Google Ad Tech Antitrust Verdict Analysis | Legal Forum

[00:00:00][Music] Hello everyone and welcome to this Federalist Society virtual event. My name is Edith Herald and I'm an assistant director of practice groups with the Federalist Society. Today we're excited to host this Fedsock forum called at it again a second Google Antitrust Verdict speaking with us today. We're very pleased to have Professor Rebecca Ha Allenssworth Bal Seed and Joel Fyer and we're thankful to Ashish Agarwal for moderating this program today. Ashish is a consultant for the American Edge Project and the US Chamber of Commerce. If you'd like to learn more about today's moderator or speakers, their full bios can be viewed on our website, fedsock.org. And throughout the program, we may turn to the audience for questions. So, if you have a question, you could enter it into the Q&A function at the bottom of your Zoom window, and we'll do our best to answer as many as we can. And finally, I'll note that as always, all expressions of opinion today are those of our guest speakers and not the Federalist Society. With that, Ashish, thanks so much for joining us today, and I'll hand things over to you.

[00:01:03]Well, thank you, Edie, and thank you to the Federal Society and the corporation's antitrust practice group and our chair, uh, Swatlana Gans, for helping us put all of this together. Um, two weeks ago, a federal court found that Google violated section two of the Sherman Act by willfully acquiring and maintaining monopoly power in the open web display publisher ad server market and the open web display ad exchange market. Though the court ruled in Google's favor in determining that there is not a relevant market for open web display advertising ad networks, the case is now moving into a remedies phase and the court's lengthy opinion could accept could affect several other pending tech cases and antitrust law generally. Today we have an outstanding panel to discuss these issues. Professor um Allenssworth is a law professor at Vanderbilt where she studies antitrust, professional licensing and evolving attitudes towards tech platforms. Bal say is in private practice at CAD Waller and adjunct professor at the Scalia Law School and the host of a very popular podcast rethinking antitrust. In President Trump's first term, Bal led the FTC's office of policy planning. And finally we have Joel theer who is the president of the digital progress institute and the founder of his own law firm the P LLC and Joel is an alumnest of both the FTC and the FCC and a famous very famous person in antirust law uh in recent weeks. Sadly um Adam uh Kvakovich uh with the Chamber of Progress had to cancel at the last minute and in light of that we're going to implement presidential debate moderator rules. So, I might press our panelists a little more than I might otherwise. All right, with that, let's get to it. And Rebecca, let's start with you. Uh, for those who haven't been tracking this case very closely, um, could you give us an overview of the complaint and the court's holding and maybe touch upon why this case has gotten so much less attention seemingly than the Google search case? Yeah, I'm I'm happy to answer those questions. So, um, this case, uh, was filed by the DOJ, um, alleging that Google monopolized the ad tech market. Now, I think skipping ahead to one of your last questions, part of why this has gotten less attention is because this market is opaque to us, you know, you could think about it like the um, energy market or the electricity market or maybe water. you know, it's like the internal plumbing or electric connections that take on the one hand um an advertisement that somebody wants to put on the web and puts it through the market through various exchanges and various processes um to a website that wants to host it. So, um it's just very complicated and unfamiliar to us. And so, we're going to try to break it down.

[00:03:41]and I'll try to break it down as simply as I can. But part of why it hasn't gotten as much attention really is this this sort of unfamiliarity and complexity with the idea. Contrast that to the search case which is very familiar to us because we use search as a product all the time. It's just such a huge part of our lives. Now, uh digital advertising actually is and if we don't we may not realize it but it actually is a huge part of our lives because it is the way that most websites are able to offer free content. So, um, you know, every time you go to look up a recipe or you go to look up an article, you're going to see you're scrolling through and you're going to see these ads pop up. And sometimes they take a second.

[00:04:18]You'll notice there's like a blank space as you're reading. And then one second later, it'll be like a car ad. Okay? So, that in that half a second, in that fraction of a second, a whole bunch of transactions are happening behind the scenes within what we call the adte stack. And it essentially looks like this. you have a um buyer side. Uh that would be the advertisers. So here we're going to put you know Chrysler J Crew.

[00:04:46]Okay, they have a campaign they want to push out their ads to the open web. Um that's the buyer side. Then in the middle you have the exchange. You could think about this like the New York Stock Exchange where uh buyers of stocks and sellers of stocks come together to actually conduct the transaction. And then the other side is you have the the um publishers. So that would be the New York Times website or the recipes.com website that's selling that blank space that has has my eyeballs in on it for a split second. Um the DOJ claimed that Google monopolized three different parts of this stack. They said and now I have to see this is so complicated I have to consult my notes. Handily the decision comes with a nice graphic that will help us. So on the ad side, on the advertiser side, Chrysler, J Crew, um they claimed that the advertiser facing uh ad network was a market and that Google monopolized it. Um they also claimed uh on the publisher side that the ad server uh product, which was where advertisers went to um you know describe and offer their their empty spots. Uh and this is you may be familiar with the company doubleclick uh this was um the double click for publishers is the is Google's product in this space they said they monopolize that market and then finally they monopolize the thing in between which is the ad exchange which in the case of Google is a product called adx so the idea here was simply put Google had this market power because of search and that was everybody wanted to advertise on Google search. Basically, if you were Chrysler, you were J Crew, you wanted to pop up in those first few answer, a few first few answers to the query on Google search. And so, Google had this tremendous advertiser demand and all these relationships with advertisers that it could use to make a product called Adwords. This is the this is the first step in the stack, which is an ad uh uh network um and have a lot of market power there. Then the allegation was that it used those that that product and that access to all the advertiser demand. It leveraged that power into the publisher side. It basically said to uh the New York Times, the publisher, look, if you want access to all these advertisers who are willing to pay a ton of money, I have them all um on on um Adwords, and so I can give you access to this. and so um monopolized that market for publisher ad servers. They alleged that the acquisition of doubleclick which happened in the I can't remember if it was the late um as or early maybe it was like 2010 or something was a part of that was earlier than that I think anyway was a part of this monopolization. So the acquisition of of um double click was a part of that. And then finally Google sort of discriminated against certain kinds of buyers and impose certain imposed a tie between their double click for publishers publisher side thing and use of their exchange in order to monopolize the exchange. So essentially they said um you know first they said if you want access to all these advertisers you're going to have to use our um publisher side ad server and and you're going to have to use our exchange tying essentially the exchange to the ad server um and those are the allegations and uh uh the verdict decided that there was a market. So of course in monopolization you have to h have a well- definfined market of which over which you have monopoly power. The court rejected one of the three market definitions offered by the government. The government said that ad networks are a market and the judge said no. So essentially rejected that advertiser side monopolization claim but did accept the government's market definition for ad exchange being something um of its own market in which uh uh Google had market power and the idea that this publisher ad server market is its own market and that Google had market power. Now I don't want to get into the details of of how the government made those arguments unless you'd like me to do that now but I'm trying to give a sort of a brief overview. So it said there is no market for ad networks. There is a market for the exchange and there is a market for the publisher ad server which is basically double click for publishers and there was monopolization based on exclusionary conduct. First the tie between the ad exchange and the ads the publisher ad server. So, by telling the New York Times and recipes.com, you basically have to use our ad exchange that tie was unlawful uses the Jefferson Parish test for that. And it says that certain contractual terms that Google put in its ad server doubleclick terms saying that, you know, you have to really privilege um our ad exchange uh uh called first look and last look. um that harmed the uh the competition in the exchanges and ultimately harmed the publisher ad server uh market were also bad acts. So that those were exclusionary acts and um that is an overview of the case. I really welcome other panelists to help me sharpen this because it is very complicated but that's that's my understanding of the case. Well really appreciate that excellent summary just a few minutes of a very complicated case. So, Bal, let's let's dive in a little bit and I wanted to ask you in particular about one of Google's primary offenses which was um to raise the American Express case from the Supreme Court um and to you know Google sort of makes this presents this view that gosh this is a two-sided market and you advertisers are just looking for the most bang for their buck you know across the internet um and this market should be interpreted very broadly. Um the court took you know much narrower view of of AMX. I'm wondering if you can share you know your um you know opinion of how the court analyzed uh that case. All right. Thank thanks Ashish. Um so I think I think this is an area where the district court's decision is is vulnerable. Right. I think um I'm not I'm not saying they got it wrong. At least I'm not saying that yet. But I think they're vulnerable on this question. Now they they clearly I think this is right. They clearly define the ad exchange um market as a two-sided market, right?

[00:11:41]Uh but they then identified, you know, separate markets for uh I'll call sellers and buyers on either side of that ad exchange platform.

[00:11:56]and and that seems not only counterintuitive um but wrong in in the analysis, right?

[00:12:03]I think what you would take from AMX is that if you've identified a two-sided market, right, ad exchange is a platform. The the district court recognized it as a two-sided market. I think then in evaluating the competitive effects of conduct on one side or the other of the ad exchange which would be the publisher ad server and the ad network. Um you you you need to take account of the effects on both sides from conduct on one side. And so this I think ties into Google's um effort to say this is an ecosystem and should have been evaluated um um in a sense collectively. The conduct should have been evaluated collectively.

[00:12:56]So, I think this is an area where the court in recognizing the exchange as a as a two-sided market, two-sided platform, um, which is correct, then made a mistake in not evaluating the the competitive effects of the restrictions that um, DOJ alleged Google had adopted for either the publisher side or the advertising uh side. Um so I I think it muddled up I think the court district court here muddled up the teaching of ammex right. Um and it's not that you can't have separate markets on the side on the other side of a plat on each side of the platform, right? Or that they can't be their own uh relevant markets. But when you're thinking about the competitive effects of conduct on one side or the other of a two-sided uh let's say two-sided platform um you don't get to ignore the effect on you know one side to just evaluate the um harm on on the other and that's what tripped up DOJ in in in MX or or the states right both states and DOJ in AMX they looked that the effects on only one side of of you know the network so to speak. So it's um I I think the decision is vulnerable here. Joel would love your thoughts on uh that annex analysis and and as you're answering if you don't mind would love to ask you about another important precedent that the court also discussed which was Trinko and the duty to deal.

[00:14:46]you know, the the core, you know, basically agreed that Google can do business with whom it chooses, but found that Google had sacrificed private profits to make life a little more difficult for competitors. Um, so if can you talk about your views of whether the court got it right or wrong in terms of its MX and FCO analysis? Happy to and for the record, I take the full endorsement of it being a very famous person for the past few weeks. So, thank you, Ashish. Um my my initial reaction to uh Bal's comments and also even uh reading our first impression of reading the the case with with respect to annex is that the I think that the analysis only makes sense if you're evaluating it through the tie and what and what the actual effects are. So it seems as seems to me uh that the court was really interested in defining very key markets as uh Rebecca very nicely outlined where the publisher side of things what happened to be a one-sided market and the question was does does their monopolization of that one-sided market force them into uh force the uh ecosystem into their ad exchange and so I don't think that the court was actually trying to clarify anything in the annex. That that's at least my read.

[00:16:08]It it I don't think they actually need to even go go the MX route if you're just trying to demonstrate that there's a section one or two violation with respect to the tie. So I also will uh admit that I am a lot of others are very confused by the AMX decision and you know how and I don't know if there's a clear path in in discerning you know when the court is going to make certain uh characterizations as a two-sided market. I think that's still a hotly contested question. Uh we saw that play out uh very publicly in the Epic the Apple case where there was this push and pull from the Ninth Circuit where uh it wasn't really clear to the court that you needed equal effects on both sides of the market in order to have an antitrust harm. Only a mere effect might might do uh be something to get you over the line. But again, I think this is we're we're seeing this get litigated in real time almost and especially now we're having these uh we're having these court cases that deal with very complicated and vert vertically integrated markets like the ad markets or like the uh digital ad market uh the app store markets all of these are I think uh that is my read of the ad tech case in particular just to sum up is that they were trying to not necessarily sidestep but just say look we considered it we could we looked at the annex decision we don't think we actually have to go there because we actually we think that the BOJ has met their burden by virtue of the tie with respect to its one-sided market. We don't need to evaluate further on AMX. That was my personal read and uh and I think the law is uh is correct that you know there there will be a lot of debate and discussion around that particular uh strategy and uh you know I but I I do think that it makes some sense if you're looking at it with within the context of the tie on uh but on to the question of Trinko. I mean uh I I uh so I think the defense uh bar at the antitrust uh in the antitrust world is uh forever the Gretchen Weiner of uh uh and where they're always trying to make fetch happen and it seems like there is a natural pension. Look, if I'm on the defense side too, I want everything to be a duty to deal case. I mean the facts are good if like if you can get the court there. Uh but the problem I think uh that occurred in this case is that there were just too many distinctions.

[00:18:39]And so just to be clear uh Trinko uh was very much considered. I mean there's a whole healthy portion of well not I would say a significant portion of the of the decision related back to Trinko to outline the specific distinctions between this instance and what happened in Trinko. And so uh just so everyone is aware the the over the underarching uh ruling in Trinko is that the possession of monopoly power can be will be found will not be found unlawful unless it's accompanied by an element of an anti-competitive conduct. So the question is what is that anti-competitive conduct? Even when uh Justice Scalia was thinking through uh the Trinko case, he evaluated not just simply what was uh what was the specific antitrust statutes alleged, but also the GI the ecosystem at large. These were highly regulated markets. We were talking about Verizon and uh Ashish, I know you and I go back on the to the telecom days where you have like versus Selex and like and uh the regulatory kufflele that that occurs through there.

[00:19:45]And so Scalia being a telecom lawyer himself was evaluating that and saying look you already have the telecom act that essentially requires these forced integrations and forced uh and basically forced dealings. You don't need an extra ump from like the uh from the antitrust side to further encourage this quote unquote duty to deal. And so I think that really did uh bite uh Google in the butt a little bit on this issue because one you can make a clear distinction there and the court actually goes out of its way to say that that one we're talking about an a virtually unregulated entity. Uh I whether you agree with that characterization is discussion for another day but that's how the court initially made its first distin distinction. The other distinction is that a the duty of the deal is separate and apart from a tying arrangement. So if you're looking at it from the context of a tie, then you get you don't uh you don't necessarily need to go to the duty the deal analysis at all unless you want to say that all tie ties are d are deacto dutyto cases which I think there's a lot of problems with that analysis. I mean for to start uh you know the Clayton Act uh identifies tying as an independent uh antitrust harm. So I just don't think you can get there. I think also there was just factual evidence that worked very much against uh Google in this case and and even if you wanted to take the Trinko case directly and say look uh it even if you uh don't want to directly categorize as a duty to deal that's fine. Uh but here's the other important also uh raised the issue of even within the note I guess you can call them exceptions but generally the outlines that Scalia had he listed three specific things that would get you out of harm's way and do a deal specifically was something offered to the public uh was something that was originally offered to the public not offered to a competitor might uh might be something that you would evaluate this may have happened here the the distinction there with with respect to Verizon was that the duty to bill in the Verizon context uh existed and I I'm quoting here from uh from Jessica Scalia only existed within the bowels of Verizon. It was never actually offered to the public in the in the first place.

[00:22:03]That I think was very much distinguished with with uh with Google because it was based uh most of the things that we're talking about here were based off of acquisitions that happened after meaning that the public did have access to this and there were certain things that they could avail themselves to pre- merger or acquisition and now they're not able to.

[00:22:20]I think that also gets you a little bit that gets you a little further out of the duty to deal uh type of arrangement.

[00:22:26]And I think that you're seeing courts get very skeptical of the the duty to deal claims. I mean, I think we also saw that in live action uh or live nation, excuse me, uh where there was a similar type of argument going back and forth with respect to duty to deal and the courts saying unequivocally, no, you this isn't a get out of jail free card.

[00:22:45]You're you're going to actually have to demonstrate why this is this is more akin to Trinko. And I think the facts in Trinko make it very hard for these tech companies in particular to try to put them in wed uh square that circle. So those are my two cents. Uh but I I think that the court really did a fairly decent job of explaining why we've why Trinko just isn't there. I think it really was to embolden the uh the tie uh aspects of the case. So hope that's helpful. Hope it wasn't muddled, but uh happy to to respond.

[00:23:19]very helpful. Um, Rebecca would love your, you know, thoughts on you the MX analysis, the Trinkco analysis, and if that wasn't enough, let me introduce, you know, another key part of the course holding which you and Joel touched upon, which was the course merger analysis.

[00:23:31]You know, the core, as you said, did did find that there was not an anti-competitive problem when Google purchased um, you know, double click or adtech or um, the other company. um in in so finding it relied heavily on the fact that DOJ had reviewed and not raised objection to those acquisitions at the time. So I'm just wondering what broader implications that part of the court's holding might have, you know, for other pending cases, you know, including, you know, the the meta case.

[00:23:58]So there's a lot there. So Rebecca Pikachu, what you'd like to talk about?

[00:24:01]Yeah, so I mean I have a lot a lot there's a lot to say about it. So American Express cannot mean what it says. It doesn't make any sense. And this poses a problem whether you are a free marketer and you know you consider yourself a libertarian, whether you're um right-wing, whether you're leftwing, whether you're neandeisian, American Express is a problem. And that's because uh it it has this idea that you know the transaction platform you have to consider both sides of the market. So if if the reason why we define markets is for one of two reasons at least in a monopolization case. The first is to figure out whether or not this company has power. Are they able to essentially you know raise price profitably on their customers and whether that's price or whether it's you know lower quality or whatever. It doesn't matter but the idea is you're able to do something bad to consumers without the consequences that we associate with competition without people being able to switch away.

[00:25:00]So under this definition under this use of market definition doesn't make any sense to think about the middle transaction where it's happening because the question is as a merchant um does that company have power does American Express or in this case does Google have power to make terms worse for you leaving you with nothing better nothing better to do now the insight from the early 2000s from economists and two-sided markets that gets distorted in AMX is that that power to raise price on merchants or to raise price uh let's say on the the publishers so the New York Times or whatever to make poor terms for them to be selling their advertising uh that that power is influenced in a two-sided market by the other side. Now that's because you know um you can't raise price too high. You can't screw over one side too badly if it means it unravels your your interest from the other side. So this economic insight that's that's that's really meaningful, really important got distorted into something incorrect, which is there's no such thing as a market on one side. What the economists were trying to say is power in that market on one side needs to account for what happens over here in this other market as well. That's not the same thing as saying there is no separate markets and there's no separate market definition. You absolutely can have power on one side of your two-sided market in a way that harms consumers and meets all of the statutory and case law definitions of how we have been thinking about market power forever. Okay, so that's one way in which American Express went wrong. And by the way, I could we could I'm going to try to keep my comments on American Express brief because they're they're multitude. The other reason why you define a market in a monopolization case is to locate the harm. Is to say here here's where the effects are. If MX is to be read to mean that you can take effects on one side of the market and balance them against effects from another, then this runs a foul of something we have had for essentially the entire history of the Sherman Act, which is you are not allowed to use effects in one market to des to to offset effects in another.

[00:27:09]This is this is true in the um 60s in a merger case u the bank merger case um Philadelphia National Bank. You can't say, "Look, we kind of screwed over the consumers in Philadelphia, but it was really because that made us a bigger player in New York." Now, you can see how left-wing people or right-wing people would be offended by this, right?

[00:27:29]Well, you know, the right the the the uh right-wing people may be offended by arguments that say, "Well, yeah, we kind of screwed over buyers of cars. We really improved the environment." Or, you know, yeah, we So, you can't you cannot justify harming one market by harming another market. Okay. So given that American Express not only is like sort of contravenes all of antitrust policy but also just internally incoherent it's very unclear what it means going forward. And I would posit that what it means is what lower courts say it means. And so when we read this opinion we are getting a sense of what AMX is going to be in practice. And I think what the lower court did here makes a lot of sense. First she says they can't we cannot treat this as a single market because the one side is not a substitute for the other. That is true. That is what AMX should have said.

[00:28:23]The problem and I will say that the law is correct. This is vulnerable, right?

[00:28:28]The problem is that's not what AMX says.

[00:28:30]AMX doesn't say that you're supposed to figure out whether the right side and the left side are substitutes. That would be the correct way of thinking about whether or not it's a market, but it's not what AMX does. So, will that get reversed? Possibly. But I think that that's the correct thing to say and I think that she handles it well in that sense. The other thing she says is that this is not a transaction platform. Again, a little bit unclear what that means given that transactions are happening between advertising sellers and advertising buyers at the exchange level. How is that different?

[00:29:00]Maybe it's a maybe what American Express meant or what we can now read it as is limited to payment systems where like there's literally just a like you're just paying. Um that's what a transaction is. So for us to have a workable antitrust going forward, we are either going to have to write around AMX or we're going to have to reverse AMX.

[00:29:20]AMX cannot stand for its most extreme and and most literal reading. And I think that that this opinion needs to be read in light of that. Okay, Trinko is a little bit different. Trinko is different because Trinko is a case that internally is coherent. Okay, you may love it or hate it. It doesn't have these kinds of problems, but it has a ton of dicta. It has all this stuff in there about we have to give people monopoly prizes otherwise we disincentivize innovation. you can't be asked to reach into the bowels to, you know, and and so what will we do with this precedent that has contributed to um uh uh sort of turn an antitrust law away from liability, especially away from liability for tech platforms that's now unpopular. Well, one thing we could do is we could start cutting away at some of the dicta and putting Trinko in its place by including only the things that are essential to the holding. And I think that that's what the judge does here. Now, is this vulnerable to reversal? Of course, because if you get a court that likes the dicta and Trinko, then maybe they'll say this runs a foul of Trinko. But what the court does here is it says that Trinko is not a license to just always deal, you know, you can you can refuse a deal with anybody you'd like. Trinko distinguishes Aspen skiing in this way that can't be distinguished here. So the court in in the adtech case kind of embraces Aspen skiing, puts puts this case more in the Aspen skiing bucket than the Trinko bucket and in particular distinguishes Trinko as being in a highly regulated industry where the refusal to deal was actually required by statute. Okay? And so there's a question there about like reaching into the bowels, right? This wasn't available on the market. This was a artificial act of of Congress. And the other is the profit sacrifice. The court here says I don't as as I do see a profit sacrifice like in in Aspen skiing that I do that wasn't present in Trinko.

[00:31:17]So that's my two cents about whether or not these distinctions that the court does of AMX and Trinko will hold up. But I will say that they are both crucial to the project of having a reasonable antitrust enforcement especially against platforms. And fin let it all out. Let let me let Bal resp respond to the AMX and Trinko discussion. Then we'll then we'll come back to you first for for mergers and some other topics. You get the last word on MX and Trinko. Well, I look I think I think what I tried to say was the decision was vulnerable because I think it does deviate from the framework in in MX. Um you know other courts may develop MX differently. Um and and I have the same view on on the on the reliance or not on on Trinko here. I mean I think in both cases the district court interpreted um whether it's Trinko whether it's AMX uh in ways that higher courts may not agree and and you know that it may be that the doctrine is muddled but it may be in unmuddling it the higher courts uh tend tend more towards you know what I'd call existing Supreme Court law than um developing district court law and and look I think I think It's a it's a it's a tough matter. So, I never like to criticize district court judges. You know, they they deal with a lot of facts and they deal with all the questions.

[00:32:44]They have to make sense of them. But every time I read a district court opinion, I remember what um Joe Katan said to me when I was in fact still a law student working with with Joe Katan.

[00:32:56]Uh he asked me to do some research for him on a point and I brought him back, you know, a memo that relied heavily on district court cases and very little on appellet court cases. And he said, "Balal," he said, "don't cite district court cases to me." He said, 'They get it wrong all the time. So, you know, I think what, and this is not a criticism, Rebecca, of you, I I agree that AMX and Trinko, u, you know, in the right cases sort of narrow application, the antitrust laws, you know, for for a plaintiff. Um but district courts don't generally get to overturn Supreme Court doctrine. And I think you know um in in both cases she um I don't know if you say either underreads or overreads those decisions and embarks on a path of interpreting them that that strikes me as as very vulnerable to um uh reversal.

[00:33:54]doesn't mean doesn't mean reversal and well it sort of reversal and remand basically to rethink it um which I think would be a killer for both and I think that's also true the search case would be a really a killer for DOJ's case and both if they have to go back you know if the district court has to write another opinion which is um more narrowly interprets um you know appellet or supreme court law. So that you know that that I think is um you know worth worth considering. I mean it it you know we AMX has been criticized when I was at the FTC in the Trump administration. AMX came down um we were very concerned about how broadly it could be read uh and so you know tried to read it narrowly. Um, but what the court has done here, I think just there are a lot of questions about whether she got it right. Um, and I and I just think it makes it vulnerable going going up. That's all. All right, Rebecca, let let's get back to another key part of the decision, you know, on on the merger question and the court's basically, you know, blessing of those two acquisitions and relying heavily on the fact that DOJ raised no objections initially. Your thoughts on that part of the opinion and what it might mean for the Medicaid. and then Joel after that I'd love to come to you to start talking about remedies. Okay, Rebecca.

[00:35:20]Yeah. So, you know, I think district court judges love to split the baby. I think that that's a strategy for um appearing to be uh sort of, you know, even-handed. And here she chose to not find in favor of the government on one of the market definitions as I mentioned and also on this idea that the acquisition of double click was anti-competitive. I don't know how much we can read into this as it pertains to the medic case. So she's she takes a lot of pains to distinguish it from the medic case. So as you point out, one of the similarities is that the regulators at the time declined the opportunity to challenge the merger, which is a little different from blessing the merger, but you know, they they they failed to to move forward. And uh but there's some diff dissimilarities. She says that at the time Google was not a monopolist in this market, which was true when it acquired um DoubleClick. Uh that's very different for the Meta case. Meta was, you know, Facebook was a monopolist at the time in terms of market share, perhaps even more so than they are now.

[00:36:20]And um that's one big difference. And then she also says that there's no evidence that they uh that Google acquired these companies in order to kill them, in order to shut them down.

[00:36:30]Well, I think that, you know, obviously uh Meta disputes this, but there's a lot of proof being offered at the trial right now that uh Facebook did acquire Instagram to shut it down or to at least throttle it in a way that you know would would would suppress competition. Um similarly uh with the way that it has used um WhatsApp uh in a way to sort of prevent these things from really fully developing into head-to-head competitors with Facebook. So, I think that um you know, I just would be hesitant to read too much into that.

[00:37:05]Um and I just think that they're different cases, but of course, yes, there's an echo here between between that case and the Facebook between this ad tech case and the Facebook case.

[00:37:15]Joel, you know, welcome your comments on that kind of merger analysis, but I did want to make sure that we turn and talk a little bit about remedies. and Joel really welcome your perspective on you know what kind of remedies would you expect the department of justice uh you know to propose here and might that include devestature you know despite the the loss on the merger part of the case and I guess I should note um you know as as part of this question that of course it's you know the the president's justice department and he has obviously times been very critical of the tech companies he's been a little a little kinder recently um during the campaign he was asked specifically whether should Google should be broken up and he said, uh, you know, quote, "If you do that, are you going to destroy the company?

[00:37:55]What can you do without breaking it up uh to make sure it's more fair?" So, with that backdrop, Joel, welcome your thoughts.

[00:38:02]What a backdrop. So, uh, well, to to start on on the merger analysis, I I completely agree with Rebecca. I'm not sure I would read too far into this case and say that there are, you know, there's clear indications that a court will go one way or the other. just I I just think there's just too much there there uh to make any firm decisions on how this implicates med the medic case moving forward or uh you know how to think through these problems just in general. So, I I actually very much deferred it to Rebecca um almost all Rebecca's analysis on that just because I I just don't know if there is it's even worth unpacking because there's just a not there's just not enough substance for me to to chew on to be able to make an assessment one more. In terms of remedy, I I'm expect to be uh very much part of the conversation with respect to remedy. I mean this administration uh even with those comments uh that the president has made has made it very clear that they are open to devestatures and this that was a very key part of uh the decision the underlying decision of this case that they the question of you know uh how much influence does Google have on the publisher side of the market was certainly uh a big part of the tying arrangement. So a devesture I think would have to be at least part of the conversation if you with respect to remedy. I I just wouldn't I think it would be naive to think otherwise. I also think there's going to be a lot of transparency requirements. So one major criticism that the adtech market has is uh Rebecca I think even alluded to is that it's happen a lot of this happens behind the scenes and so there's a lot of things that we really don't know.

[00:39:42]There are deals that are made there.

[00:39:44]There are uh transactions that it's unclear, you know, who's setting the price, where's the floor coming from? Is the floor an actual natural uh just as natural organic economic uh uh result or is it manipulated by some of the uh algorithms that Google has in place in every aspect of the stack? The answer is unclear and I think even the core goes uh into the qu into the opacity of the markets and saying really the the reality is is that uh you have these price setting features that Google can avail itself to to know uh where the floor is going to be and with the highest prices even and the competitors would would frankly not have the same access to that same same data. Is that fair? And so I think even Senator Mike Lee came out with a a bill called the America Act with I think he partnered with uh Senator Clolobachar to address that issue in particular. So part of the conversation around that bill was actually uh apppropo to the conversation you had in this case where Google itself during the trial described itself as a stock exchange where like you know we uh don't think of us as anything new or different. That's basically what we are.

[00:40:53]Of course, they backed off of that a little bit at the back end when they realized it wasn't working very well for them, but I would imagine that there uh you'll see some conversation and even some cues from things that Senator Lee has put forward with respect to transparency and also maybe even impose a uh I don't want to say a duty of care, but uh a opportunity to measure conflicts of interest. Uh and so I I think part of it I I I think absolutely uh there's going to be some form of conduct remedy that's going to be in play here, but you can't ignore the structural remedy as well. I think divevestature is clearly on the table.

[00:41:29]It was a it was a key part of the uh of the DOJ's case. It seems as if the uh the judge seemed very interested in uh in uh evaluating that particular arrangement and may even be very open to exploring that divestature remedy. But my two cents well bel welcome your thoughts on you know on all that um including potential remedies and and in in your answer I'd love it if you could talk about um you know how the court should evaluate you know ongoing market dynamics. It's a big part of the search case you know with the growth of growth of AI how should courts think about it in in the adtech space and I'll just note as I understand it Google's revenue from placing ads you know its adtech business has dropped for you know eight consecutive quarters. So I think Google would say, hey, there's plenty of competition out there right now and it's growing. What are your thoughts? So I'll I'll turn to the remedies this point in a moment, but I do I do want to comment on the merger, you know, the the finding that the two mergers were not anti-competitive. I I I don't want to uh overstate the value of a district court decision on what may be a factual question uh rather than a legal question, but I think that the court evaluated those two mergers uh at least as it wrote in the decision based on the facts in the market at the time of the decision by the FTC to clear Doubleclick and and DOJ to clear uh Admelt. um is is something that works that would work within the within the uh benefit to the benefit of Meta, right? I mean both the both their acquisitions uh they were small players not clearly well definitely not in the same market as as Meta or Facebook at the time. So I think that the courts were looked at, you know, the merger acquisitions at the time of the merger and the conditions then um is is at least potentially a positive for meta in in the FTC case. um on the remedies I I don't well first I think both of these cases the both search and adtech should settle um and I think DOJ is not likely to get um uh devestature relief in either of them uh particularly in this case um you know finding um that the acquisitions were not anti-competitive I think it'd be hard to justify um the vestature uh I think although this you Although DOJ has said, I mean Gail said this just the other day, they want to stay away from uh regulation, which I would read as regulatory decrees. Um, you know, this is a market where it's probably pretty easy to monitor uh discrimination or non-discrimination remedy, right? I mean, it goes to timing, it goes to visibility, it goes to ranking. Um, so I think if if there's a settlement here or if there's an order of relief, it'll be something like a non-discrimination clause, right? And and unlike other cases that that don't deal in in basically um computer markets, right, you know, exchanges, um you can you can monitor you can monitor whether discrimination is or is not occurring. Um, I also think as as you said, look, it it's clear from Microsoft that that it's a rule beyond just the DC circuit that the remedy both, you know, needs to be tied to the fi finding of harm, right? And so I think that's another reason you sort of tend towards um, uh, nonstructural relief if there's going to be any relief. I also think the courts recognize they have to deal with the remedy as the market exists today.

[00:45:20]Sometimes that works in favor of the of the plainif right where the market has become more concentrated and less uh easy to enter. You look for structural relief rather than behavioral or conduct relief. Here you do see that um entry is occurring. the market is changing and so requiring a you know sort of a more relief right relief that would address harm you know from a market um position 10 5 10 years ago uh is going to be is going to be overly broad and potentially uh make make Google or Alphabet um less competitive given given the competitive conditions that are developing. So to me it's either no relief or pure behavioral conduct relief. Well bel thanks and that's a great segue to my next question. You know as as the court is thinking about remedies no doubt it's also going to be thinking about it the policy impact of whatever it's going to do. You know Google and the other tech companies you know are you very much arguing that you know remedies should not be so severe that they imperil those companies ability to compete internationally. you have the president out there talking about, you know, the need to win the AI race with China. Um, how, Rebecca, let me ask you this question. You know, how, if at all, should, uh, both the Justice Department and then the courts, you know, evaluate, you know, these concerns about international competitiveness and, you know, the national security concerns as well.

[00:46:48]Oh, I I think that they should be very concerned about um about making our companies as competitive on the international market as possible and that means exposing them to the structures of competition. I mean I I think that the idea that the that the best way to compete against China is to you know sort of artificially prop up uh companies and give them insufficient incentives to to compete and to innovate is just completely gets it backwards.

[00:47:17]So, so I think the good news is that you know international competition issues are aligned with the interests of consumers in the US. You know, work harder for our dollars. Work harder to make make products better for us. This is this is what we want and it's what will make the next big thing. I mean these these companies started out in a competitive environment and that's when they did most of their innovating and then over the years they have become less competitive environments and they have become less innovative and so to me it's just it all sort of points in the same direction. Um, as far as national security, that's a little bit different, I think, than the question of how uh, you know, competitive on the international stage our companies are.

[00:48:01]And I don't I have not seen anything other than just vague sort of flag waving of security issues, you know, like so for example, I would expect that, you know, if the Meta case involved a divevestature or if this case involved a divevestature, we might not sell it to a Chinese company. I mean, that sort of seems like common sense.

[00:48:21]Um, but short of that, it's not clear to me how national security really plays into this other than just sort of as like a general sort of scare tactic to to wave that. Um, so yeah, those those are my thoughts. And and as far as like the the likely relief, I do think we need to be thinking about structural relief here. And the real I would say one of the most important legacies of this case is that it's a vertical case.

[00:48:45]It is a case at bottom about vertical integration. Antitrust has for too long ignored the economic realities of vertical integration and how it hurts consumers in the long run. Not always, not in every case, but it's given almost a total free pass to vertical integration even when it really creates real structural barriers to entry and creates real consumer harm. So this case takes a different view and I think that it is the vertically integrated nature of Google that gave rise to the anti-competitive conduct and so eliminating that vertical integration in this instance through a divevestature would be very closely tied to the conduct um as as Bal says. So um yeah I think that that's I think that's very much on the table um as a possibility and it ought to be. Joel would welcome your thoughts on this of course but let me let let me let me press you a little bit on what some of Rebecca said you know you know she commented that gosh you know Google's not uh you know you know really innovating anymore I think if Adam was here he would say well gosh you know you Google's Google and other tech companies are investing literally hundreds of billions of dollars in AI Google researchers recently you know claimed two Nobel prizes um is is that is is that is that is that a fair critique welcome your thoughts So uh there's a lot to unpack there and and I and I do appreciate the comments. Uh but I think that there's two things. One is that the role of antitrust is the role of interest enforcers are to enforce the law, right? It's this is it's not I mean there are considerations that I think that I think that administration has gone through and especially the Trump administration. It's not like they they're not aware that China is you know a threat. Uh, however, I don't know if that's a good justification for any firm to engage in anti-competitive behavior that violates our own laws. But I also push back on the notion that these companies, large tech companies, are actually our national champions. I don't think they even view themselves as our national champions. I mean you look at Apple's relationship with with with uh with the Chinese government it seems pretty clear that uh Apple doesn't give any uh any ways of of how the America how Americans feel or or even if that they consider themselves as an American companies. Google itself called itself an international company. They don't view themselves as like this pro-Americana uh company. And maybe that might change just because you know the nature of political whims, but I am very skeptical of even remotely thinking that these companies at all have the American interests and like in in like the deep heart of hearts. So I'm not and also on the other end I mean really one way to uh to encourage a win in AI if you really talking about innovation that's all happening with the lower little guys. Little tech is actually doing more in AI than any of the big guys are. the big guys are basically have been de have frequently uh demonstrated that they're slowing things down and gumming things up. And I think that's what you're that's you'll hear that in Gail's comments. Gail's comments could go out of its way to talk about how the little innovator is actually leading the way.

[00:51:44]So, uh you know, I fully appreciate that uh the big companies want to keep up and they're and they're definitely putting in investments and that's important.

[00:51:51]They should do that. However, it it just because you pay something down doesn't mean that we you get a pass on being anti-competitive or slowing down little tech. So, I I I'm just very skeptical of the claim that we need these guys in order to beat China. No, we really need our competitive markets to actually put forward really good products. And you're seeing that time and time again with little tech. you're seeing that little tech companies are like uh even folks that uh who are funded by Y Combinator which is overtly against big tech companies have have a lot of vested interest in there and a lot of money going into these little tech companies that are breaking that wedge in in the web 3 context. So I I do think that at the end of the day what the antitrust laws and the antitrust enforcers should be considering are the facts at issue. I don't know if bringing in uh you know, hey, what will this if if a divestature is the right way to go? I don't know if to resolve the anti-competitive harm.

[00:52:47]I'm not sure the value in thinking, well, how does this go to a broader uh anti-CCCP uh or or anti-China uh ethos?

[00:52:56]I just don't think that's helpful. I also don't know if that uh especially if you're trying to quell one problem. And also, that's not up to the DOJ antitrust division to solve that problem. We have the Department of Commerce to figure that out. We have uh we we have uh other aspects of the DOJ that are looking into these questions that I don't know if it has to rest on the shoulders of Gail Slater to figure out every aspect of uh of international policy. Her job as she said very rightly in her statement uh in front of Congress when she was getting uh she was getting confirmed was that she is there to enforce the law and focus on American markets. That is her job. That is what she's out there to do and I think ultimately that is what she's going to focus on. I I and whether a court should be evaluating this, I'm as a as a constitutionalist, I I I'm very wary of a court uh engaging in strict policy decisions from the bench or using specific cases to make policy assertions that exceed the scope of of what they're evaluating. So I'm very highly skeptical of uh judges sitting uh looking at the ceiling well as Justice Scalia would say pondering looking at the ceiling saying what should we uh what should we declare the law to be today. I think what they should do and what I think what they did here, especially in this ad tech case and the Google search case for that matter, was evaluate the case law at uh the case law ad issue and say and made I think a fair judgment call. And I uh I think really what what's happening here is that big tech has been demonstrated to uh uh to be in a emperor has no pants situation where we're like we I think all of us instinctually know that there's a problem in these markets particularly with with respect to search adtech and the app store issues and I think really courts are coming along to that and it goes back to my Trinko comments where some of these uh more affirmative defenses are just falling by the wayside because they just are frankly nonsensical. Sorry, I uh uh my my fire brand is kind of coming out. Uh but uh ultimately I think that this super helpful, Joel, let's let's get blah in there and we have one more question before we're out of time. Bl, what are your thoughts? Sure. Thanks. And let me let me make one comment. I could see on a on the remedy. I could see some licensing as well as sort of a non-discrimination clause. Maybe there's some intellectual property that would support entry if if they think some relief is needed. So, I could see some licensing, you know, clone type type uh effort, but I don't I don't know if that's necessary. Look, I think I think uh the executive department's uh position in the FTC versus Qualcomm case is instructive here, right? Um DOJ may have uh taken a different position. uh the antitrust division may have taken a different position than the FTC in the Qualcomm case partially because of you know Makin's Makin's view of intellectual property rights verse verse antitrust uh uh issues but I think it's well known that uh people higher up in the department of justice as well as in the other executive branch agencies like Treasury and and and and maybe commerce maybe national security agencies and the defense department were concerned about um the monopolization case against Qualcomm, not because they gave, you know, uh uh uh not because they were concerned about the sort of no license, no chips problem, right? That could be resolved without uh you know, without raising national security concerns. The current the concerns was would would the remedy in Qualcomm or the remedy here in Google affect um Qualcomm's ability to compete with Chinese um uh telecom? uh uh uh semiconductor producers, telecom companies, right? Would it affect national security in a broader way than the antitrust case um needed to be resolved? Um here the concern I assume is as you raised it also that um Google is an important player in the AI space.

[00:56:52]It need it it people may believe it needs scale to be a participant in the AI space. the president clearly wants to win, you know, the AI race and um you know, maybe a few months ago we thought we were very far ahead of China. We're clearly not very hard very far in of China. So I think um that's you know what what they're going to focus on, right? Get that you can you can resolve the antitrust issue, right? I mean I I don't think we care if China is stronger than us in ad tech, right? ad networks.

[00:57:27]what we care about is whether Google and other firms uh can support the investments to develop AI and and you know I I think at the margin the agency should take account of that but that really is not Gail Slater's call may not be Pam Bondi's call maybe it's the defense department maybe it's the national security um uh universe um and that's just another reason I I don't think you're going to get the vesture they they this administration believes in um national champions in support of national security and right now Google is a major player in the AI space and they are not going to they are not going to fix an antitrust problem um in a relatively small market uh in a way that undercuts Google's participation in the AI race. I mean it's just not going to happen. Well, thanks for that below.

[00:58:26]Look, I did want to ask, we're almost at time, but I did want to ask one more question about speech, which seems just in integral to many of the complaints about the tech company. So, if I can ask you for maybe 30, 45 seconds on this question. Um, Rebecca, starting with you, um, in her statement after the verdict, uh, the attorney general mentioned that this case would stop Google from censoring free speech. What do you make of, uh, her comments and, you know, the relationship, if any, between this case and the speech concerns?

[00:58:58]I don't think that statement makes any sense and I don't think that this case has much to say about free speech. I think that the idea that big tech censors free speech has been a sort of way of bringing along different sides to the debate about big tech market power.

[00:59:18]Um I think that at the at a very abstract level different tech companies if facing competition would provide different platforms and therefore maybe more speech options but there's a lot of problems with that also thinking about speech that way that we need to have like different siloed perspectives. You know we already have like truth social versus Facebook. So I think that unfortunately maybe unfortunately antitrust has very little to say and to do about the censorship of speech. Um and I think that's just sort of political posturing.

[00:59:53]Joel, well I obviously disagree. Uh I think that it uh that uh online censorship is actually just a way to describe the the actual consumer effect of this market concentration when it's not actually all that different from some of the concerns that were raised within the newspaper market and all the ad the ad market behind that. I mean it's the question is like who has the money and who controls the concentration and breaking up that concentration also mean could mean a net benefit for the consumer and that would be in this case as Gail pointed out uh potentially few fewer instances of censorship and I think that's all she's really getting at. So I do think that there is a close connection. I don't think it's as tenuous. Mal, you get the last word.

[01:00:33]Well, I don't I don't want to criticize the AG. Um, I think I think it will be it is hard for me to see the tie between uh the alleged censorship and and this decision, even a remedy in this decision that makes sense, but maybe maybe they're thinking a little bit um further further a field and a few steps ahead of me, you know, but I don't I don't see it. Well, guys, thank you so much for an excellent discussion. I wish we had more time to really dive in, but uh you know, on behalf of the society, really really appreciate it. And Edie, back to you.

[01:01:05]Yeah, just to echo that, thank you so much to Professor Ellensworth, Balal, and Joel for their excellent discussion today. And thank you Ashish for moderating. We're super grateful for your time and expertise. And thanks to our audience for joining us. We really appreciate you tuning in. You can stay up to date with other announcements and upcoming webinars on our website, fedsock.org, or on all major social media platforms. Thank you once more for tuning in, and we are adjourned.

[01:01:29][Music]