Americans buy a lot of stuff from Amazon. But Americans also just buy a lot of stuff. This presents a problem for anyone contending that Amazon acts as a monopoly: The company is huge, but depending on how you measure its place in the vastness of American retail, its presence isn’t necessarily overwhelming, let alone clearly illegally monopolistic.
Yes, Amazon is by far the nation’s largest online retailer — but only about 15 to 20 percent of retail sales in the United States are made online. Of every dollar that Americans spend online on consumer electronics, almost 50 cents goes to Amazon; but of every dollar that Americans spend on groceries, around 3 cents goes to Amazon (which owns Whole Foods), while 30 cents of America’s grocery dollar is spent at Walmart and its subsidiary, Sam’s Club.
Earlier this year, Amazon reported $316 billion in 2022 North America sales. Walmart reported $393 billion in U.S. sales. The nation’s overall annual retail sales are in the neighborhood of $7 trillion — giving both behemoth retailers only a single-digit slice of the American retail pie.
The Federal Trade Commission and 17 states sued Amazon this week over a range of practices they say are anticompetitive. The lawsuit asserts that Amazon strongly disincentivizes sellers on its platform from offering lower prices elsewhere online. It also says that some sellers are essentially forced to buy ads and pay for Amazon’s shipping services to get prominent placement among Amazon’s listings. Different elements of Amazon’s broader strategy aim at the same goal, the complaint says — “to keep rivals from gaining the scale needed to compete effectively against Amazon” — ultimately harming “tens of millions of American households” who shop on Amazon and “hundreds of thousands of businesses” that sell through Amazon.
But which rivals are we talking about? This is where the lawsuit runs into trouble.
In order to characterize Amazon’s share of American retail as a monopoly, the plaintiffs define its market in a way that excludes just about every other retailer, online or offline.
Pinning down the sectors in which a company operates — its “relevant market,” in the jargon — is a pivotal part of the complaint. The size of the tank determines the danger of the fish: If courts take an expansive view of the market in which Amazon competes — if they say Amazon swims in a huge tank — the company’s share of that market will appear relatively small and unthreatening; if courts put Amazon in a small tank, determining that there are relatively few substitutes for its services, it’ll look like a shark.
The plaintiffs in this case depict Amazon as Jaws in a fishbowl. They say Amazon monopolizes the markets for “online superstores” and “online marketplace services,” and under those terms some of biggest names in retail wouldn’t seem to qualify as relevant competitors to Amazon — not Home Depot, Kroger, Costco, Best Buy, T.J. Maxx, Wayfair, Dollar General or most other places from which Americans buy a great deal of stuff. Amazon’s critics have long argued that its growth has come at the expense of local small businesses, but these, too, seem to be excluded as competitors to Amazon.
“I think most competition scholars would say that the F.T.C. needs to go to the competition optometrist to get a new prescription,” David Balto, a former policy official in the F.T.C.’s bureau of competition, told me of the way the agency is defining Amazon’s market. “They’re suffering from significant myopia.”
According to the complaint, “online superstores offer a single destination for shoppers to browse a large and diverse selection of goods from multiple brands across a wide range of categories.” They’ve also got to have “sophisticated filtering and discovery tools,” “recommendations for future purchases” based on customer data and “a large volume of authentic, customer-generated ratings and reviews.”
Wait a second — isn’t all this just a description of Amazon?
The complaint zeros in on three other online superstores — the e-commerce divisions of Walmart and Target, and eBay. (The other relevant market, “online marketplace services,” refers to the money Amazon makes from outside merchants who sell to its customers; here the complaint mainly focuses on a comparison of Amazon to Walmart and eBay.) And it’s true, if you narrow your focus to include only retailers that satisfy all these criteria, Amazon wallops the competition. “Documents and data, both from Amazon and industry analysts, confirm that Amazon’s share of the overall value of goods sold by online superstores is well above 60 percent — and rising,” the complaint says.
I find this a little too cute. Amazon all but invented the notion of an online store that sells everything, and it’s been building out that idea for more than two decades; it’s hardly surprising that it dominates the category it pretty much brought into being. Citing Amazon’s dominance in the online-everything-store business seems sort of like citing Chipotle’s dominance in the choose-your-ingredients burrito business. Sure, it would be technically correct — but it wouldn’t meaningfully situate Chipotle in the larger marketplace for fast food.
Sean Sullivan, a professor at the University of Iowa College of Law whose research focuses on how markets are defined in competition law, told me that it’s possible the F.T.C. has evidence showing that customers think Amazon is offering a very different service from other stores — for instance, data that shows that customers are willing to pay higher prices at Amazon because they love its selection or speedy shipping.
But he notes that the complaint’s description of Amazon’s market doesn’t track with his own shopping habits. “If I’m going to buy something pricey, like if I’m buying a $100 keyboard, I’ll look at Amazon,” he said. “But I’ll also look at Best Buy, I’ll look at B&H Photo. I don’t really care that much where I get that from — and the fact that Amazon has lots of other products too is completely irrelevant for my keyboard purchase.”
This resonated with me — it’s rare that I’ll buy anything of significance from Amazon without also looking elsewhere online.
Another problem is the complaint’s implication that Amazon faces competition from Walmart and Target’s online stores but not their offline stores.
“Do most people think that shopping in a Walmart store is a fundamentally distinct experience from shopping on Amazon — to the point where they wouldn’t ever take out their phone in Walmart to check what the price of something is on Amazon?” Sullivan asked. If not, he said, Amazon’s relevant market should include everything that Walmart and Target sell, not just what they sell online.
If courts do use a more expansive definition of Amazon’s market, it could be difficult to label Amazon a monopoly. Of course, antitrust law evolves, but Sullivan pointed me to a landmark 1945 holding, United States v. Aluminum Co. of America, that offered a logical guideline for the share of a market a company needs to have to be called a monopoly: 90 percent “is enough to constitute a monopoly; it is doubtful whether 60 or 64 percent would be enough; and certainly 33 percent is not.”
Amazon’s share of American retail is nothing close to that, and even its slice of the overall e-commerce business — around 40 percent — falls far below the doubtful level in Judge Learned Hand’s formulation.
I’m glad that the government is finally trying to stem the power of large tech companies; Amazon’s rise and continued growth in retail, cloud services, shipping, entertainment and just about everything else gives this single company an untenable hold on society, sometimes compromising employees’ health and dignity, public safety and labor conditions in various parts of the world.
But the plaintiffs’ definition of Amazon’s relevant market — which I promise you will be the subject of endless legal wrangling — suggests that the fight isn’t going to be straightforward.
When the government broke up AT&T in the 1980s, the company controlled something like 80 percent of its market. In the 1990s, when the government went after Microsoft, the company had something like 90 percent of the market for Intel-compatible PC operating systems. If you wanted to use a phone or an affordable computer, both companies were essentially unavoidable.
Amazon’s place in the big picture of retail is nowhere as dominant. And shoehorning definitions isn’t going to make it so.
Office Hours With Farhad Manjoo
Farhad wants to chat with readers on the phone. If you’re interested in talking to a New York Times columnist about anything that’s on your mind, please fill out this form. Farhad will select a few readers to call.