Nov 25, 2024
On November 21, The Capitol Forum hosted a conference call with Tim Cowen to discuss the outlook for the Google adtech cases in the U.S., EU and the UK following the Presidential Elections and change in leadership at the European Commission. The full transcript, which has been modified slightly for accuracy, can be found below.
JACOB PARRY: Good morning and welcome to our conference call on the “Next Steps in the Google Adtech Cases in the United States and Europe.” I am Jacob Parry, a journalist with The Capitol Forum, based in Brussels.
With me today, I have Tim Cowen, Chair of the Antitrust Practice at Preiskel & Company. His practice covers a range of competition law matters across media and telecoms. And he is an expert, in particular, in digital advertising, and I think fair to call him a veteran of many of the agency actions against Big Tech in recent years.
We are going to be discussing the Google adtech cases that are ongoing in several jurisdictions on both sides of the Atlantic. The DOJ case, the European Commission case, the UK Competition Markets Authority case and, of course, the case in Texas.
We’re going to be talking a bit about where those cases are headed, given the changes in leadership that we will soon see both in the EU and at DOJ. And then I think, a bit we’ll discuss as well, potential remedies and some of the substance of the cases.
So, before that, I’d like to go through a few housekeeping items. You’ve joined this presentation using your computer’s speaker system by default. If you would like to switch to telephone, select telephone in the audio pane and then the dial-in information will be displayed. To submit questions to today’s presenter, you can type them into the questions pane of the control panel. We will collect those questions and then address them throughout the call.
So, Tim, before we get into the adtech cases, I wanted to touch on the big development that we have today, and that is the proposed final judgment in the DOJ search case. So, we’re seeing a proposal here to divest the Chrome browser, potentially even divest Android. There are other commitments around data access, ending selfpreferencing.
First and foremost, I’d like to get your reaction to this proposal. How are you thinking about this? And then maybe we can even get into how this might fit into the adtech cases, just as Chrome is such a central issue there. So, that said, I’ll hand the floor to you on that question. And let’s get into it.
TIM COWEN: Okay, great. Yeah, what timing. You wake up in the morning and you find the DOJ has put forward its proposed final judgment. So, I’m a few hours ahead of you. So, I’ve been through it a bit.
Yeah, there is a big overlap—that’s the first thing—between what the DOJ, USA I should say, is doing on the search case and also what’s happening on adtech. But they’re just focusing on search.
So, the problem, if you like, the antitrust problem, is that Google’s monopolized search and search text advertising. So, how has it done that? Well, the first thing that it’s done is to promote itself via various distribution partners. There are hardware distribution partners like Samsung where a series of agreements have been put in place where the device manufacturer carries a suite of Google products, Google search, Google Chrome, and all of the Google products that we know. And those are going to be prohibited. So, that’s the first thing.
The second thing is, again, Google has insulated itself from competition from the likes of Apple. And there is an agreement in place between Google and Apple through which Apple is paid based on Google ads. So, Google search ads revenues, it gets about 36 percent which worked out as over $20 billion at the last count. When I say the last count, I don’t mean this year. Because I think Meta was looking at the judgment on the Google search case, which was looking at 2021 numbers. So, it’s probably higher than that now.
So, Google’s going to make a saving of a huge amount of money because it won’t be able to pay Apple for that in the future, which is first very short-term cash benefit to Google. But in the longer term, it can’t promote Google’s products on the Apple platform. So, that’s a very big total shake up, the industry point.
I emphasize the fact that it’s a percentage share of Google search text ads. Because Apple users are the most valuable users on the planet. And obviously, if you’re advertising to people who have a high disposable income, you’ve got more opportunity to make sales and have adverts that are going to generate useful interactions with end users. So, this is a major, major, major issue for an advertising company. So, that’s about the agreements.
What the DOJ has also pointed out is that search is one thing, as it’s done by Google search, a search engine. But there are search access points which include the browser. So, its proposal is to prevent a search access point such as the browser from being used as a mechanism to restrict competition with search, that Google has to divest the browser. And that makes sense in the context of this monopolization of search.
Again, in all U.S. cases, the first issue is prohibition of the anti-competitive activity. And then what the authorities can propose, and the courts are bound to do something about, is the ill-gotten gains problem. So, if the beneficiary of anti-competitive activities has got the benefit of ill-gotten gains, then the remedy can address those. So, you can’t just keep the benefit of acting anti-competitively for years. And that’s part and parcel of how the remedies are designed, as the DOJ puts it, to thaw the ecosystem and finally reverse years of anti-competitive offence.
And then there’s a series of provisions. So, the third part of the DOJ system is a system of oversight and control for a technical committee which is modelled on what was really done in the Microsoft case some years ago. I would say that in the DOJ’s analysis, it cites multiple legal authorities for the position. So, that anybody suggesting that what it’s putting forward is radical, new, out of line with precedent, is just not the case. So, that’s just kind of a high-level perspective on what’s been proposed and it’s being pretty well founded. We can go a little bit further into that if you like.
JACOB PARRY: Yeah, I mean, I’d be curious to go a little bit further. And maybe in particular, we can go into how some of the solutions being proposed in the search case might come to interact with the adtech case. I mean, again, the Chrome browser there being sort of a key point that’s come up in adtech as well. So, maybe if you want to go a little bit further, in particular on the Chrome element, I’d be interested in your thoughts.
TIM COWEN: Yeah, that’s fine. So, what’s happening in the adtech case is essentially that Google controls a stack of products from—if you start with the beginning of the user journey, you sign in, you then go to the browser, the browser then enables you to search the web, but you then use Google search. That interacts with adtech and there are server-side brokerages and advertising exchanges and supply side that fills inventory with available advertising. And the essence of the case that DOJ is putting forward there, and the European Commission is putting forward, and the UK is putting forward, is this control over the entire advertising technology system creates a conflict of interest. I think the DOJ put it that it’s rather like being the broker for the stock exchange and also owning the stock exchange.
So, clearly there’s an ability to manipulate in the interests of the person who owns the whole thing. And what Google has done for 10 or 15 years in the adtech case is to block third-party access to data, is to manipulate the bidding systems and give itself either a first look or a last look at the bids so that it can top the bids of third parties, thereby winning the race, winning the competition. So, what’s likely to be proposed, certainly has been proposed by the European Commission, is breaking off the advertising exchange from the rest of Google.
Now, if you stand back and look at both of these cases together, what you’re actually seeing is that the monopolization is across a value chain. And, in essence, what all the authorities are saying is, well, the control of that value chain give rise to benefits for the organization and no benefits for anybody else. It’s enabled monopolization and massive returns on capital.
So, what we’re seeing is the authorities on both sides of the Atlantic saying, well, what we need to do is to break up this vertically integrated organization. So, you can see the advertising exchange being broken up. You can see the browser being—broken out is the way I think about it. Because I think it’s really a question of policing the remedy for the monopolization of search, that you need to have a separate browser that is doing browsing, not doing searching or isn’t being used to manipulate or bypass a prohibition on search self-preference, which is at the heart of the DOJ search case.
There’s also something, which hasn’t been mentioned in the DOJ’s proposals from yesterday, is in order to create competition in search, it’s providing an access remedy. So, it’s access to the Google search index. So, what you can see that giving rise to is huge opportunities for new search engine businesses and existing search engine businesses, the likes of DuckDuckGo and others, to get access to the Google search index.
Now, that clearly provides an opportunity for others to come in, they can take the relevance engine and then represent that which has been identified by Google as relevant to a particular query and present that in a different page or a different presentation. And that should give rise to opportunities for competition in search presentations and search results and different forms of both the presentation and display, but also quantities of relevant results by comparison with advertising. Because, at the moment, what we’ve got in Google is a pretty crammed end user experience where a lot of the page is these days really full of ads. So, the opportunity clearly exists for more relevant results to be displayed alongside different formats for advertising.
So, that’s likely to drive huge amounts of innovation, massive opportunity in competitive search, and also there’s a massive opportunity for other advertisers. So, you’re talking about how does this affect the broader advertising markets? Well, one of the ways in which Google has managed to monopolize search is to have all of the relevant data from previous searches and data from user interactions with all of its products that it can then use to target ads more accurately than its rivals and one of the aspects of the access remedy that the DOJ has proposed yesterday is that the rivals get access to the search index which is relevant results but also data which is needed for advertising. So, third parties that are in the advertising adtech game should be very happy with that. I mean, overall, I think it’s a pretty comprehensive suite of solutions. It’s not just a kind of there’s a problem here. So, we’re going to break it up. This is an identification of what the issues are and then a very careful calibrated assessment of addressing those issues through a specific set of remedies designed to ensure that there will be competition in the future.
JACOB PARRY: And these are feasible, implementable. I mean, I look at La Bruta from Brussels. We have a deck, I mean, not speaking on what’s being proposed structurally, but on the behavioral aspects. I mean, looking again from Brussels, we have a decade of enforcement against Google with behavioral remedies that have not come to much. We have the French adtech case as well. I think we’re just outside of the windows of the applicability of those remedies perhaps. But just in terms of those behavioral remedies that are being proposed, the access remedies, are those implementable? I mean, even this technical oversight committee. I mean, where do you see this fitting in? Is this a remedy that can work?
TIM COWEN: Look, I think one of the things that is not expressly stated in the DOJ’s summary or the proposed final judgment is one of the things that drives authorities when they’re developing remedies is to make sure that they will work. It’s very important that they work. And in order to do so, they need to change incentives. Because if you’ve got monopoly capability, the assumption has to be, when you’re designing a remedy, that you’re going to be able to misuse your monopoly power for your own benefit. That, after all, is what they’ve been accused of, and found guilty of, for at least 10 years. So, the idea that that leopard’s going to change their spots overnight isn’t really very feasible.
So, what’s interesting about the structural remedy is that it’s designed to do exactly that. It’s designed to change incentives. So, if you imagine the day after, if these remedies are implemented in the following form—which we can come back to because there’s a whole U.S. processor in there—you would have competition between third parties and Google in search, using the Google search index. So, there’s no inherent advantage available to Google from having its own search index anymore. Everybody gets access to it.
A new presentation of the results is likely to put a lot of pressure, and is intended to put a lot of pressure, on Google to compete more effectively to produce what users really like, which is highly accurate results to their questions. And that they would have a range of alternatives to choose from. Knowing, as everybody will, that this comes from Google actually provides the competitors and the users with a degree of assurance that what they’re looking at is likely to be accurate and useful. Because we all know that Google search results are actually pretty good. So, I think it will put a lot of competitive pressure and that changes the incentives for Google.
So, is it likely to work? Well, Google’s going to be very busy ensuring that it competes more effectively in providing better results rather than polluting the page with large numbers of ads. Because there’ll be a choice. And that’s likely to be quite a considerably different position than perhaps we’ve seen for the last 10 or 15 years. So, we’ll see what happens there.
I think the additional controls over the browser, the ability to control manipulation of the technology stack, so that obvious discrimination can be addressed and overseen is really secondary. I mean, you look at the Technical Oversight Committee, I wouldn’t think of that as being the principal enforcement organization. The way that the system is intended to be enforced is by creating competition, and that should create the competitive pressure and innovation.
I think one of the big question marks I have is how quickly that can be done. Because competitive pressure on a business that’s generating the scale, the revenues, the relationships that Google has, is going to take a long time. I mean, if you took a billion dollars a year off Google, you’d have to be—what is that? $380 billion? That’s 300 years, 380 years. So, it’s just getting to some sort of equivalent parity and scale is something that really needs to be thought about I think at the moment because I’m not sure that’s there.
You know. releasing innovation is certainly likely to happen. But doing it really quickly, so that it provides a serious amount of competitive pressure, is likely to take a bit longer. And that brings me back to the process. Because what we’ve seen announced yesterday is a proposed final judgment. That’s subject to response from Google, disclosure, hearing, final judgment and possible appeal. What’s put in place between, for the stay or the battle which we’ve seen on the Epic case and in other cases, what happens, once this is finalized as a modified final judgment and what’s then imposed on Google in the interim, will become I think a very important step during the course of the early part of next year. But yeah, the stage is set for competition, but we’re not seeing it just yet.
JACOB PARRY: I want to switch over to the Adtech case in particular. We’re midway through the process on the DOJ case. I guess we’ll have closing arguments next week. The European Commission case, I think we’re expecting a decision in the near future. Looking at this case, when we had the SO back last year, the Commission had this nice graphic depicting a sort of a snippet between the sell side and the buy side. I mean, there were reports early this year of Google maybe floating the idea of selling AdX. I mean, what is a solution in this case? We can be specific to jurisdiction. We could be general to jurisdiction but when it comes to sort of the concerns that we’re seeing, should these concerns make it to a final decision in particular in the European Commission case, what does a solution look like?
TIM COWEN: I think it’s actually likely to be quite a similar analysis. Because if you think about what the DOJ has done, they’ve said, okay, we’ve got a monopoly asset here, which isn’t likely to be replicated. It isn’t likely to be reproduced or created by someone else, which is Google search index. I mean, Google search index is absolutely massive. It very successfully creates what is effectively a duplicate web that can be accessed extremely swiftly and then searched against. So, third-party access is kind of the number one remedy in the Google search case.
So, if you’re looking at AdX, your advertising exchange is clearly problematic and the control of that’s problematic so you could sell that to a third-party. But there’s still going to be a need for access to the underlying source of the Google control or lock or choke point on the market, which is its enormous base of advertising. And that’s going to have to have some form of non-discriminatory obligation imposed on it because it’s generating huge amounts of advertising.
Now, just pushing the advertising exchange into different hands won’t limit its dependency on Google. It won’t create overnight competition with advertising exchanges. There are other advertising exchanges. AdX is the biggest. If it’s owned by somebody else, it’s still going to be the biggest and it’s still going to have a relationship with Google.
So, what then becomes important is how the—if you think about it as an access remedy for other third parties that are running advertising exchanges to the Google system. So, Google doesn’t prefer its own former advertising exchange then becomes important. So, there’ll be a non-discrimination aspect to it as well I would imagine in the advertising exchange type of case.
JACOB PARRY: And then implementation, I mean, how feasible is a remedy in Europe? I mean, I think maybe it was with you or I read elsewhere, but looking at the BT Openreach solution from 2016 as a template, is that a template for the EU case? I mean, how would you be thinking in terms of what materially this could look like?
TIM COWEN: Well, look the antitrust authorities and jurisdictions have been talking to each other for many years. So, if you look at one of the early cases, of Quarantine, it was a AT&T remedy in the early 1950s. We then obviously had the AT&T divestiture in the early 1980s, which did transfer monopoly control into other hands. So, you created the seven regional Bell operating companies.
I think what happened across Europe—and I’ll come back to Openreach in a minute—is there was a recognition that vertical integration means that those that own the means and both the production and the distribution of the products tend to prefer their own products even if they’re not the best products on the market.
Now that kind of naturally surfaced in energy before it ever came anywhere near the world in telecoms. So, production assets, whether it’s a coal-fired power station or a nuclear power station or gas, they were all separated out under what’s known as the third energy package. But it was an antitrust issue before it was dealt with as an oversight and monitoring system through regulators. And that antitrust problem is really at the heart of how all of these are being addressed, which is you need to separate out production and distribution.
So, BT Openreach, which is exactly the same really as the AT&T divestiture, there’s a control of the local access monopoly in BT. And what we did there was we separated out the downstream distribution of communications between BT Openreach which is wholly owned and the rest of BT.
Now, Openreach is an interesting model. Because one of the defects and the problems of separating out ownership is that you put the ownership of a monopoly in somebody else’s hands. That’s happy days for the person who suddenly becomes the monopolist because they get to make a lot more money. And there’s nothing very much you can do to stop it because the monopoly makes money by either increasing price or degrading quality. And in communications, IT, tech, whatever, you can do either one of those parameters. You can make sure that your system has very little investment. That’s largely what happened with AT&T in the States, which was the local access wasn’t invested in very much. So, they increased their profits as a consequence. We see that also in water companies and other aspects of monopolized industry. If they can reduce the amount of money that’s spent on investment, you increase the profitability.
So, what can you do to increase investment then becomes a critical question. And what we did with Openreach was to say, well, if the downstream only uses the asset, they’ve got an interest in making sure that it works pretty well. So, if you look at what the DOJ is doing with its access remedy in search, they’re saying Google continues to own search. Others will get access to it. But Google has an interest in making sure that search works well, subject to non-discrimination obligations. So, it’s going to invest in upgrading search. That’s great. Because that’s good for everybody. And the non-discrimination obligation makes sure that it’s good for everybody subject to an oversight and control.
I think the same issues come up in relation to adtech. How do you divest and make sure that you’re going to continue to invest? Well, if you divest in the advertising exchange business, there are a few other advertising exchange businesses that will keep the investment flowing. If you’ve got, say, ten players in the advertising exchange business, then it’s important to continue to invest to ensure that, even if you’re the biggest advertising exchange, that you’re actually the best and you work really well.
So, I talked about degrading quality in advertising exchanges, increasing quality involves investment in reducing latency, making sure that things work really fast. So, you can see that that’s likely to be a very key parameter in ensuring competition works well. So, you can see that could potentially operate pretty well in a divestiture of adtechs. Because there are other competing exchanges that will keep everybody honest. It depends who it’s—you can’t divest it to a monopoly or create a monopoly through divestment because that’s going to make the problem worse rather than help solve it. But the people who are in charge of dealing with that are competition authorities and they would have to go through an assessment. So, if any divestiture that happens in adtechs, it would be part and parcel of the solution.
JACOB PARRY: And hypothetically, there are buyers out there. This is an ecosystem big enough where there are other parties who can come in.
TIM COWEN: Correct. And whether that’s done, this is very valuable business. It’s going to be held in different hands and there’ll be different profits in different places. So, there’s real opportunities for even smaller players leveraging their positions. You can see that. And to make a competitive outcome, which is what the authority is going to look for.
JACOB PARRY: So, these cases do not happen in a vacuum. We have a change of administration coming soon in the United States. We will also soon have a new Competition Commissioner. I have a question in the chat related to these political dynamics. So, maybe we can get a bit into the U.S. part and then into the European part after. But the question is would the Commission continue to seek a divestiture remedy for adtech if the U.S. were to drop the case as a result of a change in administration? So, why don’t we talk a bit about the politics around the case?
TIM COWEN: Yeah. I mean, look. Start with search because that kind of came first. That was started by a Trump administration. Now, some people I know have said, well, yeah, but that’s Trump administration with different people in it. From everything I’ve heard so far—and there have been very few appointments—but those people who are in the running, it’s quite likely that the people who are likely to be appointed to the DOJ are antitrust enforcers that recognize that monopolies need to be dealt with. There’s a degree of politicization of that. But I think on both sides of the political aisle, there’s a recognition that these monopolies are kind of out of control. So, that’s kind of the first thing from a purely antitrust point of view.
From a political point of view, I think the original search case was launched because your President in the States has no—there’s no love lost between Trump and Google historically. And I can’t imagine that’s changed very much. We’ve also seen JD Vance talking very openly about breaking up Google. And there is a mood—and we’ve seen certainly during the course of last year—a considerable change in the attitude of Silicon Valley financiers and venture capitalists toward what’s become known as Big Tech with the launch of the little tech movement. That’s a movement of VCs, as I understand it, that are looking at how their investments are currently restricted in terms of their returns, because there’s a limited number of outlets for their products.
If you have to go through a Google or a Facebook, and you’re in the advertising world, then there’s a limited number of places that you can go. If there were a more open, competitive, broadly based marketplace with large numbers of search engines and a broader based ecosystem, that’s a huge opportunity for, first of all, growth and employment, which is actually not a bad thing for the economy, but also it’s a huge opportunity for venture capitalists that seem to be backing quite a lot of the Republicans and have backed JD Vance very publicly.
So, that looks to be reasonably aligned on the Republican side. I think it’s unlikely that there would be a big change. But then Donald Trump’s Donald Trump. And I wouldn’t want to be in the business of predicting what he’s going to do.
JACOB PARRY: No, it’s definitely a lot of uncertainty. I’ll add as well, Matt Gaetz, the current pick for AG, no friend of Google [nota bene, on November 21 Gaetz recused himself from the role]. So, it’s going to be a complex and fast-moving situation, I imagine. Which I think can maybe transition us to the question that was posed. How does the Commission respond? It is quite advanced that we have here in Brussels. How do the dynamics in the United States affect the options that the coming Competition Commissioner, Teresa Ribera, will have on her plate?
TIM COWEN: Yeah, I think it’s a fascinating question. There is a gap in the Commission’s forthcoming diary.
It does look like they have a finalized decision on the stocks ready to go. We know what their proposal would be, which is kind of a breakup of the ad exchange and the rest of it. So, that’s already been said.
I think that this announcement from the DOJ is at a high-level, politically reinforcing of the Commission. It’s not the Commission just coming out of the blue saying you should break Google up. This is intellectually the alignment of the authorities on both sides of the Atlantic.
The problem, I think, is who do you call? Because there’s nobody in place in the States. So, you can’t just phone somebody up and say, what do you think? Because there’s nobody in charge. Anybody that’s currently there is on their way out. So, it’s a very awkward period from that point of view.
So, the cautious response from the Commission would be to wait and see. But there’s also a political angle on this, which is Vestager has been at the forefront of the antitrust issues for quite a long period of time. This is on, if you like, her watch. You can see a decision which is consistent with all the jurisprudence and everything being published, which identifies what the problem is. This is the way the Commission dealt with the Microsoft case.
It said, here’s the antitrust problem. And then proposing that the solutions would include breakup, wouldn’t be politically very difficult, I would have thought. Since structural remedies would be among the range of remedies that we need to consider.
You could see a publication of that type of decision by the end of November is what we were thinking. And then there would be time within which precisely how the remedies can be coordinated across the Atlantic, take into account the court’s decision in adtech. That time window is certainly very possible between now and first quarter next year.
So, I would kind of think that’s still very much on the cards. But if there’s a lot of caution because people are nervous about the way that the Trump administration would react, then maybe they’ll just bury it for a while. And it becomes an issue to be dealt with once relations have been reset.
I don’t think that’s as likely. Because I don’t think that the starting point with the original Trump administration is really that dramatically different. So, I don’t think there’s a real need to wait and see it as a big problem. But one of the things that I think the way that the States tend to operate is where companies are primarily—I mean, Google’s a global company. But companies that are primarily headquartered in one jurisdiction tend to be, you know, the authorities defer to each other in terms of who takes the first step. So, now that DOJ said, yeah, break up some cards, that means the European Commission are not taking the first step. So, I think to that extent, you’re falling in behind the DOJ’s analysis. It’s in two different cases, but it’s a simpler sort of analysis.
JACOB PARRY: I mean, because I just want to emphasize, we have not seen the European Commission order the breakup of an American company before. So, there’s not a precedent for this, and this would really be a precedent-setting move. So, you think that there’s sort of enough (a) DOJ plowing this path, (b) it’s being sophisticated and advanced enough for the case at this stage to merit breakup as a solution as being a reason why the Commission would diverge from precedent. Because I imagine in previous cases, I think I’ve heard anecdotally from officials over the years, I mean, the European Commission doesn’t break up companies. That has been—at least not with Article 102—that’s been the hither to conventional wisdom.
TIM COWEN: Yeah, I’m not sure it’s right though. I mean, if I look back at the history of divestitures, I remember sort of the late 90s, early 2000s, RWE, a German company, which is, when I mentioned energy earlier on, RWE, the divestiture, the separation, whether it’s divestiture, which is to put assets in separate ownership hands or separate corporate vehicles, is something perhaps that isn’t focused on enough, because you can achieve similar outcomes with separate corporate vehicles without actually having to—like the BT Openreach system was one where you have separate corporate vehicles and it’s the same ownership. And that’s there to guarantee investment that everybody benefits from, as I mentioned earlier. So, there is actually an OECD report on structural separation and functional separation, and there’s over 100 examples.
Now, your point, well, this doesn’t involve American companies, I think historically it’s tended to relate to monopolies which are gas, water, electricity, telecoms. They tend to have been national by history. I don’t know whether any of them had substantial American ownership. I just don’t know. But I don’t think that’s a material consideration in the antitrust analysis. Certainly highly important in the political analysis. But I think it’s a secondary question. If you’ve got a monopoly that’s abusing its dominant position, whether it happens to be owned by shareholders of one particular country or another, hasn’t really ever been a consideration.
I think the more interesting question is, does the Commission really need to do structural separation? You know, would a behavioral remedy be enough? And I suspect that what will actually be put out there is that they’ll say, well, structural and behavioral in combination are going to be needed. And precisely what that combination is going to look at, do you need to have completely separate ownership? Maybe you do, maybe you don’t. That depends on whether you can see disaggregating of a vertical entity being more effectively policed and enforced with separate ownership. Because you’ve got assets under separate vehicles. You can see what’s going on across the asset boundary. Because you have corporate entities owning various assets, as in the Google case. You can own the search index, then there’s going to have to be transactions.
And the DOJ is talking about license agreements. Well, they’re going to be between entities. Because you can’t just have license agreements where you don’t have an entity on either side of the license agreement. So, you need separate corporate vehicles to own assets and control assets. And then you can police what’s done with them, and how those assets are used. And that’s then subject to agreements.
But ownership is a separate question from structural change. And I think that’s something that can—how you finance things can be done in many different ways. So, I think that all of that is likely to be on the cards in advertising, in the ad tech position. I think it’s probably easier to create competition in the advertising exchange markets, because there are other competitors there.
There really isn’t very much—I mean, yes, there are competitors in search, if you look at DuckDuckGo and other things, but they’re not the scale and size that Google Search is. And Google Search derives its scale and size, according to the judgment and everything we see empirically as a matter of evidence, from huge amounts of scale worldwide.
And that means you really need access to the search index in the Google case and divestiture of things on the edge. I mean, it’s surprising to say that Chrome is on the edge, but it’s actually getting divestiture of Chrome on the edge so that you don’t have a mechanism for bypassing the access regime and policing the self-preference regime. That’s the reason for the divestiture there.
JACOB PARRY: And perhaps this is a bit of a naive question, but sort of gaming out a potential scenario where we’re looking at this sort of remedy, I mean, are we waiting for the general court and then eventually the European Court of Justice to clarify before we see an effective sale? Will we see the Commission coming in earlier? I mean, I’m just trying to understand how in Europe any sort of divestiture process might interact with the appeals process.
TIM COWEN: It’s a great question. Because if you look at the Microsoft position in the States, what happened when there was an order against Microsoft that required it to be divested. That order was appealed by Microsoft.
And between the order being put into force and the appeal, it was settled under a consent decree. And that was partly because there was a new president in the U.S. So, there’s clearly a political influence over what happens at DOJ. I mean, whoever’s put in the presidency then gets to decide who they nominate to be the AG. And it’s clearly political.
In Europe, it isn’t quite so political. It’s probably a lot more bureaucratic to be fair. So, again, if you look at the Microsoft case, there was a series of remedies put in place that were simply enforced by the authorities. They didn’t apply in the U.S. They applied in Europe. Choice Cream notoriously came in here. And that was applied for quite a long period of time until people forgot it was there. And then it got switched off rather unfortunately, which led to fines and all the rest of it.
But I think that those, to answer your question directly, what happens if there’s an appeal? It’s quite difficult for the party that put forward the remedy to then say, I don’t like the thing I put forward. So, what the Commission has done, as a matter of practice, really since Microsoft, is to say, okay, here’s the problem.
This is why I think that it’s entirely likely that there’ll be a decision saying, here’s the antitrust problem. And we’ve got a range of structural behavioral remedies that may solve the problem, and then put the ball in Google’s court. So, you put it in the defendant’s court to come up with a precise remedy. When the remedy is then advanced by the defendant, the defendant kind of has to live with what they put together.
Now, we’ve seen that, and we’ve seen how that played out with Google in search. Because Google did put forward a remedy to the 2017 Google search decision by the European Commission. And notoriously, in the advertising world, it didn’t work. And that one was appealed all the way up to the European Court of Justice, and that came down on the 10th of September this year. And that’s a matter of ongoing litigation. But you can see, maybe the Commission this time would spend a bit more time and effort on making sure that whatever is proposed actually works before allowing it to continue in the marketplace.
JACOB PARRY: I want to get to questions, but maybe the last stop I want to make is the United Kingdom.
We had the CMA advance its case in September. In the briefest way possible, what is the CMA doing? And what could be an outcome of this case? I mean, obviously, the CMA has been doing work on digital advertising for years, going back to the market study reports, very high-quality work. Is there an outcome in this case? Or are they just waiting for DOJ or the Commission to act?
TIM COWEN: I think it’s one of those kind of rabbit and hare situations where the – sorry, maybe it’s the hare and the tortoise situation. But what you’re looking at is the CMA is actually behind in terms of the timetable. But when the CMA makes a determination, let’s say that there’s an agreement or a settlement or a set of commitments, I think one of the material differences with the CMA is it’s likely to be behind the others and take advantage of what the others have been deciding. That looks to be reasonably clear from the current timetable.
But once there is a decision, the one big difference between the systems is that we know that the European system has taken seven years in relation to the search case. I mean, that’s a lifetime. And Google’s doubled in size in the interim.
The U.S. court system similarly can take a long time. The U.K. court system doesn’t take that long. So, if you want to judicially review the decision of a U.K. authority, because that’s the way you would probably have to look at it, then that can be done in a matter of months. It’s not years. If, on the other hand, you’re trying to appeal the whole thing, appeal courts don’t like getting into the facts. So, you’ve got to find legal issues and they’ve got to be justified legal questions.
You then have to get, either the permission of the court that you are saying got it wrong, which is rare and very unlikely, or you get permission from the Court of Appeal, which in something that needs to be dealt with quickly and that would then be a case for expedition. So, it would be heard, I think, more quickly than it would in other places. So, that’s kind of an interesting difference between the two. I think on the substantive analysis, there isn’t any difference really between the way that the European Commission, U.K. and U.S. has been looking at these issues.
JACOB PARRY: So, we have a few questions. I want to start with one that brings us back to the search case.
So, the question, the browser Chrome was created by Google from scratch. Isn’t it a bit disproportionate to require a divestment of a business that was created from scratch and works just because it’s a better product? So, that’s the question on Chrome.
TIM COWEN: No. I mean, I don’t understand anybody who would say that because it’s – you know, the logic of a monopoly being created from scratch or not isn’t relevant to what you do by way of a remedy. Remedy is the creation of competition. And if your only mechanism to create competition involves divestiture, it doesn’t matter whether it was created from scratch or it was given to you by someone else or you completed somebody else’s half-finished work. It’s just not relevant to the analysis.
JACOB PARRY: So, that’s a short and simple answer there. I have a second question that is maybe on the broader maybe social political elements of these cases, in particular on censorship as a potential negative social outcome of Google’s actions. The question, by hiding and penalizing newspapers online, is Google deciding what media houses can and cannot talk about?
TIM COWEN: Look, I think it’s a massive issue. It’s been a question that’s been at the forefront of people’s minds here in the UK, and I know in the U.S., that if you control advertising, you control speech and democracy because they fund all of the independent publishers.
So, we have a concept here in the UK called plurality of the media, that you have to have a diversity of voices and a diversity of media that can carry those voices and have different opinions. Because if you don’t have different opinions, you don’t have a dialogue. You don’t have democratic debate. So, any concentration of power and monopolization of distribution becomes a fundamental threat to freedom of speech.
And that’s exactly what we’ve got going on in the background here. These are antitrust cases. But when we filed the original case for the movement for an open web in front of the CMA back in 2020, our entire basis for the claim was, yes, there’s an antitrust problem, but the consequence of this is a threat to democracy and freedom of speech.
I think that’s just got worse. You can see that, since then, the last three, four or five years, we’ve seen increasing, I think, awareness of just how social media is undermining the income streams for publishers. And that’s become a bigger and bigger problem to the extent that if you look at what’s happening with the next generation of search, which is AI, which we haven’t talked about. But if you’re looking at how AI is eating, literally eating, the news, ingesting news information and then creating answers, that is a massive threat to the independent publishers, which is being wielded by big platforms like Google. Incidentally, DOJ does address that and would require Google to divest all of its AI assets, which is quite a dramatic intervention.
So, yeah, there is a recognition that is really important. If you have a higher number of competitors and a more competitive environment, you’re going to have more plurality of the media and more opportunity for a variety of different sources of income to fund a broader range of publishers. So, that would be one of the major benefits of having a more diverse structure to the industry.
But then you’re looking right the way across the board of all online publishing when you’re looking at that. And there’s quite a range of range of issues there. You know, the Federal Trade Commission has suggested there should be a breakup of Facebook and all of its previous acquisitions. So, Instagram, WhatsApp.
So, if you actually think about—we talk about social media. I’ve always wondered why we call it social media.
It’s just media. And if you think about it as media, and it’s digital media, there needs to be a variety of different outlets. Otherwise, you won’t have a diversity of voices. Now, where you’ve got a concentration, as has happened, that’s going to limit a democratic debate. So, I think it is fundamental to that. Yeah, I think it’s very worrying that it’s got so concentrated.
JACOB PARRY: This seems to be an issue where it animates the Republicans as well. So, it’s somewhere where there is a bit of consensus. Final question, and then I think we’re going to wrap up. Does adtech case matter? X is a small sliver of Google’s business. Is an outcome here future proof? So, I guess going there, of course, we know the importance of Chrome and Android. Is adtech still and will it be a fundamental part of Google’s business?
TIM COWEN: I think if you, again, stand back and say it’s not really the impact—the funny thing about this is if you look at the overall impact on Google, it probably will end up making more money. And that’s one of the things that people tend to get surprised by. But in principle, if Google’s promoting its own products and people don’t like them, presenting its own products in ways that people find difficult to deal with, then breaking it up, you might be doing a huge favor. You’re going to be giving it a lot more money from not having to pay Apple all the time. But if it becomes a better competitor and it becomes more efficient at doing its stuff well, which it clearly has been in the past, then competition is likely to benefit all the participants in a more competitive world.
So, I don’t see the, whether it’s ad tech or search or being a, is that a real threat? I think the outcome is likely to be a major improvement. You know, that being said, if you look at all of the changes, so you’ve got part of DOJ’s proposal is, yes, and the headlines are we’re breaking up Chrome. But it’s divest AI, not allowed to acquire AI. That’s a big structural intervention. They also then say the preferred and simplest solution to the discrimination will be divest Android. And that’s being put in front of a judge. So, you can see the divestiture of Chrome, AI, Android and X. Now, that starts to add up to quite a significant amount of change. And I don’t know what that does to value.
JACOB PARRY: Great. We are approaching the hour. I think we can wrap up there. Thank you very much, Tim. Glad we could get you on what is a very interesting day. And thank you for your quick analysis of the DOJ’s proposals. Thank you to all the attendees who have tuned in. There’ll be a transcript of this call and I hope everyone has a good rest of their day.
TIM COWEN: Very good. Thank you very much. It’s a pleasure.
