Jonathan Kanter on the case for breaking up Google

Nilay Patel

The search case as well. So let's just start at the beginning. What's this case? Why is it different than the search case?

Jonathan Kanter

So the search case is the Google that most consumers know, which is you go google.com or your search bar or your whatever, and you put in a search and get an answer back, right? That case focused on Google's core product, which is a search monopoly. This case focuses on all the interactions a lot of people have with Google on a daily basis, but don't realize it. So when you go out to the internet, any site, and you see an ad, chances are very high that that ad is passing through Google technology. Ads are bought and sold on the internet using Google's pipes and plumbing. And so that's what this case is about. It's all that infrastructure, the stuff you don't see, that allows publishers, news publishers, or anyone who creates content to sell an ad space and anyone who wants to buy an ad to Buy ad space. Now, when people think ads, they think madmen, right? They think the old style, you know, pouring a glass of whiskey and taking someone out to buy some airtime. That's really not how ads work in the modern world. How they work in the modern world really is more like a commodities or a securities exchange, equities, where you buy and sell programmatically on these open market exchanges and people Bid in their auctions. That's how ads are sold. And this case is about exactly that technology.

Nilay Patel

So there's a lot of complicated components to that auction. The way that I always think about it is there's a lot of different software products that interact at every stage of that auction. And most of those markets where those products play, Google is the one or two player, right? It just has a dominant position across every piece of technology you need to make an ad auction go. How did you put this case together? How did you see that this was a problem? Where did it come from? And how did you know, okay, this is one we can go win? This is years in the making, but the facts here are incredibly strong.

Jonathan Kanter

So I think when you get out of the madmen and into the kind of commodities exchange mindset, the case becomes very clear. You have the buy side and the sell side, right? The tools that someone, a publisher, will use to sell ads, the tools that an advertiser will use to bid and buy on ads. And then you have this exchange in the middle. What we found in investigating is that Google has a dominant position on the buy side, the tools that buyers use. They have a dominant position on the tools that sellers use. They own and have the dominant exchange in the middle. And then Google buys and sells on its own exchange. And then they rig the results of those auctions so they win more and competitors lose less and publishers get less money as a result.

Nilay Patel

So those tools in the middle, that's the open web. What you're describing here is the open web, right? Exactly. You run a website. You put some ad slots on it. You buy some software products to auction off that inventory. Someone shows up to use some other tools to buy it. That's a smaller piece of Google's revenue today than it was before. The part that really interested me about this whole case is the connection to search, which is under all kinds of pressure, not just the antitrust case that Google now has to defend in The remedies phase, but also AI. Also, just everyone can see what SEO is doing to the web overall. And so you see the web is kind of under all this pressure. It's under revenue pressure from Google as well because of all the monopolization in the ad tech space you're describing. But then Google itself is making more money on its own platforms, on search itself, on YouTube. Was there ever a sense that you were chasing a thing that was dying all by itself? Not at all.

Jonathan Kanter

Because that felt to me like the hardest part of the puzzle here. First and foremost, right, this technology matters a lot, particularly to news publishers. And news publishers are critical for the free flow of information, which is vital to democracy. And I think that industry has been under assault for years, in part because the revenue streams have dried up in no small part due to Google's conduct, in my view, in the view of the court Here. So I think that is an important starting point. Second is, I think ahead to AI as well, right? And Google really, at its heart, is an advertising company. It makes its money on advertising. And it puts out all these products, some of which are really good and interesting to attract users in order to get their attention, in order to get their data, so then they can flip it to Advertisers and sell for money. When you think about where AI is going for a company like Google, that is likely to be how it monetizes a lot of its AI business, certainly its consumer AI business. And the infrastructure that we're talking about here is likely to be the same kind of infrastructure, whether it's this advertising infrastructure from the search case or the ad tech Infrastructure from the ad tech case that translates into how Google sells advertising alongside its various consumer-based AI products. And so getting this right now wasn't just critical to fix these industries as they exist today, but was to make it clear that they can't use these same anti-competitive practices and

Nilay Patel

Playbook in the future. Let me push on that just a little bit harder. I hear that. The nihilist in me says, well, it's already too late. The web has been scraped by all of the model makers fully and completely. The economics of starting a new website are pretty bad. Most new information happens on a closed platform. Will a case like this combined with the search case result in the economics of publishing on the web being better than they are today?

Jonathan Kanter

Well, they will hopefully result in new businesses being funded and entering the market to figure out how to monetize content, monetize AI, and sell advertising to advertisers and Help publishers figure out better ways to maximize their revenue. Right now, the system wasn't working, right? The companies like Google were not only extracting all the data from these sites through calling and indexing and training their models, but they were extracting massive amounts Of inventory through these intermediaries, these middlemen that they created to buy and sell and exchange ads. And so we need something better because what we have now doesn't work. And so either we can roll over and say, well, we'll just throw our hands up. The market's moved on to something else if they're going to dominate and use the same playbook. We can fight. We can say what they did was wrong, broke the law, and they shouldn't do it again in the future.

Nilay Patel

One of the things that really struck me about the search case was Google's pretty compelling argument that, yeah, the market might be distorted in favor of Google search, but everything Else is bad. Like Bing is just bad. There are other competitors that mostly use the Bing index. It's just bad. Sachin Della took the stand and said, I could make Bing better, but it wouldn't be worth it. And the real competition, as you and I discussed, was maybe from these AI products that might disrupt search in some other way. And then the real problem was by paying Apple for default placement, no one else could ever get the volume of search clicks to be better. But at the end of the day, Google's core argument was, look, our product is just better. Like, people like it. If we didn't pay Apple, people would just choose Google by default. We might as well get some money. They lost on that argument, but it was very intuitive the way that they fought back, right? It was intuitive to say, look,

Nilay Patel

The attorney general for antitrust. You decide, okay, we see some conduct. We've heard from a lot of publishers that they're mad. We can see the distortion in the search market. We're going to do both, right? Like we're going to chase both of these down. How did you decide to prioritize these cases? And then I want to know how you decided to put them together because no one ever really gets that view. These are really hard cases, right?

Jonathan Kanter

And so what you do is let me start kind of take a step back. For the investigation, you talk to the industry. You interview people. You talk, you learn about the products. You understand how they work. And then you send out subpoenas and you get information and documents for Google. And we found incredible stuff where Google knew that what it was doing was wrong. It talked about how owning this publisher server and these technologies and these ad exchange would be like if Goldman Sachs owned the New York Stock Exchange. That's an actual document that they wrote that we featured in our case and actually made it into the opinion. And then you put all that evidence together and then you realize that the impact here is great. Advertising, internet advertising is huge business, right? This is not a side hustle, right? You think about from the advertiser's perspective, advertising is one of the largest, if sometimes the largest, costs for most small businesses, in particular, small and medium-sized Businesses who rely on internet advertising for lead gen and customer generation. Rely on online advertising in order to get people coming in through the door. People who create content for the masses, whether they're news publishers or any other type of content who doesn't want to sell exclusively by subscription, needs to sell advertising. In the internet world, you can't sell advertising the madman way and make money. You have to have technology and tools in order to clear all those impressions. And so Google understood early on that it could gobble up through acquisition all of the components to give itself power in those tools. And then it could use market manipulation, exclusivity, and other types of course of behavior in order to lock itself in. And that's exactly what the court found.

Nilay Patel

So put me in the room. Is it like billions? Are you screaming, we're going to go get Google? Is a presentation from your staff that says, we think we have a case here. Here's how we're going to make it. Was there a PowerPoint? How does this work? It's extremely deliberate.

Jonathan Kanter

So the staff will investigate. We will, in my office, stay in tune and line with the staff and get updates on those investigations, make sure that we're sending subpoenas and interview requests out to everyone in The industry. We hire experts, get data, we analyze that data. It's expansive and very expensive, actually. And then when we get closer to thinking there's a problem, we start figuring out, okay, how do we tell this story? And does it survive scrutiny? And so we put ourselves through a very rigorous red theme style exercise to make sure that we're right. The power of the government is extraordinarily significant and we don't wield it lightly. We want to make sure that we're doing the right thing for the right reasons. And so we want to make sure it's going to hold up in the court of law that we think we're going to win and that it's the right case for the right reasons. And then you have to figure out, OK, how are we going to litigate this case? How are we going to write it in a way that, especially in a market like this, is filled with acronyms. It's incredibly complicated. Things move at the speed of milliseconds in terms of data. And so how do we tell that story in a way that's accessible? And we took a very interesting path. And I think this was consistent with our approach in a lot of these antitrust cases. Some lawyers will say you want to write a tight, short, crisp complaint. I actually took a different view. My view is we have all of this evidence. Let's lay it all out there for the public and the court to see at the outset. And so let's err on the side of saying more rather than less, but let's almost have three different stories in one. And so the way we would structure our complaints would be we have a very short intro, page or two, that gives you the thematic opening to the case, the stuff that if you're going to read One thing and you need to quickly understand what the case is about, you can read that one or two pages and understand it. And in our case, we put it forward with that Goldman Sachs quote, because I think that really told the story. Then you have a longer executive summary, right? And this is for the person who's got a little more time, is willing to read 15 pages, wants to get some more detail on what are the relevant markets and what are some of the key documents And what's some of the harm, right? Then you have like the antitrust nerds like you who want to read the whole thing from start to finish. And that's the extra, the additional 80 or 90 or 120 pages, whatever it might be in the case, where you go really deep into the weeds and you can understand how the industry functions, Where the documents are, what Google understood it was doing at the time, why it knew that it was behaving in an anti-competitive fashion, what the harm was, and how competitors were Responding. And then we position all of that to the court so that we can go to the court and say, this is not a theory. This is not just based

Jonathan Kanter

So first and foremost, you have to invest in the right team. And so when I started the antitrust division, I realized that we weren't going to be successful unless we can convince courts and juries and judges that we're right. And we had the best, in my view, and still do, antitrust lawyers in the world, but many of them hadn't had the experience of standing up in a courtroom and trying cases. And so the vision that we brought to the antitrust division when I arrived was, how do we take world-class antitrust lawyers, match them with world-class trial lawyers? And so we went out, we built this kind of elite core of first-chair trial lawyers, about 25. These are people who have grown up in courtrooms, former U.S. Attorneys, assistant U.S. Attorneys and DAs and plaintiffs' lawyers and partners at law firms. How do we assemble folks who really can be that lead lawyer in first and second and third chairs alongside the antitrust experts? And that if you combine those two expertise, you can bring the sophistication and you can bring the storytelling and in-court skill sets. And so we built that and we brought in incredible people. It was run by a woman named Hedal Doshi, who is one of my deputies and is just an extraordinary gifted storyteller and lawyer and debater and mastermind in the courtroom. And then we assembled this incredible team of what we called Senior Litigation Counsel. And so I actually recruited a former partner of mine from private big law, a woman named Julia Wood, who ended up becoming the leading trial lawyer in the case. And Aaron Teitelbaum, who is a former prosecutor in the Colorado Attorney General's Office, U.S. Attorney's Office, and then a number of others to help lead the trial team and go head-to with my former firm, Paul Weiss.

Nilay Patel

It sounds to me like you're building a little law firm inside the Department of Justice, right? You're describing a culture shift, you're describing hiring and recruiting. Is that how people should think about it, that this is a little law firm that you ran?

Jonathan Kanter

Yeah, it's a scrappy little law firm, certainly compared to the size of the big law firms. And we had incredible assets when I arrived. We just needed to shape them a little bit differently. And so bringing in this trial unit, building data scientists unit, boosting the core expertise that I thought we needed in order to go head to head with companies like Google and litigate Two cases in a year and win both of them, while at the same time suing Apple and RealPage and Ticketmaster and a number of other companies. And so we needed to scale, we needed scope, and we needed depth. And how you build your team is really important. And we invested a lot in that early on before we went to battle.

Nilay Patel

I'm going to come back to this idea that you started a little law firm and it was good at its job because I think that's hard for people to see. I think they see the Department of Justice and you're describing a

Nilay Patel

On to the remedies phase now in both the search case and in the ad tech case. What do you want those remedies to be? The last time you were on, you couldn't say, but you've told me you're free now. What do you want those remedies to be?

Jonathan Kanter

I'm free. Those remedies need to be structural and forward-looking. So structural means the company has too much power. It's used that power. It's abused that power. And particularly in the case of ad tech, it has conflicts of interest. It represents buyers and sellers, the exchange, and buys and sells its own exchange. The only way to deal with that is to eliminate those conflicts of interest and to reduce its power. Terms of forward-looking, we need to confront the market where it is today in light of a decade of anti-competitive behavior. And this is true in both search and in ad tech. And the infrastructure, whether it's search infrastructure and search advertising infrastructure or the ad tech infrastructure, is going to be used now going forward not only for Current-day products, but also going forward as things into a web that includes a lot of AI-driven content and information and products. And so I think it needs to be structural. It needs to be definitive. It needs to be forward-looking. It needs to look, again, chase, go to, escape to where the puck is going, which in this world is a more AI-driven marketplace that still relies on a lot of the same infrastructure just For different use cases. Do you think Google should be broken up? Yes. How would you break up Google? In these cases, the Department of Justice has asked for Google to sell its Chrome browser, which was critical to the finding in the search case. And in the context of ad tech, I think, again, it should not be able to own the buy side and sell side and exchange all at the same time.

Jonathan Kanter

Worth a lot more on their own? Do you think there's a Google left if you peel off all those pieces?

Nilay Patel

Again, I keep coming back to the open web. The open web right now relies on Google. It's the number one referrer of traffic to different websites. It is the number one monetization engine for all of those websites. Chrome is the dominant browser. You break that company up into a hundred little pieces. What's left? Does the web persist or does it all just become AI training data?

Jonathan Kanter

Well, I think the web may sadly be turning into AI training data regardless of whether there are antitrust remedies. The fact of the matter is, you know, Google is going to continue to exist after this case and it's going to continue to be a really significant company, just as AT&T continued after its Breakup. Well, note that AT&T reformed itself, like the Terminator.

Jonathan Kanter

Been able to exist because Google is dominant in the places where it's dominant? Absolutely. So to me, antitrust is most important when you reach inflection points in the market. So the AT&T case preceded the kind of emergence of the PC. There's an IBM antitrust case that preceded the emergence of software and operating systems. The Microsoft case immediately preceded the internet era and the launch of Google. I think what enforcers and the antitrust lawyers are concerned about is that incumbents seeing these new threats and these new inflection points coming will use their power to slow Them down and make sure that they can turn their dominance in market A into market B. They can incumbent advantage into this new technology. But that's not what you want, right? What you want as an enforcer is you want these new nimble companies to emerge without the kind of lethargy of a big monopoly aircraft carrier, and you want someone who can move quickly And decisively. And so if Microsoft had won its antitrust case, who knows what would have happened? Perhaps it could have owned browsers. Perhaps it could have owned, you know, search. It could have owned all these businesses. But instead, companies like Google and others were allowed to thrive. It's not lost on us, and we put this in our Apple case that, you know, iTunes and the iPod, which was really the precursor to the iPhone, really got traction when it went from being a Mac-only Product to a Windows product. And when you can use your iPod at the time with your Windows computer because iTunes was available, that was something that was made possible by the Microsoft consent decree that involved Competing media players being on Windows, right? And so all of these things, you know, interact with one another. And so we're sitting here now on the dawn of this new era. You had like kind of web 1.0 and 2.0, and now we're kind of moving to the AI-based web where, you know, you interact with agents and, you know, all other cool stuff that you guys talk about All the time. And we want to make sure that those new companies, those little tech companies, those startups, those other incumbents in other areas who see as an opportunity to diversify can come In here and invent new technologies and compete on the merits rather than running into obstacles. And if you look at both Google cases, the obstacles were man-made, kind of like our economic problems today. They're not due to better innovation. They're due to contractual restrictions. It's Google paying Apple $20 billion to lock up defaults or Google tying products and services together or giving itself first look or last look at auctions, which is what we proved In the Antec case, rather than having an open auction where anybody can win, right? It's putting sand in the gears or bumps on the road to slow other people down rather than speeding themselves up to innovate in order to compete. So I