Allegations
The complaint alleges that Amazon engages in anticompetitive conduct in two proposed markets: (1) the “online superstore” market that serves shoppers and (2) the market for “online marketplace services” that serves sellers. The complaint highlights several allegedly anticompetitive acts, including anti-discounting measures that the complaint alleges deter sellers from offering lower prices outside of Amazon, an algorithmic pricing strategy known internally at Amazon as “Project Nessie,” and conditioning a seller’s ability to obtain “Prime” eligibility for its products on its use of Amazon’s fulfillment service, Fulfillment by Amazon (“FBA”). The complaint alleges that these actions prevented current competitors from growing and new competitors from emerging, harmed consumers, and allowed Amazon to maintain its monopoly power in both markets at issue.
Overview of the Markets At Issue
Mr. Meador described the two markets at issue. First, he discussed the features of the FTC’s alleged market for online superstores that serves shoppers. He highlighted 24/7 availability and shoppers’ ability to shop from any location as two features that distinguish online superstores from their brick-and-mortar counterparts. He also noted that the complaint distinguishes online superstores from other online stores and individual retailers by pointing to a broader product selection and convenience provided by online superstores. He then drew a parallel with the Department of Justice’s (“DOJ”) argument in United States v. Google distinguishing general search and more specific vertical search engines. He commented that the argument over whether Google competes with narrower search engines might shed light on how the FTC’s arguments would play out as it tries to distinguish Amazon from narrower online retail outlets.
Second, Mr. Meador described the features of the FTC’s alleged market for online marketplace services that serves sellers on the backend. Amazon’s online marketplace services provide sellers with access to a large customer base and tools to reach customers, among other services. He distinguished Amazon from a vendor that purchases and resells under its own name and from software-as-a-service providers that a seller might use to handle the infrastructure needed to sell something from its own website.
Overview of the Conduct At Issue
Pricing Practices
Dr. Backus and Mr. Berger followed Mr. Meador with a discussion of anti-discounting practices and arrangements related to lowest price guarantees described in the complaint. Dr. Backus explained that the complaint alleges a de facto most-favored nation (“MFN”) arrangement with sellers, which requires sellers to offer the same or lower prices on Amazon than on other channels. While Amazon does not have an explicit MFN clause in its policies, the complaint alleges that Amazon incentivizes sellers to adhere to this pricing practice. Sellers that offer lower prices on other channels can be removed from Amazon’s “Buy Box” or be restricted from selling certain items on Amazon, among other things. The complaint alleges that this de facto MFN policy can cause sellers to raise prices on other platforms to avoid jeopardizing sales on Amazon, which could lead to consumer harm.
Dr. Backus outlined three factors that economists consider in evaluating competitive effects of an MFN policy.
- First, what are the comparison prices used to evaluate whether a seller’s price on the channel at issue is no higher than that on other channels?
- If prices offered by other sellers on other channels are used for comparison, then requiring a seller to offer the same or lower price on Amazon is arguably pro-competitive because it is competing with other sellers on price.
- If prices offered by the same seller on other channels are used for comparison, then requiring a seller to offer the same or lower price on Amazon might lead the seller to raise its price on other channels.
- Second, if a seller’s prices on other channels are used for comparison, what are the incentives that govern sellers’ choices to lower their prices on the channel at issue or raise their prices on other channels (or do neither)?
- If absent the MFN policy a seller would charge a higher price on Amazon for certain products, then the MFN policy benefits consumers of those products. If absent the MFN policy a seller would charge a lower price on other channels for certain products, then the MFN policy harms consumers of those products.
- Third, what is the empirical evidence that the MFN policy affects pricing on other channels?
- Simple parallel pricing across Amazon and other channels is not convincing, because sellers may price equally across channels even absent any MFN policy. One empirical analysis could be analyzing fee changes on Amazon and whether they lead to price changes on non-Amazon channels.
Mr. Berger added that, while not all the details are available, the complaint also addresses first-party discounting by Amazon and alleges that Amazon uses an algorithm to automatically discount its own products and undercut its competitors. Mr. Berger noted that if Amazon is automatically lowering prices on its platform to have the lowest price, it might be considered price competition, which, unless it leads to predatory pricing, may not be anticompetitive or harmful to consumers.
FBA Allegations
Mr. Berger continued to discuss allegations related to FBA. According to the complaint, sellers on Amazon must use FBA, which includes storage, packaging, and shipping services provided by Amazon, for their products to be Prime eligible. The FTC alleges that a seller without Prime eligibility is denied access to a majority of Amazon buyers. The FTC also states that sellers could “multihome”—selling their products across multiple online sales channels—more cheaply and easily by using independent fulfillment services. According to the complaint, by tying Prime eligibility to FBA, Amazon is allegedly restricting the sellers’ ability to multihome and stifling the growth of independent fulfillment services.
Mr. Berger posited that if Amazon is merely selling FBA as a service that guarantees two-day shipping to Prime subscribers and requiring sellers to use FBA to have access to Prime subscribers, it would be challenging for the FTC to argue that there is a tie between FBA and Prime eligibility.
Allegations related to algorithmic pricing
As for the allegations related to algorithmic pricing, specifically Project Nessie, Ms. Schiller noted the extensive redactions in the complaint related to this topic. Ms. Schiller referenced an Amazon blog post from 2018 that described Project Nessie as a system to monitor spikes or trends on Amazon. Ms. Schiller also referenced a Wall Street Journal article that described Project Nessie as a pricing algorithm that tested how much Amazon could raise prices in a way that competitors would follow. For example, if competitors did not raise their prices the algorithm would revert to old prices, but if a competitor offered a discount, then Amazon and potentially other competitors would match that discount, and prices would stay low. One Amazon spokesman was quoted, stating that Project Nessie was intended to stop Amazon’s price matching from resulting in unsustainably low prices.
Ms. Schiller raised questions about how Project Nessie might align with other conducts alleged in the complaint, given its abandonment in 2019. She also emphasized the importance of understanding whether the algorithmic pricing is inherently different from manual pricing and whether Project Nessie can be considered price matching at a larger, automated scale. Overall, Ms. Schiller highlighted the need for more information about Project Nessie to better assess its role in the case.
The Case and the Broader Landscape of Technology Antitrust Cases
The conversation then shifted to the broader context of antitrust enforcement and legislation related to large technology companies.
Mr. Berger stated that enforcement agencies’ efforts to rein in perceived antitrust violations by big tech companies are occurring in a world where the same tech companies are often revered by the public which oftentimes does not perceive the problems that antitrust enforcers see. He noted that enforcers often target conduct regarding sellers or advertisers rather than consumers and pointed out two overarching issues with technology cases. First, there is a lack of a clear discussion about what the antitrust enforcers believe the pro-competitive world should look like and how consumers would benefit. With respect to this matter, the FTC might be envisioning a world in which consumers have multiple superstores to choose from. Second, there is no specific remedy being sought in these cases. Specifically, the Amazon complaint does not explicitly state a remedy for the various problems it alleges. Mr. Berger suggested that the agencies are looking to the Microsoft case from the 1990s in which no remedy was specified at the start of the case.
Mr. Meador then discussed the potential legislative impact of this case. He noted that Amazon was widely seen as one of the targets of Congress’s proposed American Innovation and Choice Online Act, which aimed to prohibit self-preferencing by large tech companies. Mr. Meador commented that if the case is successful, existing antitrust enforcement mechanisms could be seen as sufficient for checking large tech companies. However, if the case fails, it could be used to argue for a greater role of legislation or stronger enforcement. Mr. Meador also highlighted that the FTC’s complaint includes broad allegations about various aspects of Amazon’s conduct. He argued that such breadth could provide legislators with potential ideas for new bills to address perceived issues, even if they do not constitute clear antitrust violations under current Section 2 of the Sherman Act or Section 5 of the FTC Act.
Ms. Schiller pointed out that this case would provide more insight into the FTC’s approach outlined in its policy statement on Section 5 of the FTC Act. She noted that this case would be a test for whether Section 5 extends beyond the Sherman Act and that the outcome of the case and potential appeals could offer more clarity on the application of Section 5 in the context of antitrust enforcement.
The FTC’s Vision of the But-For World and the Alleged Harms
Dr. Backus discussed two types of alleged harm. First, there is the typical consumer harm caused by higher prices due to anti-discounting practices and tying Prime eligibility to FBA. Second, there is a more speculative harm involving the stifling of competition in the market for online platforms. Dr. Backus elaborated on the FTC’s dual narrative of barriers to entry and denying scale to competitors. With respect to barriers to entry, he explained that if Amazon’s practices prevent competitors from effectively undercutting and competing with Amazon through lower costs or seller fees, that constitutes a barrier to entry that is familiar to economists. With respect to denying scale, he pointed out that such language could invite criticism that the case is about protecting competitors rather than competition. He argued that Amazon’s defense might be that it grew to what it is by focusing on growing through a better product and offering a superior value proposition to consumers.
Dr. Backus also emphasized the intersection of the anti-discounting/MFN claim and tying claim in this case, noting that the de facto MFN claim potentially strengthens the tying claim. The traditional efficiency defense in a tying claim is that consumers value the combination of the tied and tying product (for example, fast shipping times and low prices). But this efficiency defense may be inconsistent with the de facto MFN, if the de facto MFN is found to shield Amazon from facing lower prices on other, potentially more efficient, sites.
Mr. Berger noted that the key alleged punishment following the violation of the de facto MFN provision is demoting sellers’ products in search results, according to the FTC. Mr. Berger questioned the effectiveness of this alleged punishment if, as the FTC claims, Amazon is saturated with advertising. He suggested that the FTC’s characterization of advertising on Amazon might undermine the harm to sellers that the FTC is alleging.
Ms. Schiller offered some thoughts about the implications of the case in the context of evolving market practices over time. She pointed out that the complaint focused on practices that Amazon has abandoned, such as an explicit MFN provision and Project Nessie. This complicates the litigation process and the drafting of decrees and poses a challenge for the FTC in shaping effective remedies.
Questions from the Audience
Next, Mr. Murray opened the floor for questions from the audience.
The first question addressed the naming of individual Amazon executives in the case. Mr. Berger explained that while the executives are not being sued, it has become more common for enforcement agencies to name individuals in civil antitrust cases.
The second question concerned whether the FTC’s allegations can be best understood as a concern regarding monopsony in the online services market and FBA. Dr. Backus commented that the focus of the complaint appears to be more on a two-sided platform market rather than monopsony.
The final question asked to what extent lowest price guarantees in other retail contexts might be analogous to this case. Mr. Meador suggested that Amazon might argue that its practices are analogous to lowest price guarantees, while the FTC contends that Amazon is punishing sellers for attempting to offer lower prices on other platforms. The key factual dispute would be which of those statements better characterizes the situation. Mr. Berger added that there is a significant difference in effectiveness between Amazon’s alleged anti-discounting practices and traditional lowest price guarantees, with Amazon’s automated algorithms being almost 100% effective.
Closing Comments
The discussion ended with the panelists’ final thoughts and predictions of the case. Ms. Schiller emphasized the importance of relevant market definition, especially in the context of online superstores. Dr. Backus expressed excitement about the empirical analyses based on data produced by Amazon as well as the choice of experts for the case. Mr. Meadow predicted that the FBA-related allegations might be a heavier lift than the anti-discounting practices, if the data support it, and that the FTC Act Section 5 claim remains a wildcard. Mr. Berger noted that the progression of the case may be influenced by a change in administration, as a more risk-averse office might be more willing to accept a settlement. Mr. Murray concluded the panel by highlighting the complexities and uncertainties surrounding the case, acknowledging that much will depend on the facts as they emerge during the litigation process.
