All the Hottest Parts of Google’s Antitrust Ruling

The opinion in the Google search antitrust case, published on Monday, is extremely long. As it was a bench trial, Justice Amit Mehta was in charge of making factual as well as legal findings. Thus, there are more than a hundred pages of factual findings and even more legal conclusions, totaling a 286-page document full of footnotes, essays and even a graphic illustrating a search result for “golf shorts” ( the which apparently came up a lot at trial).

The decision in United States x Google It’s a lot to absorb. Some of this was previously reported in the press during the weeks-long trial; But here, the judge inadvertently compiled the trial’s greatest hits: snide quotes from executives, embarrassing internal studies, and a bunch of surprising details about that multibillion-dollar contract that keeps Google as the default search engine on Safari.

Google pays Apple billions of dollars a year to be the default search engine on Safari. But according to Eddy Cue, Apple’s senior vice president of services, there is no other meaningful alternative. During the trial, he said there is “no price Microsoft can offer” to Apple for the company to preload Bing in Safari.

“I don’t believe there is a price in the world that Microsoft can offer us,” Cue said at another point. “They offered to give us Bing for free. They could give us the whole company.”

For Google, this is a sign that they have earned their default status (which, by the way, they pay Apple a lot of money to maintain). Judge Mehta says this is an indication that “the market reality is that Google is the only real choice as the default GSE [general search engine].”

(Of course, Cue’s opinion doesn’t mean that Bing is objectively bad. Elsewhere, opinion notes that Bing’s search quality is comparable to Google’s on desktop, though it lags behind on mobile.)

In addition to Apple, Google also has contracts with wireless carriers and device manufacturers to be the default search engine on Android devices (these contracts work a little differently, as they depend on Google’s control over the Google Play Store).

It’s not just Eddy Cue that refuses to pay attention to Bing – all of these companies recognize Google as the only game in town. None of these “Fortune 500 companies” have a real choice in this matter.

“Google understands that there is no genuine competition for defaults because it knows that its partners cannot afford to go elsewhere,” the judge writes. “Time and time again, Google’s partners have concluded that it is financially infeasible to change the default GSEs or seek greater flexibility in search offerings because doing so would mean sacrificing hundreds of millions, if not billions, of dollars that Google pays them in revenue share. ”

According to the opinion, “[i]In exchange for exclusive and non-exclusive default placements (i.e. user-downloaded default Chrome and Safari bookmarks), Google pays Apple a [redacted] percentage of its net advertising revenue, which totaled $20 billion in 2022.”

Apparently, this is “nearly double the payment Google made in 2020, which at the time represented 17.5% of Apple’s operating profit.”

Google and Apple signed their current agreement in 2016. Their negotiations go back much further, but around that time Apple launched Suggestions. (Think, for example, when you type something into Spotlight and Apple suggests a website for you — that’s not the same as Google Search.)

This was significant. A Google analysis estimated “a query loss of 10-15% of Safari traffic and a revenue loss of 4-10% of iOS Safari revenue based on Apple’s suggestions.” The new 2016 contract includes a specification that “Apple’s implementation of the Safari standard must ‘remain substantially similar’ to prior implementations” so that Apple “could not expand beyond what it was doing” so that Apple would not “suck traffic.”

These days, when it comes to iPhones specifically, “Google receives nearly 95% of all general search queries.”

The terms of the 2016 contract appear to have worked for both companies. Google and Apple extended the agreement in 2021: the contract will expire in 2026. Apple “may unilaterally extend the agreement for two years” and, if both parties agree, they can extend the contract even further, until 2031. Part of the contract obliges both Google and Apple to defend this agreement “in response to regulatory actions” (e.g., DOJ antitrust lawsuits like this one).

According to the judge, it’s not just that Google pays Apple not to challenge its search supremacy – it would be incredibly difficult for Apple to take action. Not surprisingly, both Google and Apple looked into this and their own internal estimates were released at trial.

Apparently, Apple calculated that “it would cost $6 billion annually (on top of what it already spends developing research capabilities) to run a GSE.” Meanwhile, “in late 2020, Google estimated how much it would cost Apple to create and maintain a GSE that could compete with Google.” Apple would have to spend something “in the approximate order of” US$20 billion to “reproduce [Google’s technical] infrastructure dedicated to research.”

And neither Amazon nor Meta.

Firstly, United States x Google draws a distinction between general search engines (GSEs) and specialized vertical providers (SVPs). The heavy use of technical acronyms may make your eyes water, but the essence is actually very simple. A GSE is a search engine in the sense that everyone understands it – Google, Bing, DuckDuckGo and so on.

If you get really worried about this, there are thousands of little “search” boxes all over the Internet. Sometimes you even use them in a similar way to Google Search – say, for example, to look for cheap flights to a specific destination or to buy a pair of baggy black leggings. However, Booking.com and Amazon.com are simply not the same as a general search engine that indexes the World Wide Web. Do you, the average person, need to logically justify this knee-jerk reaction? No. A court has already done that for you, in a flurry of words you probably don’t need to read.

So much for SVP. But the little search bar on social media platforms like TikTok works slightly differently – at least in terms of user behavior and certainly in terms of whether Google views certain companies as competitive threats. Apparently, in 2021, Google conducted a survey on “younger users”. One of their findings: “Among ‘Gen Z’ participants (defined as participants ages 18 to 24 who use TikTok daily), 63% reported that they use TikTok as a search engine.”

However, says Justice Amit Mehta, social media platforms are distinct – they are walled gardens of content. And most importantly, “there is little evidence that they actually compete with the GSEs for search queries.” The TikTok study, he says, doesn’t look at whether the platform’s search quality results are competitive with Google’s – just because kids like TikTok doesn’t mean it’s in the same relevant market as Google Search. And TikTok isn’t the only social platform. One study, he says, suggests that Facebook usage corresponds with an increase in Google Search usage.

For Mehta, when it comes to an antitrust analysis, Zoomers’ internet habits are not relevant information. “Imagine if Google search quality were substantially degraded, whether on purpose or through negligence,” he writes. (Yes, imagine. Who. Could. Imagine. That.) “Would senior vice presidents or social media platforms be able to shift resources to launch a product that resembles a GSE and thereby capture a number significant amount of dissatisfied Google users? The answer is obviously no.” It would take “extraordinary costs and expenses” for even a juggernaut like Amazon or Meta to fill this gap in the market.

Perhaps AI research is the future, but the future is not yet here – at least, not in a way that is relevant to antitrust law. “Someday AI may fundamentally alter research, but not anytime soon,” writes the judge. Elsewhere he writes that “[c]Currently, AI cannot replace the fundamental building blocks of search, including web crawling, indexing, and ranking.”

He also discovered that – factually speaking, even – “Generative AI has not (or, at least, not yet) eliminated or materially reduced the need for user data to provide quality search results.” The opinion’s conclusions quote Neeva co-founder Sridhar Ramaswamy as saying that “the intermediate problem of figuring out which pages are most relevant to a given query in a given context still benefits greatly from the query’s click information. And it is absolutely not true that AI models eliminate this need or replace it.”

In other words, when you search for “golf shorts”, it’s not just that you get (hopefully) relevant results for golf shorts – Google more or less automatically receives important information about what do you think the relevant results are, based on the pages you click on. This feedback loop is not happening with AI chatbots.

The opinion also quotes Google’s own vice president of search, Pandu Nayak, as saying that it is vitally important that Google continues to “have an infrastructure that [it] to understand[s]” – that is, the traditional classification system. According to Nayak, “it doesn’t make sense that we have transferred our classification to these systems. We still exercise a modicum of control over what’s happening and an understanding of it.”

Apparently, in 2020, Google conducted a study to see what would happen to its bottom line if it “significantly reduced the quality of its search product.” The conclusion was that even if the company made Search worse, Search revenues would be good.

“The fact that Google makes changes to products without worrying about its users going elsewhere is something that only a company with monopoly power could do,” writes the judge.

Fundamentally, antitrust regulation is underpinned by the idea that competition is good for everyone – the market, the companies themselves, but especially the average customer. It is debatable whether “consumer harm” is still the right barometer for defining monopolistic behavior in the Internet era. Yet, United States x Google suggests that even one of the most innovative companies of the last 20 years can hurt consumers the old-fashioned way – by forcing competitors out of the race, it can deliver an increasingly worse product and still make the same money.