Google has been, in many ways, a positive force in the worlds of technology, society, convenience, and information dissemination. It’s hard to argue that any of the other tech giants (save, perhaps, Microsoft) have done as much to improve numerous aspects of the lives of people around the world.
But in building services outside their core web search business — Google Maps, Chrome, Android, YouTube, Google Hotels, the ill-fated Google+, Project Fi, and dozens more — there’s strong evidence the company leveraged their monopoly position in search to unfairly compete in other sectors.
In the comments below this post, and in the tweets and shares about it on social media, I can almost guarantee that some poor-reading-comprehension visitor will posit something to the effect of:
“Google can’t be abusing its monopoly power because the competition is just a click away!”
This argument is erroneous and misguided in three ways:
- Google, as the chart above illustrates, is absolutely, unquestionably a monopoly in web search. They have more than 90% of the market in the United States, and over 95% globally. Only in China, South Korea, and Czech Republic are those numbers lower. Three good sources back this up: Statcounter, Jumpshot, and Merkle. The old Comscore data you might find from 2009-2014 showing Google with “only” ~67% of searches is no longer accurate (and it probably never was). Tragically, in much of the early reporting on this issue, a source called “NetMarketShare” was cited as showing Google with ~70% market share. That number is horribly wrong, and it appears the reporters did not understand how to use the source they referenced (NetMarketShare also shows Google above 90% market share in the US when the correct filters are applied), nor do they include Google’s other properties (most notably Images, Maps, & YouTube).
- Lacking cost does not preclude the possibility of having a monopoly. Radio is free to listen to. Broadcast TV is free to watch. Yet both have been long subject to US government regulation and guidelines that ensure a diversity of companies hold sway over the airwaves. These are perfect corollaries to Google — they require a device to access them, but then provide (at least some) content without further charge.
- The crux of the argument isn’t whether Google is a monopoly (although it is). The DOJ isn’t trying to ascertain whether Google has dominant market share (at least, that’s our best understanding from the initial reports). Instead, they’re (probably, wisely) asking if Google’s practices violated antitrust law. As Representative David Cicilline of Rhode Island said in a news conference Monday, “This is about how do we get competition back in this space.”
So… What are the United States’ Antitrust Laws? I know, I know. You want the short version. OK friends, I’ll summarize the official FTC guidelines (but they’re honestly pretty easy to read and grasp on your own):
- The Sherman Act outlaws “every contract, combination, or conspiracy in restraint of trade,” and any “monopolization, attempted monopolization, or conspiracy or combination to monopolize.” It doesn’t outlaw all of these, just those that are “unreasonable,” or “so harmful to competition that they are almost always illegal.” Things like price-fixing between competitors or dividing up markets and agreeing not to compete would fall under this. For example, when Apple, Intel, Google, and Adobe formed secret non-poaching agreements with each other, the courts gave them tiny slaps on the wrist. Most of the $450mm went to lawyers. ~66,000 tech employees got the equivalent of a ten-year-old, used car.
- The Federal Trade Commission Act bans “unfair methods of competition” and “unfair or deceptive acts or practices.” When Microsoft used its near monopoly on operating systems in the 1990’s to distribute Internet Explorer, crush Netscape, and become the dominant web browser, the government sued the company and eventually won a long, drawn-out, and at the time unpopular battle. If this reminds you of Google’s requirement that every Android installation come with the Google Search App and Chrome pre-installed (you know, the one that cost them a 4.34B EU fine last year), you’re not alone my friend.
- The Clayton Act addresses practices like mergers and acquisitions that could create competition-unfriendly environments or harm consumers. It’s the one that requires big companies to notify the government of these types of transactions (like how Sprint & T-Mobile are waiting for regulator approval to merge). And it prohibits individuals from holding too much sway over multiple, potential competitors in a market (as when the FTC’s 2009 investigation resulted in then-Google-CEO Eric Schmidt stepping down from Apple’s board of directors).
Perhaps even more important than understanding anti-trust law as its been in the past is understanding why it exists — because that speaks to its potential future.
Long story short, economists, regulators, and those who study capitalism understand that unfettered market forces are immensely dangerous to a functional society. A significant part of that danger comes when a powerful corporation with market dominance leverages that dominance in one field to unfairly compete in others. That unfair competition could even benefit consumers in the short term (e.g. by lowering prices or giving a great service), but if the dominant player drives away competition and becomes free to compete, unfettered by oversight or other players, consumers suffer and often, society at large does, too. See examples like Kodak offering film only they could develop, the five big Hollywood film studios owning theater chains and controlling film distribution, or Apple’s attempt to fix prices on e-books to compete against Amazon’s Kindle.
As Michael Bloomberg said this week “I’m as much of a capitalist as you will ever find, but anyone who believes that unfettered capitalism works hasn’t read history.”
So what has Google done that could be considered “unfair competition” and where should the Dept. of Justice investigate?
It’s my opinion that the strongest cases for antitrust against Google rest in the following three areas:
- When Google’s search engine has made highly visible portions of its search results accessible exclusively to its own properties, often before those properties are the clear, obvious, or only high quality result.
- When Google uses their search engine to aggregate data from multiple third-party publishers, then presents it to search consumers as though it were created by Google alone, without any attribution or credit.
- When Google used its Android operating system and its dominant search engine to distribute its Google Chrome browser (a practice that continues, albeit with less forcefulness now that Google’s, arguably, won the browser share market).
I’ll go through each of these in a little more depth.
#1: Google Gives SERP Real Estate Exclusively to Its Own Properties
On July 17th, 2014, my now cofounder at SparkToro, Casey Henry, worked at the video software company, Wistia. He authored a blog post that day sharing bad news — that the video snippets long offered by Google to anyone with a video embedded on their web page was going away. Casey updated the post a few days later with data crawled from Google’s results showing where all those, historically-open-to-anyone video snippets had gone:
That’s right. YouTube.
Google (mostly) removed the video snippets offered to multiple providers in order to give preferential treatment to their own property. As a result, competition in the space was massively harmed (as Google web search and YouTube combine for well over 95% of all video-based search on the Internet).
They’ve done this in numerous sectors. In the local search world, clicking a result would send a visitor to the business’ website until Google made a change sending all clicks on local-pack search results to Google Maps instead. In the world of flights and hotel searches, any website could compete for visibility in the rankings until Google launched their own flights and hotel aggregators and gave them preferential treatment above the rest of the organic (non-paid) results.
Google did this when they rolled out Google+, offering visual snippets in the search results (that, ostensibly increased visibility and click-through-rate) to participants in their social network’s ecosystem. Danny Sullivan (now tragically a Google employee, though I wish he was still the one writing posts like this) authored a scathing piece noting that Google pushed their Google+ product over relevant results after the G+ launch.
They did it with the Google Play Store, giving ranking opportunities only to apps that were part of their app world. They did it with Twitter, forming an exclusive partnership that shows recent tweets at the top of many searches for individuals and brands, but doesn’t allow participation by any other social network (old and powerful or new and emerging).
Some of these are certainly stronger cases for antitrust activity than others, but in all of them, there is absolutely harm to competition and in many cases, less than ideal experiences for consumers.
On this latter point, Google Maps is an excellent example. When it first launched, many Google searchers and search professionals complained that Google Maps had very few reviews and often incomplete, inaccurate, or entirely missing information about local businesses. Yelp, TripAdvisor, Urbanspoon, and dozens of other local business aggregators did a far better job… But a few years after Google put their own Maps results front and center, they’d received the data they wanted — data that allowed them to compete and, many times be as good or better than the competition. They didn’t compete on even footing with the rest of the field; they gave their entrance into local business data a massive headstart by forcing business owners who wanted the correct info to show on Google searches to enter it into Google and nudging consumers who wanted their reviews to be visible (and business owners who wanted to have those reviews associated with their search listings).
In my opinion, that injection of Google Maps atop the search results, long before it was truly competitive, is a perfect example of using monopoly power in web search to unfairly compete in local business data.
#2: Google Aggregating Results and Answers Without Citation or Credit
You’ve almost certainly seen results that look like this when searching Google:
Google doesn’t offer any explanation of how it forms those blocks of conveniently scrollable results, but professional search marketers are well aware that the engine is merely scraping the listings from the top results, aggregating them, and displaying those most often-mentioned by the web pages Google deems relevant, authoritative, and timely.
As a consumer, this is a nice feature. As a publisher, it’s maddeningly infuriating, and fundamentally unfair.
Google’s bargain with publishers has always been that they get to crawl websites without paying any fee, sweating trademark or copyright issues, or reimbursing site owners for the huge amounts of bandwidth the search engine crawler consumes. In exchange, when Google shows data they got from your website, *you* get a citation — a mention of your brand and a link back to your site.
With these new aggregated results, that’s no longer the case. Teams of writers, editors, bloggers, journalists work hard to make that content, and Google’s automated algorithm merely scoops up the work and serves it to searchers without any credit for anyone but Google. 90%+ of searchers probably have no idea that Google isn’t doing their own editorial work to form these answer blocks.
This phenomenon happens in searches for software tools, for places to visit in a city, for certain types of restaurants, for entertainment, even for legal cases.
The lack of any citation or link in these cases is only slightly worse than what happened to sites like CelebrityNetWorth, where Google’s featured snippets removed a massive amount of traffic from the site (after previously asking if they could partner with the company). The case is described in detail here: How Google Eats a Business Whole.
Here’s to hoping that these practices changes soon, and publishers get the credit their hard work deserves. Fast answers for consumers are great, but removing the incentive to get those answers is a short-term win at best.
#3: Google’s Use of Its Search Engine and Android O/S to Distribute Chrome and Restrict Other Browsers
Do you remember how Google Chrome became the dominant web browser after its launch in 2008?
First, in all fairness, Google Chrome, at launch, was an excellent product and filled many of the gaps left by Microsoft’s Internet Explorer and Mozilla’s Firefox, especially around speed. But it wasn’t just a great product that helped Chrome achieve enormous, near-overnight adoption.
Yup. Google’s homepage, at the time the most popular homepage on the Internet, nudged all of their hundreds of millions of users to install the browser. They did this on search results pages. They did it on YouTube. This Reddit thread remembers.
In January of 2018, Tom Warren authored a piece for The Verge arguing that Chrome, with its adware bloat, its exclusive content, and its Google-ecosystem promotions had become the new Internet Explorer. Six months later, the EU issued their record-smashing 5.4 Billlion Euro fine to the company, in part for this exact type of nudging, favoring, and bundling behavior.
This is probably the most cut-and-dry, obvious example of classic antitrust behavior, but to be honest, I hope that the Department of Justice considers the other two issues above as well.
How Will Google Respond?
This is going to sound very strange, because for the last couple years, I’ve been a pretty harsh critic of Google. But my honest hope is that the company does not get fined or penalized… Wait, WHAT?!
It’s true. I hope that instead of a many-years-long investigation with millions poured into political lobbying by Google (already the largest lobbyist corporation in the United States), the search giant takes a long, hard look at their behavior and realizes that they don’t need to unfairly compete to be an amazing company. And then they change. Voluntarily. Before the DOJ or Congress come cracking down.
Google can and should open up their myriad types of results to any entrant who meets a reasonable set of guidelines for search inclusion (y’know, like they do with the normal paid and organic results). They can and should give credit and provide links whenever they get data from sources, even if its aggregated. They can and should stop unfairly distributing their own products using their powerful distribution through search, Chrome, Android, and others, and instead do some old-fashioned marketing (or some new-fangled marketing; they’re pretty good at that, too).
Ultimately, I’m thankful to Larry and Sergey and the hundreds of thousands of people who’ve built wonderful products at Google over the past quarter century. I think y’all have more greatness ahead. And I don’t believe for a second that you can’t do it without violating the spirit of antitrust. Please, for your own sake, and for the future of a fair, useful, open Internet, Don’t Be Evil. I know you remember how.
p.s. I’d be remiss not to mention that, since 1978, US antitrust law has often conflated “harmful to consumers” exclusively with “higher prices.” There’s lots of speculation that this precedent may be finally be reconsidered, and in light of the EU’s actions against Google, and the evolution of the information economy, I’m assuming in this post (as many who’ve written about the topic elsewhere have) that the US is thinking beyond a direct connection from unfair competition to consumer prices. Even if they don’t, though, I think there’s probably cases to be made that Google’s unfair competition in all these ways drives up advertising costs (and their own ad revenues), which then has to get figured into the margins of consumer spending.