ITIF Search

Assessing the FTC’s Complaint Against Amazon

Assessing the FTC’s Complaint Against Amazon
Tuesday, October 17, 202312:00 PM to 1:00 PM EST
Virtual Webinar

Event Summary

Since her famous law review article “Amazon’s Antitrust Paradox,” now FTC Chair Lina Khan and the rest of the neo-Brandeisian movement have made Amazon the top target in their crusade to reshape antitrust policy and take on “Big Tech.” With the FTC having now filed its long-awaited complaint challenging several of the company’s business practices, FTC v. Amazon will no doubt prove to be one of the defining cases that determines not just Chair Khan’s legacy, but that of the broader antitrust reform movement. And yet, following a string of recent losses against large technology companies in FTC v. Qualcomm, and the Meta/Within and Microsoft/Activision merger reviews, is the Amazon case set to be the next big defeat for the FTC? Does the case against Amazon double down on the bold neo-Brandeisian theories in Chair Khan’s law review article, or take a more mainstream approach? Is the FTC out to win, or instead just bring cases that highlight the limitations of existing antitrust laws as a ploy to spur congressional action?

Watch this expert discussion about the merits and implications of the FTC’s landmark challenge to Amazon.


Joseph V. Coniglio, ITIF: Okay, I think we can get started here. We’re just at the hour. We’ve got a lot to cover today. My name is Joseph Coniglio. I’m the director of the Antitrust Innovation Policy and the Schumpeter Project on Competition Policy here at ITIF. On behalf of ITIF, I want to welcome you all to this really excellent panel assessing the FTCs complaint against Amazon. The FTCs complaint was filed three weeks ago to the day, and it alleges that Amazon has engaged in any competitive behavior to maintain monopoly positions in two markets. First, the online superstore market, which encompasses Amazon’s overall e-commerce business. And then second, the online marketplace services market, which focuses on Amazon’s third party business connecting sellers with online buyers.

The FTC claims that Amazon has monopolized these markets through two distinct but mutually reinforcing sets of practices. First, quote, exclusionary anti-discounting conduct whereby Amazon has said to have artificially inflated prices and prevented price competition from rival retailers in marketplaces. And then second, allegedly coercing sellers to use its fulfillment services fulfillment by Amazon if they wish to be eligible for its prime product, Amazon Prime.

And finally, in addition to the monopolization claims, the FTC alleges that Amazon’s anti-discounting conduct and it’s coercing of sellers to use its fulfillment service is an overall course of conduct that constitutes a standalone violation of Section 5 of the FTC Act as an unfair method of competition. And additionally, the FTC claims that Amazon’s use of an algorithm known as Project Nessie is itself a distinct standalone violation of Section 5 of the FTC Act.

So to discuss the FTCs case, I’m very fortunate to have with me a truly August set of experts in antitrust scholars. First we have, I think on the far right of your screen, top far right, Kathleen Bradish. Kathleen is the acting president of the American Antitrust Institute. She served in government, including as assistant chief, international counsel at the US Department of Justice, Antitrust Division, in addition to having a very distinguished career in private practice at Cleary Gottlieb here in DC.

Second, right below her we have Herb Hovenkamp. Herb is the James G. Dinan University professor at the University of Pennsylvania Carey Law School. Herb is known affectionately by many as the dean of Antitrust. And among his vast corpus, he’s the co-author of the magisterial Areeda-Hovenkamp Antitrust Treatise. Third, on the bottom left of the screen is Steve Salop, professor emeritus of economics and law at Georgetown University Law Center. Steve is, I think, generally recognized as a, if not the founding father of modern exclusionary conduct analysis. Very relevant today, and he also has the added distinction of being my antitrust professor. So thank you, Steve. And finally, we have Bilal Sayyed, senior competition counsel at TechFreedom. Bilal is also a counsel at the law firm, Cadwalader Wickersham and Taft here in DC, as well as an adjunct professor at the Antonin Scalia Law School at George Mason University.

Bilal has had a long career, both in private practice and government, including most recently in the last administration as director of the Office of Policy Planning at the Federal Trade Commission.

So I’d like to welcome all of our panelists and get started with Herb. So Herb, I want to start on the threshold issue here of the FTCs monopolization claims, which is the question whether or not Amazon has monopoly power in a relevant market. So what do you make of the FTCs market definitions here? And specifically, why don’t we start with this online superstore market? It seems to have the best of both Philadelphia National Bank as a cluster market, but also a Brown Shoe submarket all kind of rolled into one. Is that going to hold up in court?

Herbert Hovenkamp, U. Penn. Carey Law School: I think FTC has got an uphill battle here. First of all, market power attaches to products, not to firms. For example, Microsoft may be a monopolist in its Windows operating system, but certainly not in its Bing search engine with a 3% market share. So the real question is, can you group all of these non-competing products together into a single one for determining market power? Brown Shoe said yes, you could provided that the shares of all the individual products in that case and children’s shoes were more or less the same. That’s clearly not the case in Amazon. We’ve got a little bit of precedent for doing it in merger cases such as Staples in which you have one firm acquiring the entire line of a different firm, we don’t have any equivalent precedent that I’m aware of in a merger case that lets you point to the store and then say, well, the store is a relevant market, and so we don’t need to worry about the individual products in that market.

And the fact is that Amazon’s individual product market shows has 12 million products. That’s why it’s so big range from somewhere in the low 60s for eBooks, assuming eBooks is a relevant market. And then for many, many products, well below 20%, 4% for tires, a little under 20% for toasters and so on. So Amazon’s individual market shares are all over the place. And furthermore, the customers buy individual products. That is to say from the customer’s perspective, the question is how many alternatives are there to that toaster or sweatshirt that you’re looking at on Amazon? Now, we do have some precedent, as you mentioned, for cluster markets, but those are typically where the only way you can get the product is from the cluster. So for example, a hospital is a cluster market. Anesthesiology and thoracic surgery are not competing products, but you still need to go to the hospital to get the product.

And the courts have generally rejected cluster market definitions in cases involving multi-product, department stores. Pay ‘n Pak is the leading example. The simple fact that a store sells 1500 different products, the court said doesn’t justify creating a single market for all of those products. So on that particular market definition, I think the FTC has an uphill climb. I think the online marketplace services is a little more defensible here. They’re more common to the overall range of products. The problem I have with that market definition is that everything in it looks incredibly easy to duplicate. I mean, it lists things like the ability for sellers to set the prices, the ability of sellers to maintain product pages or display customer reviews. Well, if anybody can do those things and at fairly low cost, then that doesn’t sound to me like a grouping of sales that’s capable of being monopolized.

Joseph V. Coniglio: Thanks, Herb. Kathleen, I wanted to ask you to follow up on that. So are these really easy services to duplicate? I mean, we know in these types of markets, our platform markets, a lot of it is having sufficient customer scale. But what’s to stop a brick and mortar retailer that’s very strong from opening an online marketplace here and competing with Amazon if it has a very large customer base?

Kathleen W. Bradish, American Antitrust Institute: Well, I think this entire case is about the barriers to entry and precisely that market. I think you have the perhaps unparalleled diversity of products, and you also have just the network effects of having so many different sellers and the need to be on Amazon in order to gain the eyeballs of shoppers. So I think while it may on the surface seem like something anyone could do, I think the barriers to entry that are described in this case that are in part created by Amazon go to the fact that it’s not that easy to replicate.

Joseph V. Coniglio: So Steve wanted to bring you in here, obviously ask you to comment, but also specifically, I know it’s a Section 1 case, but obviously the Supreme Court’s decision in Amex seems like it has some relevant material for this type of a marketplace definition. What’s your thought on that?

Herbert Hovenkamp: Joe, there are no Section 1 allegations in the complaint.

Joseph V. Coniglio: Right. No, but I’m wondering if those general principles can still be relevant or Steve, what you think.

Steven C. Salop, Georgetown Law Center: Well, I think that it would’ve been a lot simpler to define a symbol two-sided online superstore market and then treat the downstream consumer market as the primary focus. I think this would’ve actually solved hers problems as well. So let me explain. I generally take the view that market definition should be set out in the context of the anti-competitive conduct. So let me start with the conduct. The central allegation here is that Amazon’s two restraints, as you mentioned, the MFN or price parity provision and the FBA or fulfillment by Amazon requirement, those two restraints raise the merchant’s costs of selling on rival platforms and raises the retail prices charged on those platforms. And that applies to all the products one by one or collectively. Amazon then collects its monopoly profits on Amazon merchant sales by charging high FBA fees and other merchant fees such as pay-to-play advertising.

And so as a result, there’s harm on both sides of this two-sided market. Consumers are harmed by the high prices and again, responding to Herb that’s on all the products. And secondly, merchants are harmed by the higher Amazon fees, the higher costs of selling on rival platforms and the lower sales on rival platforms from the barriers to entry as Kathleen stressed. So that would’ve been a simpler way to do it.

Joseph V. Coniglio: Thanks, Steve. Bilal, I want to bring you in here. And Steve, if you wouldn’t mind muting. One of the things that the FTCs markets definitions exclude is brick and mortar. And I think we’ve already heard from Amazon that something like 80% of all retail sales occur offline. Is there really justification for the FTCs decision to wholesale sort of exclude brick and mortar retail from the relevant market?

Bilal Sayyed, TechFreedom: Well, Joe, the short answer is it’s hard to say, right? It’s a factual question. And I think what we have in the complaint is not a clear setting forth of what I’d call economic or econometric evidence that helps define the relevant market here. But let’s assume that what you might call a narrow online superstore market is a relevant market. I think though a relevant question still is to what degree the firms or offerings or alternatives outside of that market definition, constrain pricing and other conduct within that relevant market and market definition or products or services are not usually in or out, right? It’s sort of a series of concentric circles. How close are alternatives, how strong is the constraint?

So here I could see it. I don’t know that it would be the right relevant market to define. I agree with Steve that a two-sided market would’ve been better than two separate relevant markets. But the descriptive evidence of the online superstore market doesn’t strike me as very compelling. And it leads you to naturally question whether, even if it is a relevant market, what’s the constraining effect of alternatives? And I think the brick and mortar experience is a different experience in some positive ways, in some negative ways. An online shopping experience, not in a superstore is also a different experience. But there’s very little in the complaint, and I think likely very little in practice, this suggests that those alternatives do not have at least some and maybe significant constraining effect with respect to so-called online superstore market.

Herbert Hovenkamp: Joe, can I add from there?

Joseph V. Coniglio: Yeah, of course. Herb, please.

Herbert Hovenkamp: I agree with Bilal. I think this is an incredibly product specific inquiry. Not only is it a fact question, but it’s one, for example, for streaming, there is no offline market. The streaming has hijacked practically the entire market for videos and digital music today. At the other extreme, Amazon really struggles in markets like try on clothing where it’s tried various devices to substitute for trying on. Tires where they have to be installed by customers. Groceries, which are excluded, but they’ve had the same issues there. So product by product, the question of how much competition Amazon faces from offline sellers varies immensely. It varies from zero to a hundred percent in fact. And I don’t think you can come up with a one size fits all online only market definition under those circumstances.

Joseph V. Coniglio: Yeah, no, thank you, Herb. I think there’s definitely a sense of over breath here. But turning to the conduct itself, Steve, you started to get into that, but I want to turn to Kathleen specifically the anti-discounting practices that are at issue here. Kathleen, could you talk a bit about what the FTC is actually getting into and the extent to which is this a most favored nation? Is it something else? What are we looking at?

Kathleen W. Bradish: Well, I think we can understand it as a most favored nation’s policy, a defacto, most favored nation’s policy, I think it’s avoided in the complaint because it’s one of many aspects of conduct that is being challenged here, and it tends to confuse people. So I think it’s not called that specifically in the complaint for that reason. But what we see with most favorite nations clauses is a practice that has been challenged both here in the United States and outside the United States and Europe, for example, in which the requirement is that that lower prices not be offered on any other platform. So sellers using Amazon can’t discount to other online platforms. The effect of this, which makes MFN challenges sometimes controversial, can be both pro-competitive on one side and anti-competitive on another. And I’ll explain a bit about that in a second. And the difference between the two can sometimes be very hard to tell. And that really comes down to the facts of the case.

So on the pro-competitive side, there are arguments that MFNs, most favored nation discourage free writing and encourage investment. So you don’t have a platform or other service that’s investing a lot in making themselves a great location, a great service, and then lower those services are used by customers that then go and get a lower price on another platform and therefore discourage investment in the platforms in quality and service, et cetera.

On the other hand, there are some very real concerns that most favored nation policies, especially when the entity imposing the most favored nation’s policy, is very large, very important, or there are a number of MFNs are prevalent in a particular industry or market that this really acts as a form of price control, that it effectively means that it is way too expensive for a seller using this platform to discount to anyone else because they also must discount to, in this case, Amazon, and that this has an effect on customers because you don’t get the benefit of that discounting, selective discounting.

And on the other hand, it makes it very difficult for new entry to happen because a seller can’t say, I’m going to discount to you a new platform, or the platform can’t offer lower fees, for example, and attract new business by having lower prices than Amazon. So it has the potential for adverse effects for both consumers and for other potential platforms who would compete on fees, who would compete on service with Amazon. So I think that gives a basic picture of what we’re looking at, but we also have a lot of experts on MFNs here on the call. So I think we can open up on the floor on this because I think everyone here has a lot of background in this area.

Joseph V. Coniglio: Well, thank you, Kathleen. That’s very helpful. So speaking of one of those experts on exclusionary conduct, Steve, I want to go to you here. So I actually kind of remember our class on MFNs, and as I remember it, there’s some coordination and exclusion that’s at issue here with respect to this type of conduct. And I guess I wanted to ask you, is a predicate of the FTCs exclusionary theory of how these alleged MFNs or defacto MFNs, everyone characterize it, any discounting practices, exclude rivals? Is that predicated on Amazon being able to sort of maintain coordinated behavior across these very large markets that the FTC has defined? And if so, to what extent is that really even possible?

Steven C. Salop: Well, somebody’s not muted. My understanding is that what Amazon does is it has algorithms that scrape all the relevant websites, and if it discovers that the merchant’s product is being sold at a lower price anywhere else, it contacts the merchant and says, you’ve got to lower the price to us, or you’ve got to raise the price to them. And faced with that, the merchant raises the price on the other platforms or actually withdraws its product from another platform. If it can’t change the price, it’s like the vicarious liability on the merchant. If the merchant is selling to someone who’s reselling and the reseller is charging a lower price, Amazon holds the merchant responsible for that lower price and says, shape up or ship out buddy.

The other thing that’s going on here, I don’t know whether you want to talk about it now or later, is this Project Nessie issue?

Joseph V. Coniglio: Let’s bracket that for now, Steve.

Steven C. Salop: Okay.

Joseph V. Coniglio: But no, I do want to follow up on the price... The complaint talks about alleged price surveillance in this vast network that Amazon has to, as Steve suggested, make sure that this coordination would be occurring. But turning to you Bilal, if Amazon is a monopolist sort of connecting the two threads and it has this unilateral pricing power in these markets to raise prices, it seems a little bit odd that it would be monitoring all these retailers for pricing. At the same time, I would think it would have the ability to control prices on its own without having to engage in the sort of vast network of surveillance to see what its purported rivals are pricing. Does that make sense?

Bilal Sayyed, TechFreedom: Joe, that’s an interesting point. I hadn’t thought about it that way. I don’t think we have sort of an either or situation here, at least from the FTCs perspective. People have unilateral market power in a sense, when they’re able to punish, when they have the ability to punish other firms either through output or other punishment forms. So I don’t know that the fact that they monitor and in a sense follow up undercuts the idea of unilateral market power or monopoly power. But I also think it’s sort of a maybe not incredibly dispositive of how the courts will treat this. MFNs, clearly, I think everyone accepts MFNs can have anti-competitive consequences. The problem is, I think for the agency or any plaintiff is they can also, as Kathleen mentioned, have pro-competitive benefits and how the court structures the inquiry I think is going to matter a lot here.

Monopolization cases either have sort of rules of thumb for the evaluation of whether certain conduct by a monopolist is illegal, and that includes things like loyalty programs and other discounting practices. In other cases, they have in a sense formulas like predatory pricing. And then in a variety of cases or situations, you either, I think the courts either pretend to have apply sort of a cost benefits test, or if defendant comes forward with a benefit justification, the defendant wins. So I think the FTCs likelihood of success, which I think is low under just about any framework the courts use, but their best opportunity is I think if the court really tries to balance the effects here. And that is seems to me going to be very difficult across either individual products or bundles or baskets of products. Because if you think the basket in the first Staples, Office Depot merger analysis was complex, imagine how somebody might put together a basket or a bundle of Amazon sold products, either first party or third party, and show an effect, a negative effect or positive effect.

I really think the FTC has hurt, I’m getting a little off the question, but I really think the FTC has hurt itself here by bringing a very broad case involving basically a single company alleging that they have seized control over much of the online retail economy, I mean, this is a huge case. I don’t see how they’re going to manage it effectively and including in the presentation of evidence of effects.

Joseph V. Coniglio: No, thank you, Bilal. That’s exactly where I wanted to go. And Herb, looking at you here, we do have a private class action going on where this claim is being judged in the same court, Western District of Washington where the FTC brought its complaint and we had an motion to dismiss decision where the court basically said, I think what a lot of folks are saying here that this kind of makes sense, it’s a plausible economic theory, the benefits harms. But it also, I think, noted that there’s never been a case where a court has actually found a policy like this to actually be anti-competitive. Is that true? And if so, how does that bode for the FTC?

Herbert Hovenkamp: I think it’s true, and especially on this MFN or quasi MFN issue, MFNs only work on a product specific basis. We generally, they’re under the rule of reason. So if I’ve got a 10% market share in my toaster, I’m buying it at wholesale, I want an assurance that other retailers are not getting a better deal than I’m getting, think of that as perfectly legitimate. It creates an issue only in situations where my market share is pretty high and I’m using it to exclude rivals. Well, what that suggests is you really have to go product by product. And for the vast majority of the individual products that Amazon sells, there’s nothing approaching market dominance.

I think the only one that I know of that gets above 50% is eBooks, assuming it’s in market. And for most of these others, it’s much smaller and you just can’t aggregate a product specific vertical practice like this and say, well, we’re doing it for a whole bunch of products. And as a result, even though it’s not anti-competitive, when you look at each product individually. And to answer your question, I don’t know of another case that is permitted aggregation outside of the merger context.

Joseph V. Coniglio: But even on the MFN issue, I think the court said they’ve never seen a policy like this. Depends on how you read it, but it seems like the court, however you want to characterize Amazon’s policy, there hasn’t been any precedent to find it any competitive. But looking at the next conduct, the alleged coercion of fulfillment with Prime, and Steve, I’m turning to you here. What is the FTCs basic theory? Is it let’s say kosher within modern exclusionary conduct analysis, or are we looking at a tying or bundling claim here? And if so, what are exactly the two separate products that are being joined?

Steven C. Salop: Okay, well, let me just make one quick remark about the MFNs. Something like Delta Dental, which it didn’t go to trial but passed the motion to dismiss it changed the law, and that was a different price for every dentist. Yet the court was willing to uphold that on the motion to dismiss. I mean, sometimes aggregates do work, and I think in my discussion of the FBA coercion claim, I can show that as well. So an answer to your specific question, Joe, is yeah, I think conventional antitrust does reach this claim. I mean, the claims got several moving pieces, but when you break it down, it’s a fairly conventional raising rivals case, though there is one refinement that I want to stress at the end. So let me explain what I mean.

The point is that merchants need to use FBA to qualify for Prime eligibility and the Buy Box and they all comply because of the large customer base involved. And so it amounts to a defacto requirement that you use FBA, even if it’s not a literal contractual form. So maybe an answer to the point Herb’s stressing, we would want to focus on the merchants that use FBA. It would only be the markets where the merchants actually use FBA where there’s harm. But for those merchants, the FBA requirement raises their costs and it raises the prices that they charge in selling to rival platforms in three ways. And because it raises the prices on the other platforms, then it raises the prices charged to consumers on Amazon as well.

So let me go through these three ways. So first, the requirement to use FBA raises the merchant’s costs of selling through other online retail platforms, and that increases Amazon’s power to raise online prices to consumers for all those products. Second, the requirement to use FBA raises the costs of rival fulfillment and delivery platforms by reducing their scale. And that will need to be shown as a matter of evidence. And if so, that raises the cost of selling through rival platforms as well. And actually it raises the cost of selling through rival platforms, both for Amazon merchants, but also for non-Amazon merchants who also are then selling through fulfillment people that have lower scale so that even the people that don’t use FBA are getting harmed.

Third, the high fees associated with FBA and the other provisions such as pay-to-play requirements that I mentioned earlier, those high fees raise the cost of the Amazon merchants in selling on Amazon, but in conjunction with the MFN price parity provisions that Kathleen stressed the high fees raise the retail prices that need to be charged on other sites. So there’s all the ways in which it raises rival’s costs and raises prices on rival sites as a result.

Now, Amazon’s power, this is where the refinement comes in. Amazon uses the high fees as the mechanism for collecting its monopoly profits on all the products. That is Amazon can set its FBA fees and the other fees such that the merchants charge the monopoly retail price. And then Amazon’s monopoly profits the difference between the monopoly retail price and the merchant underlying costs. So that’s how they collect their monopoly profits. So it all fits together in a very nice, fairly conventional package.

Joseph V. Coniglio: Well, thanks for that, Steve. I think I want to discuss that a bit further, and I’m looking at Kathleen to start. So obviously Kathleen, you’ve been an enforcer, you’ve worked in private practice just reading the complaint and thinking about these theories, reading what the FTC said about how sellers prefer to have a loan fulfillment provider, and then sort of in turn having more fulfillment providers will enable rival online superstores and marketplace services providers to come up and compete with Amazon. It seems a bit attenuated to me, but what do you think?

Kathleen W. Bradish: Well, I think one of the things we haven’t really talked about in this complaint is that there appear to be sellers and sellers in particular who are willing to say that these practices have an effect on them, that they’re afraid of the anti-discounting policies and that affects their behavior, that they won’t go with another fulfillment company because it doesn’t make any sense for them to have different fulfillment companies given the scale that Amazon has and that they would prefer in some instances to go with a different fulfillment company. So from my perspective, even though, and I realize that the fulfillment part of it, people love their Prime. I love my Prime and love the delivery services in part because they don’t know what other options might’ve happened without Amazon’s policies. But to have these potential witnesses who are cited in the complaint, I think that may make all the difference if they can really get these folks on the stand, if they can get them to testify to the ways in which it has changed their behavior.

Joseph V. Coniglio: So Herb, I want to follow up with you on that. It seems, hearing Steve and Kathleen reading the complaint that a lot of the FTCs theory here depends upon Amazon having some market power in fulfillment. What the complaint discusses doesn’t, I guess formally define as the online fulfillment services market. Does Amazon have any market power in this regard?

Herbert Hovenkamp: Well, I’m not sure it’s a fact question. I think it conceptually it holds together better than the online superstore market simply because the services are more common across a wider variety of practices. The hard part is going to be for the FTC to convince the fact finder that that’s a grouping of sales that’s capable of sustaining a monopoly markup and it may be able to do that. If you look at the services individually, most of them look like pretty routine billing, packaging, maintenance services that anybody can supply. In the aggregate and given their scale, it may be the case that a very large provider of those services has a cost advantage over a smaller provider, but that’s going to depend on fact findings.

Joseph V. Coniglio: Thanks, Herb. Bilal wanted to make sure you had a chance to quickly weigh in here before we move on to the Section 5 issues, but Kathleen mentioned the benefits of Amazon Prime and you worked at the FTCs policy planning office. I think a lot of consumers like their Prime, and the FTCs case really seems to sort of strike at the court of the value Amazon is offering there. Do you think this is the sort of case that in the policy planning office they’d say, yes, this is exactly what consumers want, they want Amazon’s Prime to go away or change?

Bilal Sayyed, TechFreedom: Well, Joe, I think we’d be careful about, and we should all be careful about assuming what a diverse group of hundreds of millions, I don’t know of customers or individuals want. Some people probably would like an unbundled Prime, some people probably like a bundled Prime. I was thinking with respect to the fulfillment question that it is a core, I think, reputational issue for Amazon, that products that are offered, purchased off of Amazon will be delivered to you quickly, safely in one piece. And that I think is a very powerful reason for them to want to their third party sellers to use their service. So this strikes me as an area where to return it to your question on would policy planning support something like this or a policy office supports something like this, I think we would think very hard about the benefits associated with the service with respect to Amazon’s reputation as a provider of goods quickly, safely in one piece.

And this idea that parties should be able to take advantage of the platform and the Amazon brand, but then give the merchandise to a third party that may or may not particularly does not meet the same delivery fulfillment and delivery standards really strikes me as a very dangerous ground for the agency to hang a case on because there seems to be a real reputational concern there by Amazon, basically free riding off of Amazon’s brand and then look for alternative cheaper price, maybe lower quality service. So there is a clear, to my mind, clear competitive justification for that, and I’d expect that to carry the day with the court.

Joseph V. Coniglio: Well, let me have you keep your... Steve, I want to move on to Section 5 here if you don’t mind, but just to keep your FTC hat on here, the FTCs obviously brought to Section 5 standalone claims and Bilal, why don’t you to get your take on how are these going to be received by the courts?

Bilal Sayyed, TechFreedom: That’s a very good question. I think there will be skepticism of standalone Section 5 cases sort of divorced from Section 2 standards. It’s been a long time since the FTC litigated a standalone Section 5 case or claim. I think the last time was in the WIC case, formula case back in the early 90s. Antitrust has moved so much. Here’s the real failure there. I think we all understand the reason this case was brought in the district courts rather than the administrative system. And in many respects, I’m supportive of that, in many cases, cases that are not so novel. This is a novel case, and the administrative system at the FTC or the FTC in a sense was designed to do novel cases and they’re bypassing that.

And again, procedurally I think it’s bad, but I think it’s going to hurt them because they don’t get the chance to write the first, actually any opinion here. They’re going to be relying on district court, appellate court, maybe even the Supreme Court without articulating themselves a framework why Section 5 should apply here, even if Section 2 does not. Again, I think we all know why they brought it in district court, but it strikes me as a real serious mistake and procedurally suspect.

Joseph V. Coniglio: Yeah, Herb, I wanted to follow up with you on that. I mean, even in the glory days of the FTC, so to speak, when they were bringing these types of cases every once in a while, how did they typically fare in court and-

Herbert Hovenkamp: Not as well as they should. This is one area where I think the FTC could make some positive law, but I just want to step back a second. First of all, why no Section 1 claim in this case? I don’t understand that several of the practices are being carried out by means of licensing agreements or other contracts. Seems to be pretty clear that the agreement requirement can be met and that the liability standards are more aggressive. And typically when we think of freestanding Section 5 claims, we think of claims that expand beyond Section 1 of the Sherman Act. For example, exclusive dealing under FTC versus Brown Shoe where the Sherman Act or Clayton Act requirements couldn’t be met. Or the most vexing set of problems is proof of consciously parallel or oligopolistic behavior in situations where the agreement requirement of Section 1 of the Sherman Act cannot be met.

And one of the elements of this complaint is that what Amazon is doing is acting as an oligopoly facilitator. And I think that is a useful place to get a little bit more expansion in the law because it basically is arguing that Amazon is able to get these prices to go up in tandem without there being an agreement that would meet Section 1 standards. And that I think is due to systematic under enforcement by the courts in cases like DuPont Ethel in the 1980s that basically said, if you want to go after parallel conduct in the absence of an express agreement, you have to meet a very high burden and they found it not to be met.

Joseph V. Coniglio: No, thanks for that, Herb. I wanted to turn to the Ethel case, but I mean I think similar, again, hitting this point of Amazon as a monopolist for some parts of the client complaint, but then again, it’s another parts, it’s using oligopoly type behavior to harm consumers. It’s an interesting contrast. But Steve, could you maybe talk a bit about sort of, I guess what we would call this non express unilateral collusive type conduct that seems to be an issue with Project Nessie? What’s your take on that?

Steven C. Salop: Okay. Project Nessie is really unclear because the entire section is redacted, but I did a bunch of searching in preparation for our panel and what it seems like is that Project Nessie is a combination of Amazon charging a high price, but having a defacto meeting competition provision so that if it’s higher prices and followed it reverts its price back to the level that the rival platform is charging. So it sounds like Amazon’s setting a high price and telling rival platforms or what you might say, telling rival algorithms of platforms, if you don’t match this price, I’m going to cut mine tit-for-tat. And that’s what they were experimenting with, whether their algorithm could communicate with other algorithms and teach them to charge high prices.

Joseph V. Coniglio: Well thanks, Steve. I mean, I think, again, it’s interesting to see the facts that come out Amazon I think is yet to file any sort of response. But Kathleen, assuming what we know, her mentioned the DuPont case asking you to put on your sort of practitioner and enforcer had here, what are the chances the FTC is going to meet that standard? I think it has an intent requirement. Do you think they can actually win on this?

Kathleen W. Bradish: Well, are you asking whether they could win if there were a Section 1 [inaudible 00:48:06]-

Joseph V. Coniglio: Section 5, on the Section 5 claim they brought? Yeah, given the DuPont case and other precedent?

Kathleen W. Bradish: Yeah. Well, I think it works in conjunction with the Section 2 claims, so I’m not sure they have to win separately on that. So on that point, I think that’s maybe a moot question to a certain extent. There is always the appeal to a judge of, especially if there is some novelty to the case of not creating the potential for a lot of class actions, a lot of private actions, which is always the advantage of a Section 5 decision, at least from the perspective of a judge who may not want to open floodgates, for example. So I think there is still some appeal from that perspective if a judge is feeling nervous about going someplace new.

Joseph V. Coniglio: Okay. Well, I think we’ve got 10 minutes left here and I wanted to open it up and also turn to some audience questions. So one of the questions we’ve got, and I think it is relevant to sort of the bigger picture here, turns to the broader neo-Brandeisian movement. And Bilal, I want to start with you something you tweeted this morning about this case being the FTCs Afghanistan. Obviously there’s a lot of Middle East policy going on right now, but it’s an interesting metaphor. So the neo-Brandeisian movement really started with Lina Khan’s, now Chair Khan’s article about Amazon. Does it ultimately stand and fall with this case? What do you think?

Bilal Sayyed, TechFreedom: Well, look, I don’t think it should stand and fall with this case. I’m not a supporter or supportive of sort of this neo-Brandeisian view, but incremental change with respect to enforcement is probably a positive thing. The Obama administration, I think at the end of its tenure was thinking maybe a little differently about some cases. The Trump administration continued that. I mean, it’s no secret that the FTC was investigating Amazon in the Trump administration. The reason I suggested this Amazon case would be called the FTCs Afghanistan, a bit of a play on the FTCs Exxon or attempt to break up the oil companies in the 70s being called the FTCs Vietnam.

It is just too big. If you want to make progress, and progress should be made, I think you don’t take on such a large complicated matter. And so this case should outlast the tenure of any of the commissioners at the FTC that can lead to changes in the way it’s approached in the courts, which could be positive, could be negative, but it is just hard to see how this turns out as a positive model for neo-Brandeisian approach. But it could have if it was more narrowly drawn, and I think if they wanted to extend Section 5 or rely on Section 5 brought through the administrative process, could have resulted in maybe small, modest, meaningful changes in the law or even clarity whether or not Amazon’s practices were illegal or not. So it’s too much, too fast, and honestly it’s just not strong enough I think. So I don’t think that neo-Brandeisian will fall or rise on this, but it just strikes me as a mistake the whole way they approached this case.

Joseph V. Coniglio: Okay. I want to make sure everybody has a chance to weigh in on this one. Steve, I see you smiling. Go ahead. Take us off.

Steven C. Salop: Yeah, I couldn’t disagree more. First of all, the case, this is not a neo-Brandeisian case at all. It’s a straightforward raising rivals cost case in the context of a two-sided market. MFNs concerns about MFNs are as old as the hills. It goes back to the Ethel case. Then the other cases that were brought in the 80s and 90s. The FBA is pretty simple, raising rival’s costs. So I don’t really see... This case is unrelated to Lina Khan’s Law Journal note, so I just don’t see that you’re saying it is so novel. It’s also not so broad. They could have been clearer about it in the complaint, but they’re not really attacking Prime. They’re viewing Prime as evidence of market power on the online marketplace, not attacking the provisions of Prime.

But they’re a little sloppy there in a paragraph about dark patterns, and they slip in how Prime is tying. They could have left that out. That’s not what the case is about. Nor for that matter, are they attacking Prime. They’re not attacking Prime at all. What they’re attacking is fulfilled by Amazon. And I think Bilal got to that as the heart of it earlier. The FTC says that you don’t need FBA in order to enforce the quick delivery guarantee. That’s a legitimate guarantee by Amazon, and that’s going to be their primary argument for FBA. So the FTC says, well, no, there was the Seller Fulfilled Prime program SFP, and it worked. The sellers did meet, Amazon was able to monitor, and sellers fulfilled the quick delivery guarantee. But Amazon dropped SFP because it feared loss of monopoly power. So that’s what the FTCs got to prove. But if they can prove that, then they can win. And so I don’t see it either as neo-Brandeisian or as too big.

Joseph V. Coniglio: Thank you, Steve. It looks like post-Chicago is alive and well. We don’t need neo-Brandeisian. Herb, what do you think?

Herbert Hovenkamp: I’m kind of in between Bilal and Steve. On the one hand, it’s not a neo-Brandeisian complaint for the simple reason that the FTC knows they couldn’t win. So it’s an attempt to fashion a complaint in ways that will get it passed the motion to dismiss under existing antitrust standards. Did they succeed in doing that? Well, I’m doubtful. I think they’re really pushing the boundaries of how you measure market power more than they’re going to be able to get away with. But I’m willing to give them the benefit of the doubt at this juncture.

Joseph V. Coniglio: Kathleen?

Kathleen W. Bradish: Well, I would take issue with the one point that this is too big, and I think the importance of it being big is that the FTC has to challenge these big issues because otherwise they become immune to antitrust challenge. So for them to not attack something because of the size, the diversity of products, or the number of behaviors, et cetera, I think would be to avoid applying the antitrust laws to certain areas of the economy, which doesn’t make, I think is a real problem. And I would say that in terms of likely success, there’s a lot of stuff in this complaint that could be very interesting and very convincing to a judge if it manifests. Like the story Steve is pointed to of a period of time when there was no requirement that sellers had to use Amazon’s logistics under Prime. And if the FTC can put that case forward, can bring the witnesses, can show that there were alternatives, then I think they stand a chance.

Joseph V. Coniglio: Okay. Well, thank you Kathleen. Herb, would you mind muting please everybody? Just apologies for the technical difficulties again, but I see we’re right up against the hour here, so I think that’s going to have to be the conclusion of our panel. But I wanted to thank everyone again for joining us and for really this very August group of panelists here to discuss this very timely issue. I think that this will certainly, it’s just the beginning of a long conversation, so thank you all very much.

Steven C. Salop: Thank you.


Kathleen W.
Kathleen W. Bradish
Acting President
American Antitrust Institute
Joseph V.
Joseph V. Coniglio@JosephVConiglio
Director, Antitrust and Innovation Policy
Information Technology and Innovation Foundation
Herbert Hovenkamp@Sherman1890
James G. Dinan University Professor
University of Pennsylvania Carey Law School
Steven C.
Steven C. Salop@stevesalop
Professor Emeritus
Georgetown University Law Center
Bilal Sayyed@BilalKSayyed
Senior Competition Counsel; Counsel
TechFreedom; Cadwalader
Back to Top