: “Because the reasonableness of a restraint turns on its anticompetitive effects, and not the identity of each actor who participates in imposing it, Apple and the dissent’s observation that the Supreme Court has refused to apply the per se rule to certain vertical agreements is inapposite.” Id.
6 This theory has rarely been successful, with only two notable cases: Broadcast Music, Inc. v. Columbia Broadcasting System, 441 U.S. 1 (1979), and NCAA v. Board of Regents of the University of Oklahoma, 468 U.S. 85 (1984).
7 See, e.g., US v. Delta Dental, 943 F. Supp. 172 (D.R.I. 1996) (holding that DOJ had survived a motion to dismiss on grounds that the MFN clause induced participating dentists from disaffiliating with other plans and caused dentists to maintain higher than normal prices to avoid losing significant revenue from Delta patients); US v. Blue Cross Blue Shield of Michigan, 2011 U.S. Dist. LEXIS 89849 (E.D. Mich. Aug. 12, 2011) (holding that BCBS’s MFNs with Michigan hospitals may have resulted in anticompetitive effects such as raising competitors’ costs, increasing premiums, and increasing costs to self-insured employers); Fiona Scott-Morton, Deputy Assistant Attorney General, Antitrust Division, Department of Justice, Presented at Georgetown University Law Center’s Antitrust Seminar: Contracts that Reference Rivals (Apr. 5, 2012), available at http://www.justice.gov/atr/public/speeches/281965.pdf.
8 The dissent, though largely discredited by the majority, made one startlingly stark point: the per se rule “does not apply to arrangements with which the courts are not already well-experienced.” Dissent at 22 (citing Leegin, 551 U.S. at 887). Traditional antitrust theory has, for over a century, held that the per se rule applies only in cases where the behavior or restraint so obviously harms competition that complete rule of reason analysis is unnecessary. Here, as the dissent points out, “no court has previously considered a restraint of this kind.” Dissent at 22-23. Apple has not indicated whether it intends to seek Supreme Court review of the case, but the dissent certainly provides the vehicle for formulating a petition for a writ of certiorari.