Procedural History
In December 2023, Amazon filed a Motion to Dismiss (MTD), arguing that the alleged anticompetitive conduct reflects “common retail practices that presumptively benefit customers,” and that the FTC had not pled facts sufficient to support a monopoly maintenance claim. With respect to the FTC’s authority to bring the case under its standalone authority, Amazon argued procedural error, claiming that the FTC lacked the statutory authority to bring the case directly to an Article III court, without an a priori “unfairness” finding by the FTC’s administrative tribunal.
On September 30, 2024, Judge John H. Chun denied most of the MTD (with the exception of a few of the state law issues), finding that Amazon’s responses stated procompetitive benefits that were not appropriate to consider at this stage. Notably, the Court found that there was no procedural error with regards to the appropriateness of the forum for the standalone claim, distinguishing the factual and substantive context of the Sperry case. The Court also noted that the FTC’s 2022 Policy Statement Regarding the Scope of Unfair Methods of Competition Under Section 5 of the Federal Trade Commission Act (“2022 Statement”) was not contested as a legal matter by either party.
Burden of Proof/Substantive Claims
The FTC’s substantive claims fall into two broad theories of harm. The first is Section 2 monopoly maintenance, which includes elements of foreclosure and “monopoly leveraging” theories. The second is the Section 5 UMC, which can be broken into a “non-standalone” and a standalone component.
The respective burdens of proof for these claims will be critical in dictating the parties’ respective strategies and possibilities for success.
A. Section 2 Claims
In order to meet its burden of proof for a Section 2 claim, the plaintiff must prove that a monopolist has engaged in exclusionary behavior (rather than the mere “exclusion” of a competitor). Under the Supreme Court’s Grinnell framework, this requires proving two elements: (1) the possession of monopoly power in the relevant market and (2) the willful acquisition or maintenance of that power as distinguished from growth or development as a consequence of a superior product, business acumen, or historic accident.”
As a general matter, “monopoly power” means the power to control prices or exclude competition. The plaintiff can establish the possession of monopoly power through either direct or indirect evidence. Direct evidence would be “direct proof of [an] injury to competition which a competitor with market power may inflict” (e.g., restricted output, supracompetitive prices). Through the more common indirect evidence method, the plaintiff must define the contours of that market, establish that the defendant has sufficient market share to exercise monopoly power, and show that there are barriers to entry to the market.
To prove the “willful” maintenance of monopoly in a relevant market, the plaintiff must demonstrate that the defendant undertook anticompetitive conduct with the intent to control prices or exclude competition. “Anticompetitive conduct” is conduct that impairs the opportunities of rivals and does not further competition on the merits (or does so in an unnecessarily restrictive way).
Under the Microsoft burden-shifting framework, proving a Section 2 claim requires that a series of “rule of reason” based steps be met. First, the plaintiff “must demonstrate that the monopolist’s conduct indeed has the requisite anticompetitive effect…it must demonstrate that the monopolist’s conduct harmed competition, not just a competitor.” (“Step 1”). Second, if established, the “monopolist may proffer a “procompetitive justification,” defined as a “non-pretextual claim that its conduct is indeed a form of competition on the merits.” (“Step 2”) Finally, if the defendant proffers a procompetitive justification that is unrebutted, then the Court conducts a “balancing” test, where the plaintiff “must demonstrate that the anticompetitive harm of the conduct outweighs the procompetitive benefit,” sometimes construed as whether “procompetitive efficiencies could be reasonably achieved through less anticompetitive means.” (“Step 3”).
Step 1: The FTC’s Prima Facie Case
Market Definition
In his Order, Judge Chun did not consider the nature of the relevant market, noting that Amazon conceded that any dispute as to the relevant market need not be resolved at this stage, while maintaining their argument that the FTC’s “highly-gerrymandered market is unlikely to survive that [relevant market] test.” The FTC alleges two relevant markets: (1) a market for online superstores (OSS), and (2) a market for online marketplace services (OMS).
The OSS market is defined as effectively a submarket of all retail outlets, distinguished by factors such as the breadth and depth of its product selection (i.e., “one stop shop”), its 24/7 availability, data aggregation and consumer purchase pattern recognition capabilities, and consumer-friendly research and filtering tools. Amazon contends that the FTC undersells the ability of customers to shop at physical or digital stores that sell individualized products, and that there is nothing inherent about the “one stop shop” nature that would prohibit a consumer from treating these product-specific or physical stores as reasonably interchangeable.
The OMS market is defined by the unique characteristics that constitute a “marketplace,” as well as the technical and structural factors that make it unique and not reasonably interchangeable with other physical and digital outlets. The FTC contends that businesses use a marketplace to “sell goods directly to U.S. shoppers [and] rel[y] on the marketplace to attract shoppers rather than attracting shoppers solely on their own.” Further, the FTC claims that the variety of pricing and transaction structures involved when vendors sell to retailers (e.g., bulk sales, seller forfeiture) varies from what is offered in a one-stop-shop purchase of online marketplace services.
During discovery and trial, the parties will likely heavily focus on providing expert reports and other empirical support to quantify their perceptions of consumer behavior. As Jonathan Rubin has noted, the economic coherence of a market is related to the cross-price elasticity of demand between competing products. This elasticity (or responsiveness of one product to a price or other variable change in another product) is not necessarily uniform across different groups or classes of customers, because the nature and needs of a particular customer might affect the degree of elasticity between a product or groups of products. For example, an online superstore that offers various firm-friendly products and services might appeal to large offices with recurring supply needs, but an individual consumer might be more willing to consider alternative physical or digital options.
Under the hypothetical monopolist test, a “candidate” market is deemed a relevant market if a hypothetical monopolist could profitably enact a small but significant, non-transitory increase in price (SSNIP) of 5% over a certain period of time. Any quantitative and/or qualitative data that can capture consumer behavior under different conditions would be persuasive in dictating the bounds of a relevant market (e.g., data on amount of Amazon shoppers that would substitute away from one-stop-shops under a given price increase, or consumer behavioral surveys that show consumers consider local stores that sell individual products within a certain geographic radius), although econometric and other high-quality SSNIP data can be difficult to obtain.
Monopoly Power
Once the markets are properly defined, the FTC would need to establish that Amazon possesses the requisite monopoly power in those markets, through either direct or indirect evidence. In its Complaint, the FTC claims that Amazon’s ability to profitably increase prices to its sellers and degrade its search quality to prioritize paid or sponsored advertisements is sufficient to constitute direct evidence of its monopoly power. Amazon has re-framed this as evidence of legitimate price competition and the featuring of customer-preferred search results. Given the high threshold and limited range of acceptable evidence of direct effects courts have generally accepted, the Court will almost certainly focus on indirect evidence.
For indirect evidence, the FTC claims that Amazon’s share of the overall value of goods sold by online superstores is well over 60%, and more than 71% of the online marketplace services market. With respect to barriers to entry, the FTC argues that certain unique features of platforms make it more likely that Amazon’s dominance has or will result in anticompetitive effect, citing economics of scale, network effects, reputational barriers, and shopper switching costs.
To make its case, the FTC will need to demonstrate how Amazon’s anti-discounting, self-preferencing, and seller coercion exist within an environment of high entry barriers. The FTC attempts to utilize Amazon’s statements about the importance of scale as a “key differentiator” from other online stores to show the barriers to entry, but the strongest evidence would demonstrate how scale creates a “lock-in” effect sufficient to prevent rival firms from reaching minimum efficient scale (MES), thus creating the conditions for durable monopoly power (e.g., the importance of user-generated reviews on Amazon to keeping customers in-house and bringing in new customers).
This ties into the notion of network effects, which can combine with scale to create a commercial environment in which customers are not willing to look elsewhere because of pre-existing platform dynamics (e.g., feedback loop between product ratings/reviews and attracting new customers). At the same time, any lock-in and network effects are premised on the isolation of customers and the durable value of data. As the FTC acknowledges in the seller coercion context, multihoming is an “especially critical mechanism of competition in online markets,” and can enable “rivals to overcome the barriers to entry and expansion that scale economies and network effects can create.” The durability and usability of data can also vary by context, meaning that particular qualities of the locked-in customer data are important in determining whether scale and network effects can actually deprive rivals from reaching minimum efficient scale (e.g., can Amazon’s rivals use another data collection method or approach to attract marginal consumers, get more value out of their existing data, or might Amazon’s existing data go “stale” over time). In this way, the FTC would have to demonstrate not only the existence of scale and network effects, but their ability to create durable monopoly power for Amazon, under the context and conditions of the market.
The FTC also argues that Amazon has been able to establish a positive reputation with its customers by enabling repeat purchasing, such that it creates reputational barriers to entry (i.e., sort of a lock-in of a fixed amount of trust for Amazon). This may be the weakest argument for the FTC, in that it is premised not on the inability of competing firms to reach MES, but the mere difficulty of competing against a more established competitor. The FTC would likely need to bring very strong evidence that customers are unwilling to consider trusting other firms (e.g., consumers treat trust as a finite resource).
Finally, the FTC claims that data retention and technical convenience create high switching costs for consumers. To prove this, the FTC would need to offer evidence that these inconveniences present such an obstacle that rivals could not reach MES. For example, the Complaint argues that Amazon’s ability to retain and analyze certain customer data (i.e., payment, shipping, order history information) means customers would find it too inconvenient to switch to a rival, and that they have developed routines that can be hard to break. However, Amazon could argue that multihoming creates low switching costs, or that customer data grows “stale” over time. The FTC would need to offer good evidence that can at least somewhat quantify how high and durable these switching costs are (e.g., consumer behavioral surveys that show large majorities would not switch for these reasons).
Competitive Effects
If the FTC can prove the possession of monopoly power in a relevant market, it would still need to show that this monopoly power results in competitive harm. The unifying theme of the FTC’s claims are that they are premised on a foreclosure or foreclosure-adjacent (i.e., “monopoly leveraging”) theory of harm. Demonstrating anticompetitive harm sufficient to meet the FTC’s prima facie burden would require showing more than mere harm to competitors, but harm to the competitive process itself (i.e., substantial foreclosure sufficient to deprive rivals of the ability to reach MES). The anticompetitive conduct alleged can be broken into three categories: anti-discounting (e.g., price-surveillance, algorithm manipulation), self-preferencing (e.g., quality degradation of search and advertising results), and seller coercion (e.g., leveraging of Prime-badge control over use of order fulfillment services).
Anti-Discounting
The FTC defines “anti-discounting” as tactics meant to discipline sellers who offer lower-priced goods elsewhere. Amazon is said to use a “price-surveillance group” that monitors whether third-party sellers who use Amazon Marketplace offer lower prices on competing platforms, punishing these sellers by depriving them of access to the Amazon “Buy Box” (i.e., display feature where buyer can easily click “Add to Cart” or “Buy Now”). The FTC also alleges that Amazon used an algorithm that would copy competitor changes in price “to the penny,” disallowing a “perfectly competitive market.”
Though Judge Chun found that these claims were sufficient to survive Amazon’s MTD, he noted that Amazon’s responses were procompetitive justifications that should not be prematurely considered. Looking at Step 1, Judge Chun seemed to provide enough legal and factual direction for the FTC to build its case-in-chief. For example, the Judge found that the anti-discounting claims were in effect foreclosure claims (and not predatory pricing claims, as Amazon argued). The upshot of this, however, is that the FTC would have to prove substantial foreclosure sufficient to deprive its rivals of the ability to reach MES.
Similarly, for the algorithm manipulation claim, Judge Chun frames the claim with respect to scale, citing portions of the Complaint that claim that rivals are deprived of the ability to gain scale by offering lower prices, including a portion that shows that the algorithm matches price movements in either direction. For purposes of the MTD, the Judge accepted that an Amazon employee’s statements could be evidence of anticompetitive intent, which is relevant for an anticompetitive effects analysis under Aspen Skiing. But it is well-established that anticompetitive intent is merely circumstantial evidence of anticompetitive effects, and that these effects need to be proven to meet the plaintiff’s prima facie burden.
To meet its prima facie burden, the FTC will need to show more than rivals’ inability to lower prices for fear of Amazon’s retaliation, but that the threat of this retaliation substantially foreclosed a sufficient number of sellers from listing on rival platforms (e.g., quantitative evidence that a critical mass of sellers did not list on rivals for fear of being deprived of access to the Buy Box). On a MTD in a parallel private class action in the Western District of WA, Judge Martinez allowed a similar MFN-based claim to proceed, but noted that “no Court has ever found a policy like [this] to violate the Sherman Act,” although this “does not, in itself, render these claims implausible.”
Self-Preferencing
With respect to quality degradation, the FTC claims that Amazon punishes sellers for offering a lower price on a rival platform by moving the displayed results for that product down the search list and/or concealing the prices for those products, even if the price for the product is the lowest one on Amazon’s platform. Amazon is also alleged to prioritize paid and private label advertisements over organic results, resulting in inferior results being prioritized and the gaining of a competitive advantage over third-party sellers (i.e., third-party sellers have to pay advertising fees in addition to transaction fees if they want to be placed at the top of the list, giving Amazon’s own products a cost advantage).
Though Judge Chun did not seem to address the advertising claim directly, he addressed Amazon’s general claim that its self-preferencing conduct constitutes a refusal to deal, finding that Amazon’s reliance on the Dreamstime.com case is distinguishable because the conduct both does not constitute a refusal to deal (i.e., more like exclusionary or foreclosure-type conduct), and because refusal to deal doctrine does not hold in any case when the purpose is to create or maintain a monopoly.” At the same time, he acknowledges Amazon’s argument that “steering” customers toward particular search results or items is a procompetitive justification, and Amazon’s MTD suggests it will argue that this steering is a legitimate design quality decision.
Although Judge Chun rejects refusal to deal as the applicable doctrine, the FTC would still need to demonstrate that Amazon’s self-preferencing substantially excludes or forecloses third-party sellers from offering lower prices on rival platforms. The FTC will likely have to bring strong, causal evidence of foreclosure or exclusion in the relevant markets to meet its Step 1 prima facie burden (e.g., the number of sellers foreclosed from rival platforms, attributable to Amazon’s threat of search results demotion).
Seller Coercion
The FTC also makes a form of a monopoly leveraging claim, claiming that Amazon uses coercive tactics over third-party sellers by tying the provision of a “Prime” badge (i.e., a graphic marker that denotes which items are eligible for Prime-shipping benefits) to the use of Amazon’s order fulfillment service (“FBA”) to deny rival platforms the scale necessary to compete in the relevant markets. Amazon argues that it does not tie the Prime badge to FBA, offering an alternative through the “Seller-Fulfilled Prime” (“SFP”) option that allows for third-party order fulfillment, as long as sellers meet Amazon’s terms and standards for FBA. Furthermore, there is a factual dispute as to the current availability of SFP, which the Judge did not consider for purposes of the MTD.
As described in Dreamstime.com, a monopoly leveraging theory includes allegations that the defendant gained or attempted to gain a monopoly in a second, downstream market. For the FTC, this would mean proving that the provision of a Prime badge exists within the context of an independent market in which it has monopoly power, and that power is then leveraged over an independent, downstream market in which the FBA exists (e.g., Prime badge is construed within the OMS market, and leveraged to gain power in a downstream OSS market).
Besides the fact that it is unclear whether the Prime badge and FBA exists within two different relevant markets (e.g., the badge is given to third-party sellers within the context of the Amazon Marketplace, which would seemingly implicate the OMS market, and leveraged over forcible use of Amazon’s fulfillment service, which would also seem to imply third-party seller involvement and thus the OMS market), the FTC would still have to demonstrate that monopoly power was gained on the fulfillment side, or that Amazon had a specific intent to monopolize order fulfillment with a dangerous probability of success. The Complaint alleges “twin mechanisms” through which this occurs: (1) raising the costs for sellers of using multiple sales channels; and (2) artificially stunting the growth of independent fulfillment providers, which work to deny rivals meaningful scale.
For the multiple sales channels claim, the FTC claims that Amazon’s tying of the Prime badge to FBA imposes “unnecessary and additional costs” that “reduces sellers’ incentives to offer their products and invest resources into selling on multiple online superstores.” However, the FTC also seemingly acknowledged that sellers benefit from platform-based aggregation, and that Amazon recognizes and takes advantage of this. This would seem to create space for Amazon to argue that it offers efficiencies to sellers and consumers, and has at most gained a competitive advantage over rivals by doing so, with no specific intent to monopolize. As such, the FTC would likely need strong evidence that speaks to an intent to monopolize (e.g., internal documents that speak to pretextual efficiencies and leveraging goals), and plausible evidence to support rival foreclosure.
For the artificial stunting of rival growth claim, the FTC seems to segment the OMS market into two, claiming a submarket of “online retail fulfillment services,” over which Amazon uses leverage over Prime badges in the OMS market to reduce volume in the online retail fulfillment market, and maintain its monopoly in the OMS and OSS market. But as with the multiple sales channel claim, the FTC seems to acknowledge the benefits of aggregation, claiming that “Amazon recognizes that scale is necessary to build an efficient online retail fulfillment network,” and that “independent fulfillment providers, too, would benefit from large fulfillment volumes that can help them scale and reduce costs.” If true, the FTC would need to show that the efficiencies gained represent something more than mere competitive advantage to make a cognizable leveraging claim.
Moreover, to prove monopoly power in a second, downstream market, the FTC would have to prove that the use of independent fulfillment providers raises input costs on sellers to the point of preventing them from using rival platforms, thereby preventing those rival platforms from being able to reach MES. This will require evidence that speaks to average costs and the amount of foreclosure that occurs from Prime-FBA tying (e.g., analyses of how much seller costs are raised from having to involuntary split costs between FBA and other fulfillment providers), tied to the impact of that foreclosure (e.g., is a Prime Badge actually necessary to compete on Amazon; even if so, does any forced FBA usage cover a sufficient volume of commerce to keep rivals below MES).
Step 2: Amazon’s Procompetitive Justifications
Under Step 2 of Microsoft’s burden-shifting framework, the defendant may offer non-pretextual, procompetitive justifications for its conduct. The procompetitive justifications come in three forms: price efficiencies, design efficiencies, and delivery efficiencies.
Price Efficiencies
In its response to the FTC’s anti-discounting claims, Amazon argues that its price surveillance and third-party tracking algorithms are meant to create value for consumers on price by matching discounts from rival platforms. Amazon notes that the FTC does not make out a cognizable predatory pricing claim, framing the claim as attacking legitimate discount-matching competition. Though the FTC’s claim is really a broader, foreclosure-based claim, the fundamental issue at this stage would be whether Amazon’s “anti-discounting” conduct (e.g., disqualifying sellers from the Buy Box if a seller is found to be offering a lower price on a rival platform, contractual restrictions on certain important sellers) offers cognizable procompetitive justifications. The Ninth Circuit has seemingly accepted the broad definition of potential cognizable efficiencies noted in Microsoft, defining a procompetitive justification in Qualcomm as “a nonpretextual claim that its conduct is indeed a form of competition on the merits because it involves, for example, greater efficiency or enhanced consumer appeal.”
Under this rubric, Amazon will be able to claim that any contractual restrictions or Buy Box demotions are made in order to offer consumers a unique, differentiated, and trustworthy product. The FTC will likely claim that Amazon’s justification is pretextual, arguing that the real purpose of Amazon’s conduct is to punish sellers who attempt to price compete with Amazon through rival platforms, which it will likely seek to demonstrate through internal Amazon statements and documents.
Design Efficiencies
In response to the FTC’s self-preferencing claims, Amazon argues that its conduct is aimed at offering a better in-store experience. Similar to the Buy Box or “Featured Offer” demotions, Amazon claims that this is a design efficiency meant to enhance competition (i.e., economizing on consumer search costs by featuring offers Amazon thinks consumers may like, and facilitating competition by encouraging comparison-shopping and building trust with consumer). For the claim that Amazon prioritizes paid advertisements which degrade search result quality, Amazon will likely argue that these paid advertisements allow Amazon to enhance the consumer experience (e.g., funds through ad payments are reinvested in platform infrastructure) and that organic search results are still accessible, albeit they appear further down the results page. The FTC will likely again argue that Amazon’s self-preferencing behavior is a pretext for anticompetitive exclusion (e.g., quality degradation intended to punish sellers out of listing on rival platforms), but the case law with regards to deferring to a firm’s design decisions as a procompetitive justification appears to be strong.
Delivery Efficiencies
In response to the FTC’s monopoly leveraging claim, Amazon argues that, even if the Prime badge was only displayed for those sellers who utilize FBA, there are reputational (i.e., represents Amazon’s promise of fast and reliable delivery) and quality-based (i.e., effectuate the Prime guarantee from order placement to delivery) justifications for doing so. Like price and design efficiency rationales, the delivery rationale would seem to have strong support in the case law. Even if the FTC could meet the Ninth Circuit’s stricter “intent” standard for making out a monopoly leveraging claim, the FTC would likely have to have to show pretext or how the Prime badge/FBA tie combines with some other anticompetitive conduct to have an overall anticompetitive effect.
Step 3: Balancing Test
If Amazon is able to proffer a procompetitive justification that is not rebutted, the FTC would have to show that the “procompetitive efficiencies could be reasonably achieved through less anticompetitive means.” The Ninth Circuit has read a “substantiality” requirement into this test, finding that courts must “give wide berth to [defendants’] business judgments.” In defining the deference afforded to defendants’ business judgments, the Court has found that an alternative is “substantially” less restrictive if it is “virtually as effective in serving the defendant’s procompetitive purposes…without significantly increased cost.” The Ninth Circuit seems to define “significantly increased cost” in terms of its plain economic meaning. In a Section 1 tying and Section 2 attempted monopolization case involving a change in the credentialing criteria for physicians to operate at a hospital, the Ninth Circuit defined “significantly increased cost” as the burden on a hospital of individually evaluating the skill of particular physicians, rather than relying on Board certification. In a Section 1 restraint case, the Ninth Circuit addressed causality and degree, finding that the Court should intervene only when a restraint is “patently and inexplicably” stricter that what is needed to accomplish the stated procompetitive objective.
For the anti-discounting conduct, the FTC might argue that price competition can be facilitated without algorithmic surveillance or Buy Box demotion, but Amazon could argue that surveillance is necessary to efficiently find competitor prices (rather than individually scouring the Internet), and that Buy Box demotion is necessary to achieve the procompetitive justifications (i.e., build consumer trust, elevate lowest price on Amazon’s platform).
For the quality degradation/advertising claim, the FTC might argue that there are alternative methods by which revenue can be raised to fund design and consumer experience improvements, but the Court must give significant deference to Amazon’s business judgment (and almost certainly would not rise to the “patent and inexplicable” standard). Similarly, if the monopoly leveraging claim reached this stage, the FTC might argue that Prime’s reputation/brand quality could be maintained in other ways (e.g., greater availability of SFP and distinguishing of FBA), but this would likely also fall under the Court’s obligation to defer to business judgment (and Amazon could argue significant cost/reputational harm to reducing FBA in this way).
B. Section 5 Claims
Though traditional Section 2 monopoly maintenance seems like a difficult case to make, the FTC might have a better case under its Section 5 UMC authority. Critically, UMC claims are no longer bound by the scope restrictions of the 2015 Statement of Enforcement Principles Regarding “Unfair Methods of Competition” under Section 5 of the FTC Act [“2015 Statement”], which has given the FTC the doctrinal space to test out novel theories of harm under its UMC authority.
Having cleared certain procedural hurdles, the FTC’s case may now hinge on if and to what extent UMC doctrine is found to be more permissive. Specifically, the FTC re-alleges the monopoly maintenance-based counts as a UMC (“non-standalone” claim), and makes a separate, standalone allegation that an algorithm-based initiative called “Project Nessie” constitutes an unfair method of competition (“standalone” claim). We will note the general framework for applying Section 5 as a UMC, and then apply to each count in turn.
Background
The substantive scope of UMC law under Section 5 has been a subject of serious debate for decades. Though Congress delegated a wide range of authority to the FTC and the courts to determine the meaning of “unfair methods of competition,” there have been few substantive constraints imposed. Though the 2015 Statement tethered standalone authority to the language and scope of the Sherman and Clayton Acts, the 2022 Statement replaced this constraint with more ambiguous language. In addition to covering practices that violate the Sherman and Clayton Acts, the 2022 Statement lays out general categories of conduct that might be actionable as “incipient” violations, or for violating the “spirit” of the antitrust laws. Furthermore, the 2022 Statement limits the procompetitive justifications that can be used in defense, and notes that there are “non-quantifiable harms” that make a net efficiencies/ROR framework inappropriate.
The 2022 Statement defines “unfair” methods of competition as conduct that (1) may be coercive, exploitative, collusive, abusive, deceptive, predatory, or involve[s] the use of economic power of a similar nature; and (2) negatively affects competitive conditions.
In defining the first factor the FTC cites, among others, the Sperry and Ethyl decisions. In Sperry, the Supreme Court gave some content to what may constitute “exploitative” conduct in a price fixing and stamp-exchange suppression case, citing the R.F. Keppel case in finding a public-policy dimension grounded in the common law and criminal statutes. The Court also said that the there was “no indication in the Commission’s opinion that it found S&H’s conduct to be unfair in its effect on competitors because of considerations other than those at the root of the antitrust laws,” although it also found that “the Commission’s findings…go beyond concern with competition and address themselves to noncompetitive and consumer injury as well.” Though the FTC cites the Ethyl case for the proposition that the UMC reaches a broad range of conduct, it would seem to stand for the more limited proposition that Section 5 can reach noncollusive, interdependent business practices under some circumstances.
While the Sperry and Ethyl cases seem to remain tethered to the antitrust laws, the Texaco case may have taken a step outside of them. In Texaco, the Supreme Court found that, in a situation where an oil company had dominant economic power over its service station dealers, the dealers were effectively induced to buy tires, batteries, and other accessories from the company in a manner that “tended to foreclose competition,” which would seem to at least open to door to something less than the Section 2 foreclosure standard being sufficient for a standalone violation. But the case is also distinguishable, in that even if sellers were found dependent on Amazon Marketplace for their livelihoods and thus hesitant or unwilling to list on rival marketplaces, there are different competitive dynamics and acceptable procompetitive justifications that have arisen since the 1968 Texaco case, as well as subsequent case law that recognizes some form of a limiting principle.
In defining the second factor, the FTC seems to focus on citing cases that obviate the need for the plaintiff to prove anticompetitive consequences or effects. For example, in a footnote to In re Coca-Cola Co., the FTC in a final order cites the Sperry and Dean Foods cases for the proposition that Section 5 standalone authority was meant to combat incipient acts or practices that might result in a violation of the Sherman or Clayton Acts. Though the Qualcomm case did not reach the merits of an alleged standalone violation, the CA District Court notes the breadth of the UMC, though qualifying this by saying this authority is not “unbounded.”
Read together, the case law seems to suggest that UMC authority, though not unbounded, maintains a wide degree of latitude to encompass business conduct that is broadly anticompetitive (e.g., coercive or exclusionary, without any actual anticompetitive effect) and conduct that may grow into (when viewed in the context of the firm itself or its relevant market) an anticompetitive violation. In this way, though Section 2 may not have the breadth to reach Amazon’s conduct, there is nothing in the case law that necessarily precludes the FTC from trying to reach the conduct through the principles enumerated in the 2022 Statement.
Monopoly Maintenance-Based Claim
In terms of the claim that re-alleges the monopoly maintenance-based counts as a UMC, the FTC may argue that Amazon’s price surveillance and algorithmic price manipulation is anticompetitive in spirit, not because it has any direct anticompetitive effect at present, but because it will over time discourage sellers from listing on Amazon Marketplace (e.g., Amazon’s growing dominance will increase the competitive salience of a Buy Box demotion, thus discouraging sellers from listing on rival platforms). Furthermore, the FTC might argue that, though Amazon has currently opened up SFP and claims it no longer ties FBA to Prime badges, the fact that Amazon can restart this conduct constitutes a potential future anticompetitive effect.
Given the capaciousness of the UMC’s statutory language and case law, the Amazon case will be a significant test case as to how far this authority can be pushed. A critical piece of this will be how Judge Chun construes the scope of a Section 5 UMC claim that is co-pled with a Section 2 claim (e.g., whether the fact that the UMC claim covers the same conduct as a concurrent Section 2 claim affects how he views the scope of UMC authority; whether he conceptually separates the scope of these “non-standalone” claims from the scope of standalone UMC claims).
Project Nessie Claim
An even more significant test might come from the standalone claim: that Amazon engaged in unfair competition through an algorithm that raised prices by manipulating rival pricing algorithms into matching price increases on its own platform. According to the Complaint, Amazon’s algorithm would predict when it thought rival stores would increase prices on a specific product; once identified, Amazon would raise prices on those products, thereby setting a new, higher competitive baseline. When public scrutiny would increase, Amazon would allegedly turn off the algorithm (though Amazon claims it discontinued use in 2019).
Though Judge Chun allowed the standalone claim to proceed under his reading of Sperry and because of Amazon’s ability to turn on Nessie again, the critical substantive piece is that Judge Chun sided with the FTC’s interpretation of Ethyl, finding that the so-called “parallelism plus” language in the case was capacious enough to allow the FTC to survive a MTD. In particular, the FTC was said to have pled sufficient facts to indicate that Amazon plausibly had “anticompetitive intent or purpose” in considering the structure of the FTC’s complaint, and the Judge distinguished a case that might have disallowed this claim under the Ethyl standard.
In this way, Judge Chun may be suggesting that the substantive scope of the UMC doctrine could include parallel price movements in a heterogenous market structure, when there is any indication of a potential anticompetitive intent or purpose pled by the plaintiff. The Ethyl court stated that there must be some “workable standard” to define this scope to avoid the risk of subjecting target companies to “uncertain guesswork rather than workable rules of law.” In the Abbott Laboratories case, the D.C. Court of Appeals considered a case in which the FTC accused the largest provider of infant formula of bid rigging, and “with anticompetitive intent…provid[ing] information that showed to competing bidders that defendant preferred…an open market instead of a sole source system,” acting “unilaterally…to obtain an anticompetitive outcome in the bid.” Though the Court found Abbott not liable because of the Government’s role in bringing about the outcome, the Court engages in an extended discussion of Abbott’s intent and knowledge throughout the duration of the conduct, noting that throughout litigation, Abbott “had nothing to hide and turned over all its information and records to the FTC…[and] waived all its legal privileges…[and] at no time took any action to prevent the FTC and the Court from obtaining all the relevant and material facts.” If the scope of the UMC is dictated by the defendant’s intent, knowledge, and good-faith during the litigation process, it would seem to leave the scope fairly open-ended, with the result possibly turning on the extent of Amazon’s compliance with court and discovery mandates.
At the same time, as noted in a dissenting statement to the 2021 rescission of the 2015 Statement, the Commission has not successfully litigated a standalone Section 5 case since the 1960s. As Bill Kovacic and Marc Winerman note in a 2010 article, judicial psychology has tended towards tying UMC doctrine to “standards familiar to [judges] from Sherman Act and Clayton Act cases,” and that the “cost-benefit concepts devised in rule of reason cases supply the courts with natural default rules in the absence of something better.” In a discussion of the interplay between Section 5 and Section 2, Kovacic and Winerman note that deference to agency expertise is highly contextual, and that non-competition arguments for application of the UMC might be treated with suspicion.
If, however, Amazon was found liable under the UMC framework, its procompetitive justifications would become highly relevant. Under the 2022 Statement, the FTC notes that the burden of persuasion remains with the defendant, but the benefits must be “of the kind that courts have recognized as cognizable in standalone Section 5 cases.” The FTC admits there is limited case law that speaks to UMC-based efficiencies, and seems to give itself broad authority to follow or not follow what the courts said in the earlier cases. To the extent that Judge Chun ties his efficiency analysis to a Section 2-based precedent, Amazon’s price, design, and delivery justifications would be well-grounded in the law, but the “sliding scale” framework in the 2022 Statement might create doctrinal space to limit or disregard the import of these efficiencies.
Conclusion
The merits of the FTC’s case are somewhat mixed. In terms of its Section 2 claims, the FTC faces a challenging task at each step of the analysis. It is not clear whether the relevant markets are defined properly (and Amazon has already indicated it plans to contest this vigorously). Even if they are, the FTC would have to put forward evidence that rival superstores/marketplaces are being deprived of the ability to reach MES, and/or demonstrate the intent or an attempt to monopolize a second market for the monopoly leveraging claim.
If the FTC does meet its prima facie burden, Amazon seems to have a good case that its price, design, and delivery procompetitive justifications are cognizable and non-pretextual. If one or more of the Section 2 claims are deemed cognizable and non-pretextual, the deference courts typically afford to business judgment and the difficulty in proving the viability of less restrictive alternatives would give Amazon a strong case at this step at well.
The larger questions will revolve around the Section 5 claims. Though Project Nessie would appear to be similar to the Section 2 algorithmic manipulation theories, the difference lies in the Court’s reading of Ethyl as opening the door to UMC expansion through the “anticompetitive intent or effect” language. Though the case law the FTC cites is from before the Chicago and Post-Chicago schools of economic thought began to take hold, the Amazon case will be the first significant test of the scope of the UMC doctrine since at least Abbott Labs, and the first since the inception of the 2022 Statement. The key question will be the extent to which the Judge takes the “workable standard” language of Ethyl and Abbott Labs seriously, and what evidentiary and proof standards he might place on any parallelism-plus type approach (e.g., what type of anticompetitive intent or purpose is sufficient to sustain a Section 5 claim; what are the relevant evidentiary and proof standards for different types of Section 5 conduct that might exhibit anticompetitive intent; how does the “sliding scale” approach of the 2022 Statement affect the analysis).
Finally, there are at least a couple of other issues that may have an impact on the case. For one, Judge Chun acknowledged that there are cumulative harm arguments that might be relevant for the trial. The 2022 Policy Statement acknowledges that it is within the FTC’s perceived scope of its powers to pursue these types of claims, and the Complaint makes allegations about negative spillover effects and inter-market harms.
Second, there is a question as to what causation standard will apply, and how it should be applied in this context (brought to the fore through the recent Google Search decision). The Amazon case offers the additional issue of how this causation debate might play out with respect to the Section 5 claims.
