Return of the $2 Trillion Secret Trial Against Google

It's been a long Winter, but the antitrust trial of the century (thus far) returns...

Apr 30, 2024

Over the course of 40 days last Fall, Big Tech on Trial brought you daily updates on the first major antitrust trial of the 21st Century. By the time closing arguments commence this Thursday, May 2, nearly 6 months will have passed since the conclusion of those headline-stealing proceedings. It’s the litigation equivalent of an ice bath, enough time to forget what the case was about or why it mattered, and part of a larger pattern of frustrating litigation delays. The Federal Trade Commission is pushing for a 2024 trial date in a case filed in December 2020 against Meta’s alleged social media monopoly. The FTC’s case against Amazon is expected to go to trial in 2026, three years after the filing of the FTC’s long-awaited case against the multinational conglomerate.

Compare these timelines to the Justice Department’s landmark antitrust case against Microsoft in 1998. The case against Microsoft was filed on May 18, 1998. On May 22, a trial date was set, and trial commenced 5 months later on October 18, 1998. If you put the timeline of the Microsoft case against that of the current case against Google Search, the Microsoft case had already been appealed to the DC Circuit, remanded to the lower court, and almost completely resolved before trial would have begun in the Google Search case.

It’s been a long Winter, and much has happened outside the walls of Judge Amit Mehta’s courtroom since the conclusion of trial proceedings last November. Most notably, Google is now a bona fide illegal monopolist. That’s according to the unanimous voice of the jury in Epic v. Google, which deliberated for just 4 hours to decide that Google had maintained an illegal monopoly over app distribution and in-app payments through its Google Play Store. The judge overseeing that case, Judge James Donato of the Northern District of California, is now deciding what to do about it, and there’s no reason why a remedy in that case should stop short of fundamentally restructuring the Android ecosystem. And it’s possible we’ll know what that remedy is before Judge Mehta rules on Google’s liability in the current case.

Then, on March 21, 2024, the Justice Department filed a sweeping antitrust lawsuit against Apple, alleging its monopolization of the smartphone and premium smartphone markets. The complaint against Apple is replete with references to Apple’s exclusive default search agreement with Google, which the Justice Department claims has allowed Apple to “wrap itself in a cloak of privacy, security, and consumer preferences” while boxing out Google’s more secure and higher quality search rivals. As we inch toward a ruling in the Google Search case, the Justice Department is doubling down on this same troubling fact pattern in its separate litigation against Apple.

Protracted litigation delays have real consequences beyond the inevitable cynicism that flows from delayed justice. While we wait for Judge Mehta to decide whether Google’s $26 billion per year (at least) agreement to secure default search on Apple devices is legal, Apple and Google are moving forward with talks to embed Google’s Gemini AI in its next iPhone update. It’s as clear an indication as any that the Google Search case is about the future of AI innovation, too. Despite a looming smack-down by a federal court, Apple and Google are conducting business as usual.

Adding further salt to a festering wound, Google held its quarterly earnings call last week where it announced $80.5 billion in quarterly revenue, up 15% from the year prior. According to CEO Sundar Pichai, the star witness in the Google Search case, that growth is based in part on the strength of its Search business. Which may or may not be a legal business model. We don’t know yet. But if it is, the delay is only serving Google.

Time is on the side of the monopolist.

A TALE OF TWO GOOGLES

That brings us up to date. Starting this Thursday, the Justice Department and Google will present their respective cases, adhering to a meticulously structured agenda spanning two full days in court. There will also be a public audio line for closing arguments, which would have been nice (!) to have for the duration of the trial. We’ll hear two very different stories about roughly the same set of facts. And all eyes will be on Judge Amit Mehta.

There are a few ways to get up to speed on what the parties will say this Thursday. You could go back and browse our archive of daily trial summaries, which provide most of the facts, legal theories and musings you’ll need. The other way to get up to speed is read the parties’ post-trial briefs. The government’s brief is here, and Google’s is here. If you want to go (much, much) deeper, the parties have also filed proposed findings of fact and conclusions of law.

But the larger point is we already know what the parties will be arguing. Of course, there are important outstanding issues of secrecy and the looming question of what to do about Google’s deletion of litigation records (an issue that Judge Mehta has stalled on, while counterparts in other jurisdictions have been quick to sanction Google for the same conduct.) What will make this week’s closing arguments interesting is how Judge Mehta engages with those arguments.

We can expect that Judge Mehta has reviewed all of the post-trial briefings. He may have even started drafting an opinion. That’s not to say Mehta has made up his mind. Rather, it means that any comments or questions will speak volumes as to where his mind is at. And in this case, Judge Mehta is not just a judge, but the jury and executioner, too.

Let’s recap what we can expect…

The Government’s Case

The overarching theme of the government’s case is that Google has subverted the competitive process to “freeze the ecosystem” of what should otherwise be an innovative and competitive industry.

There are several components to the government’s case, beginning with proving that it has defined appropriate relevant markets. Relevant markets are defined to capture where competition actually exists - or where it should exist - in terms of both product competition and where geographically that competition occurs. (The geographic market in this case is the entire United States and is not disputed.) Defined too narrowly, relevant markets can ignore actual competitive pressures that constrain an alleged monopolist’s conduct. Defined too broadly, and they excuse a firm’s actual market power, incorporating “rivals” that don’t actually compete with the alleged monopolist. The scope of these markets is hotly contested in this case.

The government defines three relevant markets: General Search, Search Advertising, and Text Ads. For all three markets, the government cites unique product characteristics, unique use cases, and industry recognition (including from Google’s own executives) that the proposed markets reflect the actual contours of market competition. The government argues that General Search includes services like Google, Bing, Yahoo, and DuckDuckGo, which actually compete with each other because they are interchangeable. Google argues that users also conduct search queries on “Specialized Vertical Providers,” like Amazon or Yelp, and social networks like Facebook and TikTok, and that because these other forms of search exert competitive pressure on Google, the government’s proposed market is too narrow. This is a threshold dispute that we’ll hear about early on.

Next, the government argues that Google possesses monopoly power in each of the relevant markets. As of 2020, Google’s market share in general search services was 89%, greatly exceeding even the more conservative benchmarks for identifying market power. Google has maintained its market power through exclusive distribution contracts with device manufacturers like Samsung and Motorola, carriers like AT&T, Verizon, T-Mobile and Spring, and of course its mammoth exclusive distribution agreement with Apple.

Google’s exclusive distribution agreements are central to the government’s case. And whether they are truly exclusive is hotly disputed. Early in the proceedings, we learned about the “power of defaults,” a term coined by Google’s chief economist Hal Varian over two decades ago. The maxim provides that the reason defaults are so powerful is because users are generally unaware of their ability to switch to other search engines, and even if users are aware, they are hobbled in their attempts to do so. Some of the most compelling testimony in support of the government’s case came from behavioral economist Antonio Rangel, who spoke to the bias generated by defaults and the technological barriers they create to competition. The government argues that these defaults are the most important and efficient method of distributing Google’s search engine, and that’s why Google doles out billions of dollars per year to secure them.

The government points to other barriers to entry that protect Google’s search monopoly. So-called “economies of scale,” which enable search engines to improve search results based on troves of user queries and click-through behavior, fortify a barrier to entry for rivals by locking up data necessary for rivals to achieve competitive quality. Further, the government points to the enormous capital costs required to bring a rival search engine to market, citing Google’s own former heard of search, John Giannandrea, who acknowledged that building a competitive search engine is “very expensive.” To wit, Microsoft has invested $100 billion in Bing over the past two decades, but has largely been locked out by Google’s control of key distribution channels.

The government argues that Google has generated persistently high profits, capturing significant surplus from its distribution deals, while experiencing little to no market response to search quality reductions and advertising price increases. At trial, we heard about Google’s ability to manipulate ad auctions to extract additional revenue from advertisers, and that advertisers accept the price increases because they have no alternative for replacing the volume of Google’s ad market monopoly.

Throughout its post-trial brief, the government argues that Google’s exclusionary conduct has harmed competition in each relevant market. Google has deprived its rivals of the scale necessary to create a competitive search engine. Further, its contract with Apple has discouraged Apple from entering the market itself. At trial, Apple’s lead negotiator, Eddie Cue, testified that if Apple had not received the massive payments it sought from Google, Apple would have developed its own search engine. Instead, Apple accepts billions each year in revenue share payments in exchange for Apple’s agreement to route search traffic to Google.

Google’s maintenance of its monopolies have harmed users, too. Here, the government argues that by foreclosing competition, Google’s conduct has reduced its incentive to innovate, invest in, and improve its own products. When Google invests less in search quality, users receive slower and less accurate results than they would in a competitive market. The result is increased latency, decreased privacy, and eroding search quality.

Antitrust law being what it is (or what it has become, over time), the government can successfully meet its burden of proof on all of the above and nevertheless lose if it fails to account for rebut alleged “procompetitive justifications.” The burden shifting goes like this: Once the government makes its affirmative “prima facie” case, Google will argue that it had non-pretextual procompetitive justifications for its conduct. If the court agrees, and the government is unable to demonstrate that Google could have accomplished its goals through narrower means, Judge Mehta will weigh the anticompetitive harms against Google’s procompetitive justifications.

If Judge Mehta is put in the position of balancing anticompetitive harms and procompetitive justifications, the result could very well be a line call. It might feel like a coin toss. It’s an aspect of the law that undermines predictable outcomes, and places more focus on how Judge Mehta will exercise discretion in the case at hand.

On that note, let’s talk about Google’s version of events.

Google’s Case

Google’s overarching theory of the case is that Google has faced extraordinary competition from a wide array of online rivals - more than the government gives it credit for - which has fueled its “relentless commitment” to improving Google Search. Far from excluding competition, it has won the competition fair and square.

Google begins with an interrogation of the government’s proposed relevant market. Here, Google claims that the government has defined the market too narrowly, excluding actual competitors like Amazon, Facebook, AirBnB, Yelp, and several others. Contrary to the government’s characterization of a frozen market dominated by Google’s search monopoly, Google characterizes online search and digital advertising as “incredibly dynamic spaces with seemingly endless potential for growth.” The competitive pressures from SVPs and social media platforms exert competitive pressure on Google to constantly better serve its users and advertisers.

To make its case, Google distinguishes between general search queries and shopping queries, where Google claims to face greater competition from Amazon and Yelp than it does from a general search engine like Microsoft’s Bing. Google characterizes social media channels like Facebook, Instagram, Pinterest and TikTok as increasingly popular places for users to turn for their information needs, and growing targets for advertisers, too. For Judge Mehta, the question is whether users actually actually substitute any of these other sources of information for Google’s general search engine. According to Google, the governmnet’s construction of markets, even if based on Google’s own characterization of those markets, cannot be squared with commercial realities.

Even if the government’s relevant markets are valid, Google contends that the government has failed to demonstrate that Google possesses monopoly power in any of them. In support of this claim, Google points to a steady increase in the number of general search engine queries as evidence of its “relentless pursuit” of improvements to its search engine. Further, Google contends that the government has failed to prove that the quality of their search engine has dropped, or that ad prices have increased. Where the government has presented evidence of Google’s ability to raise ad prices without losing market share, Google counters that prices have only increased as quality has improved.

Google further argues that any barriers to entry are insignificant, are falling rapidly, and that rival search engines can compete with only a small share of the alleged market. Google points to DuckDuckGo as one profitable alternative (even though it has a less than 5% share of the search market.) Google also points to Neeva, the upstart search engine started by former Google executive Sridhar Ramaswamy. (Of course, Neeva famously threw in the towel when it failed to gain enough traction in the market.) But what matters to Google’s case is that Neeva was able to enter and innovate, and they did so successfully… until they no longer could.

Google then argues that its default search agreements are not “exclusive,” because they don’t prevent customers from accessing rival search engines. This is Google’s rebuttal to the government’s evidence (and their own chief economist’s characterization) of “the power of defaults.” In Google’s version of the case, users can capably switch to rival search engines. To the extent they don’t, that’s because users actually prefer Google’s product, as demonstrated by Google’s high share of the search usage on non-default devices. The government will argue that this is also a figment of Google’s high brand name recognition, which it has secured over time through its various exclusionary actions.

Where the government claims that Google’s default search agreements prevent rivals from competing, Google argues that there is actually robust competition for those default search agreements, and that Google can’t be faulted for winning that competition. Google argues that rivals can compete for 100% of all queries, first by competing to be the default, and second, by competing for users even if they fail to win the default.

It should come as no surprise that Google dedicates significant analysis to one of its chief would-be competitors: Microsoft Bing. Where the government argues that Google has deprived rivals of economies of scale, in the form of search query data necessary to improve any search engine, Google points to Microsoft’s 2009 agreement with Yahoo as evidence that Microsoft had access to data sufficient to improve its search engine, but nevertheless failed to compete on other metrics. As we have written, an open dispute in this case is whether Google’s search engine is superior because of that access to data, or because Google has gone above and beyond to create a superior search engine irrespective of that data.

Making the case about Microsoft serves Google in another key way. Not only does Google point to Microsoft’s Bing as an inferior search engine that has been unable to compete for default placement in Apple’s Safari browser, they are also hoping to persuade Judge Mehta that this case is not really about competition, but about a single competitor. Antitrust, after all, protects competition, not competitors. And Google has won the competition fair and square.

Finally, Google argues that its conduct is supported by legitimate business justifications that generate procompetitive benefits - even outside the market for general search. Google argues that default search engines are a feature of browser design that are not driven by Google itself. The very fact of a default search engine is a benefit to consumers, not a tool of exclusion. Google points to testimony from both Apple and Mozilla executives that they chose to employ defaults because those defaults best serve the interests of their consumers. And they chose Google Search because it is a superior engine. According to Google, default agreements improve browsers, encourage users to user the web more, and increase overall search output.

By Google’s estimation, it’s a win-win, and the beneficiaries of its revenue share agreements aren’t complaining about the billions of dollars they receive to secure those distribution channels. But this is where the procompetitive justifications analysis is on a fundamentally shakier legal footing. Instead of pointing to benefits within the proposed relevant markets, Google points to benefits in outside markets like browsers and smartphones, too.

The government argues that consideration of these out-of-market benefits contradicts the plain meaning of Section 2 of the Sherman Act, and that, from Eastman Kodak to Microsoft, the law actually precludes consideration of outside market benefits, or at least leaves the issue unresolved. This unsettled legal question is one that I’ve been thinking about from the outset, and could be an area where Judge Mehta pushes the law into a realm of much broader implications.

This is, of course, an abbreviated version of the facts and legal arguments at play. I’ve not touched on the States’ case here, which is important, too, but we’ve written about some of the overlap between the cases already.

I’ve also done my best to reserve my own personal judgment, although it’s probably not hard to deduce that I think the government has demonstrated a powerful prima facie case. I invite you to disagree with me in the comments, or, more amicably, to share aspects of the case that you’re watching closely.

And remember…

You Can Tune In To Closing Arguments, Too.

I’ll write later in the week about what we learned from closing arguments, but you can tune in, too.

While it’s great that there’s an audio line for closing arguments, it begs the question why there wasn’t an audio line all along. We’ve been arguing for greater public access throughout this trial, often bumping against the Court’s institutional allergy to allowing that access. Ultimately, these decisions rest with Judge Mehta, who has been far from consistent with respect to issues of courtroom transparency.

There’s a direct correlation between courtroom transparency (and public trials, generally) and the interest in this trial, which may fundamentally restructure the internet and tech innovation as we know it.

Remote Public Access Line:

Toll Free Number: 877-848-7030
Access Code: 3218747

May 2, 2024: Violation of Section 2 of the Sherman Act in the Alleged Market for General Search

Market Definition and Market Power (9:00 a.m.–10:30 a.m.)
1. Plaintiffs: 40 minutes.1
2. Google: 40 minutes.
3. Plaintiffs’ rebuttal: 10 minutes.
4. Break: 15 minutes.
Prima Facie Case (Anticompetitive Effects) (10:45 a.m.–1:00 p.m.)
1. Plaintiffs: 60 minutes.
2. Break: 5 minutes.
3. Google: 60 minutes.
4. Plaintiffs’ rebuttal: 10 minutes.
Lunch (1:00 p.m.–2:00 p.m.)
Procompetitive Benefits & Balancing (2:00 p.m.–4:05 p.m.)
1. Plaintiffs: 50 minutes.
2. Google: 50 minutes.
3. Break: 15 minutes.
4. Plaintiffs’ rebuttal: 10 minutes.
Open Summation (4:05 p.m.–4:50 p.m.)2
1. Plaintiffs: 20 minutes.
2. Google: 20 minutes.
3. Plaintiffs’ rebuttal: 5 minutes.

May 3, 2024: Violation of Section 2 of the Sherman Act in the Alleged Markets for Search Advertising

Market Definition and Monopoly Power (9:00 a.m.–10:40 a.m.)
1. Plaintiffs: 45 minutes.
2. Google: 45 minutes.
3. Plaintiffs’ rebuttal: 10 minutes.
4. iv. Break: 10 minutes.
Prima Facie Case (Anticompetitive Effects) (10:50 a.m.–12:30 p.m.)
1. i. Plaintiffs: 45 minutes.
2. ii. Google: 45 minutes.
3. iii. Plaintiffs’ rebuttal: 10 minutes.
Lunch (12:30 p.m.–1:30 p.m.)
Procompetitive Benefits & Balancing (1:30 p.m.–2:50 p.m.)
1. Plaintiffs: 35 minutes.
2. Google: 35 minutes.
3. Plaintiffs’ rebuttal: 10 minutes.
4. Break: 15 minutes.
SA360 and “Duty to Deal” (3:05 p.m.–4:15 p.m.)
1. Plaintiff States: 30 minutes.
2. Google: 30 minutes.
3. Plaintiff States’ rebuttal: 10 minutes.
Sanctions (4:15 p.m.–4:40 p.m.)
1. Plaintiffs: 10 minutes.
2. Google: 10 minutes.
3. Plaintiffs’ rebuttal: 5 minutes.
4. Break: 10 minutes.
Open Summation (4:50 p.m.–5:25 p.m.)
1. Plaintiffs: 15 minutes.
2. Google: 15 minutes.
3. Plaintiffs’ rebuttal: 5 minutes.

That’s all for now. Thanks for reading!