Amazon paradoxes are proliferating. Here’s another: to the extent that Amazon is engaged in anticompetitive conduct, it is the conduct of opening its website to third-party sellers, not, as Amazon critics hold, the conduct of failing to be even more welcoming to those third-party sellers.
As the Times’ David Streitfeld, who has perhaps done more than anyone else to advance the notion that Amazon is unreasonably severe with third-party sellers, seems slowly to be realizing, Amazon’s third-party sellers are, well, a problem. They sell junk. They sell defective products. They fool their customers. And then they disappear.
As the Wall Street Journal alerted us more than two years ago now: Amazon’s open door policy with respect to third-party sellers, which sellers constitute more than 50% of sales on Amazon.com, has caused Amazon effectively to “cede control of its site,” badly degrading the shopping experience.
Which begs the question: why? Why would Amazon let this happen? The answers is: “dreams of monopoly.”
Every other retailer in the world seems to understand that one of the biggest pieces of value retail can deliver to consumers is: curation. The retailer does the hard work of sifting through the junk and the fakes and the defectives to find the good stuff, so that consumers don’t have to do that themselves. Why do you shop at Trader Joe’s instead of your local supermarket? Because you know that if Trader Joe’s is selling it, it’s probably not only of reasonable quality, but likely tastes great too. That’s the value of curation.
But, as Streitfeld correctly notes, Amazon has all but given up on it. Anyone can list products on Amazon. And the company makes almost no effort to flag the best products for you. Ever since that Journal article, the public has known that “Amazon’s Choice” is just an empty label slapped on a piece of third-party seller junk by an algorithm parsing sales trends. No one at Amazon can vouch for the underlying product’s quality, usefulness, or safety.
You might have hoped, as I once did, that at least the products Amazon itself sells under its own brand names, like Amazon Basics, are competently curated. But those, too, have turned out to be no more than the outputs of sloppy and stupid algorithms. The programs troll Amazon’s site for popular third-party products and flag them to Amazon product teams, which then contact the original equipment manufacturer in China, slap on an Amazon Basics logo, and bring the rebranded product back to market. The result is that the Amazon-branded products can blow up in your face, just like the stuff sold by third-party sellers.
What would cause Amazon intentionally to forego delivering curation value to its customers and so risk alienating them from its website? The answer must be that Amazon gets something that it thinks is even more valuable in return for running this risk.
That thing is scale.
By opening its doors to third-party sellers, Amazon was able to bring much of the commercial internet onto its website, ensuring that if a consumer wanted to find something, he didn’t need to search the Internet, he just needed to search Amazon.com. And that in turn ensured that most consumers would do their online shopping on Amazon. And that in turn ensured that if you wanted to sell something online, you wanted to do it as a third-party seller on Amazon. And so on. In econ speak, opening the door to third-party sellers created massive “network effects” for Amazon, effects that make Amazon.com essential for both buyers and sellers.
Curation would destroy that because curation is costly. Algorithms aren’t good enough to curate effectively, as the piles of fakes, defectives, and junk on Amazon.com today shows. So if Amazon were to take curation seriously the company would need to pay people to do it. And even Amazon can only afford so many people. So Amazon would only be able to curate so many products. Which means that Amazon would not be able to offer everything on its website anymore. Which would mean that everyone would no longer shop on Amazon. Which would mean that fewer third parties would need to list their products on Amazon. And so on. Amazon would be better. But it would also be smaller.
And Amazon would face more competition, because now Amazon’s advantage wouldn’t be its network—the fact that Amazon carries everything and so everyone shops there—but rather the quality of its curation. There can only be one retailer that carries everything and at which everyone shops. But there are lots of retailers that curate—and compete on curation.
So, Amazon’s open-door policy toward third-party sellers, and the danger and frustration to which that exposes its customers, is anticompetitive. At least in the sense that it is meant to extract Amazon from the fierce competition based on curation that confronts most retailers, and to put Amazon instead in the unique position of being flea market to the Internet.
Amazon clearly believes that the benefits it gets from avoiding having to compete on curation outweigh the costs to the company of forcing its customers to wade through oceans of junky, fake, or defective products on its website. How could Amazon not, when imposing those costs on consumers makes Amazon indispensable, and hence immune to any consequences associated with alienating those consumers?
Well, not completely immune. There are still other retailers out there. And the more toxic Amazon’s site becomes—the more it really does come to resemble a flea market—the more willing consumers will be to put up with the cost and inconvenience of shopping elsewhere. I personally no longer buy books from Amazon because I hate dodging fakes on its site—buying elsewhere costs more, and sometimes I have to wait weeks longer for my books to arrive. But I’m happier.
The key for Amazon is finding a way to engage in just enough curation to prevent consumers from leaving in droves, but not so much that sellers abandon the platform and it ceases to become indispensable to consumers.
The irony here is that the anti-Amazon zealots, in fighting, under the banner of “self-preferencing,” every attempt by Amazon to impose order on third-party sellers or to curate by promoting its own brands, are effectively pushing Amazon to retain its monopoly position by continuing to welcome the entire commercial Internet onto its website.
Amazon critics: If you really want to make Amazon small, and quickly, help Amazon to engage in more self-preferencing. Ask Amazon to sell only its own branded products on the site, kicking all third-party sellers off. Or ask it to discriminate more heavily against some third-party sellers in favor of others, until Amazon.com becomes like every other retailer: offering a relatively small selection of products that Amazon believes consumers will like the most.
It should be clear that Amazon’s policy of being a flea market, instead of a normal, curating, retailer, is anticompetitive. But just because something is anticompetitive—in the sense that it harms competitors and hence competition—doesn’t make it bad, or an antitrust violation, unless the conduct also harms consumers. So, does it?
The answer must be no. Because, as I have said, Amazon is not completely immune to consumer dissatisfaction. You can find almost everything sold on Amazon elsewhere; it just takes more time and expense to get it. So Amazon today presents the following choice to consumers, who can shop elsewhere: Speed or safety; A low price or the genuine article; One stop shopping or purchases free from defects. And consumers so far have chosen the former, which suggests that they prefer it.
Antitrust cannot eliminate the tradeoff that seems to exist here between scale and quality. But consumers can decide which they prefer. If Amazon doesn’t do something about its site, if it doesn’t strike a better balance between scale and quality, if the junk and the fakes and the defectives continue, consumers will rebel. They will learn that the extra time and expense required to secure curation is worth it. And Amazon will go down; or change to save itself.
I for one don’t plan on buying any more books from Amazon anytime soon.