FTC's case against Amazon is misunderstood
The case makes a nuanced argument about how Amazon abused monopoly power through anti-discounting measures and tying in Prime with Fulfilled by Amazon
This is the last of three articles in Unpacked’s “Tech Policy September” series (well, almost September).
Disclaimer: The views expressed in this article are solely my own and do not reflect the views or positions of any organization with which I am affiliated, including my employer.
If you look at recent media coverage about DOJ v. Google, there is general consensus that the DOJ’s arguments are well-founded. Most articles rightfully point to Google’s distribution deals as the crux of the trial, and in last week’s analysis, we talked about how these onerous deals that Android device manufacturers had to sign might be the DOJ’s smoking gun. The coverage on FTC v. Amazon however has been mixed. If you read FTC’s press release and then Amazon’s rebuttal, I wouldn’t blame you if you sided with Amazon.
I have been skeptical of the FTC given their repeated missteps and back-to-back failures, and I was ready to write about how this is yet another misstep. Then I read the FTC’s 150-page complaint against Amazon. My takeaway was two-fold - 1) it’s a very well-framed case that is poorly understood, and 2) the FTC needs a better PR person. In this article, we’ll dive into:
FTC’s arguments against Amazon
Non-arguments that are being misunderstood as arguments
What lies ahead
FTC’s arguments against Amazon
I won’t harp on this since we’ve covered this a bunch before, but here’s a quick refresher on anti-trust / anti-monopoly laws in the US:
It is not illegal to be a monopoly; it is however, illegal to engage in anti-competitive tactics to maintain monopoly position
For a monopoly to exist, a definition of “relevant market” is required and needs to be accepted by the courts
Interpretation of anti-trust laws in the US went through many phases in the past century. The most black and white version of it applies the consumer welfare standard (i.e. if a monopoly causes price increases / directly lowers consumer welfare, that’s an anti-trust violation). A broader interpretation of the law, which is required for tech products given they’re often free, looks at whether there is short or long-term harm created for consumers (eg. if a company killed all competition and if that results in price increases / stifled innovation in the long term, that could be considered an anti-trust violation)
With that context, let’s lay out the outline of FTC’s case against Amazon. The argument goes something like this:
Amazon is a monopoly in two relevant markets - the “online superstore market” and the “market for online marketplace services”
Amazon engaged in anti-discounting tactics that prohibited sellers from offering lower prices outside of Amazon. This, in turn, increased prices for consumers on and off Amazon.
Amazon used Prime’s popularity to force sellers into using Amazon’s fulfillment services. This, in turn, made it more difficult for sellers to use a single fulfillment provider across multiple marketplaces (“multi-homing”), and therefore made the market less competitive for rivals
If you’re not sold on it, that’s okay - it gets better. Let’s unpack these arguments one by one to see how much merit they have.
Argument #1: Amazon is a monopoly in two relevant markets
The FTC argues that Amazon is a monopoly in the “online superstore market” and “the market for online marketplace services”, and they provide pretty robust reasoning for the definition of these markets.
Paraphrasing, the online superstore market is characterized by a wide breadth & depth of product selection, 24/7 shopping access, distinct features to easily filter & identify the right products, lower effort to shop (vs in-store), recurring shopping engagement, and ability to personalize shopper experiences. The FTC argues that this market is not reasonably interchangeable with both brick-and-mortar stores, as well as online stores that lack breadth. FTC also explicitly excludes the grocery market from this definition due to the perishable nature of grocery products. In the online superstore market, FTC argues that Amazon has a 75%+ market share.
The second relevant market that the FTC defines is that of “online marketplace services”. I’ll spare the specifics but the key difference is that sellers have the ability to list products and set prices, unlike retail where the store usually buys products, gets inventory into their own books, and then sells to consumers.
One of the criticisms of the FTC’s case is that the relevant market definition is too narrow. For example, Amazon might be dominant in the online marketplaces market but if you include all retail (online + offline), Amazon has a much smaller share. While I agree there is some truth to it, this is literally every monopolist’s argument - in DOJ v. Google, Google argues that search advertising is too narrow a relevant market despite being ~$160B+/year.
I think this argument is running out of steam given how large Big Tech has gotten, and the fact that DOJ v. Google went to trial is clear evidence that the courts aren’t sympathetic to the argument anymore. And broadly, I agree with that - there needs to be a line beyond which a large player cannot make the “oh we’re big but if you 10x the denominator, we’re not that big” argument.
Argument #2: Amazon’s anti-discounting measures increased prices for consumers across the internet
This argument from the FTC’s summary seems questionable at first glance:
Anti-discounting measures punish sellers and deter other online retailers from offering prices lower than Amazon, keeping prices higher for products across the internet. For example, if Amazon discovers that a seller is offering lower-priced goods elsewhere, Amazon can bury discounting sellers so far down in Amazon’s search results that they become effectively invisible.
Some strong reactions have come out to this statement. Amazon innocently argued that of course they are promoting lower prices because it’s a good deal for consumers, and it is ridiculous that the FTC is prohibiting that:
Some of the businesses selling on Amazon might still choose to set prices that aren’t competitive. Just like any store owner who wouldn’t want to promote a bad deal to their customers, we don’t highlight or promote offers that are not competitively priced.
The FTC’s case alleges that our practice of only highlighting competitively priced offers and our practice of matching low prices offered by other retailers somehow lead to higher prices. But that’s not how competition works.
But that is not what the FTC is alleging. The FTC complaint states a few things. It starts by stating that Amazon crawls websites across the internet to see whether the seller is offering lower prices elsewhere. So far so good - there is nothing illegal about it and Amazon rightly points out that this is a common practice among other retailers.
Where this gets anti-competitive is when Amazon unfairly and aggressively punishes sellers for having lower prices outside of Amazon, while continuing to increase Amazon’s take rate steeply over the years. The punishment for having lower prices outside of Amazon is the following:
Being booted from the “Buy Box” at Amazon, which is how most Amazon consumers buy (the exact % is redacted); Amazon acknowledges that this can tank a seller’s business
Heavy de-prioritization in Amazon’s organic results
Threat of exile / being banned from Amazon
Let’s play that out again for a second. If you’re a seller on Amazon, and you decide to have lower prices outside of Amazon, your business on Amazon is essentially killed. The popular counter argument is - they are a marketplace that wants to provide competitive prices, so why can’t they force sellers to have the lowest prices on Amazon?
But play this out in the context of Amazon increasing their take rate from sellers. Say a seller would like to sell a pair of earphones for $25 to cover costs and make a profit. If Amazon had a 25% take rate back in the day (inclusive of fulfillment, advertising, etc.), this means the seller would list the product for $33. The FTC argues that Amazon’s take rate today (including the dollars sellers need to spend on ads to make sales happen) is ~50%, i.e. the seller would now have to charge $50 to make the same level of profit. Sure, whatever, that’s the free market.
But now, let’s say another marketplace like Walmart or eBay has a lower take rate of ~30%. In a competitive market, the seller should be able to price the product at $36 (to maintain their $25 goal). But no - Amazon’s anti-discounting measures prohibits the seller from pricing below Amazon prices anywhere else. You could argue that the seller would drop their price on and off Amazon to take a lower profit but in a world where Amazon has dominant market share, there is no incentive for the seller to do that. Therefore, as a consumer, you’re now paying $50 for a product that you could’ve been getting at $36 if Amazon was not engaged in anti-discounting measures. In other words, Amazon suppressed market-based price discovery by using their monopoly, and forced all its competitors to be price followers (even in situations where their competitors and sellers wanted to drop prices).
I have to say that it’s a pretty impressive argument (and you can see why this is misunderstood, re. FTC needs a better PR person). The incredible thing about this argument is that it ties very strictly to the well-established consumer welfare standard of anti-trust, i.e. if a monopolist is causing price increases to the consumer, that’s a clean argument for consumer harm.
Now the big question - the accusations above are substantial and the FTC is making bold claims (being booted from the “Buy Box”, de-prioritization from organic results). Can they prove it? If you’re skeptical, I can understand that, but the FTC gives some glimpses of proof in their complaint. For example, they talk about the existence of Amazon’s contractual agreements with sellers which prohibit them from having lower prices elsewhere. These contracts were ruled anti-competitive in the EU and Amazon started backing down, but the FTC argues that the punishments continued to exist - Amazon just moved from contractual to algorithmic implementation.
The FTC also has a meaningful amount of redactions in their complaint about an algorithm that Amazon built specifically for penalizing sellers who had lower prices elsewhere, and several mentions of a related “Project Nessie” (note that Amazon does not refute any of these claims). From my past analyses of the FTC and DOJ’s cases, there are usually no factual errors in their claims. If they claim something exists (like the algorithm), they likely have robust proof to back it up. Where the FTC has failed in the past is with the arguments they have made in cases and not with the facts themselves. So, I’m inclined to think that the FTC has a strong case with this argument and the evidence to back it up.
Argument #3: Amazon used Prime’s popularity to force FBA adoption and stifle competition
This one is a little less black and white than anti-discounting but I think it’s well-crafted. Let’s break down the argument.
Amazon created Prime, primarily tied to two-day delivery (and now one-day delivery for many products), and got consumers used to always looking for Prime enabled products. The FTC states that Amazon increased Prime’s cost from $79 to $139 over the years, and points to Amazon’s motivations with Prime (corroborated by internal comms) - Amazon does not care about making money off of subscriptions but the commitment to the Prime subscription makes consumers order more from Amazon. A redacted % of orders on Amazon are Prime-eligible products, and you can imagine the percentage is high.
Nothing illegal about this - Walmart, Costco, several other retailers have subscriptions. There have been a bunch of posts/tweets getting mad at the FTC for making a luddite argument criticizing subscriptions - that’s not what the FTC is doing. The details about Prime, that Amazon built a highly valued brand that is now used as a default by most customers, is merely context for the next part of the argument.
After making Prime a prominent brand (because customers love it), and while continuing to scale up Prime, the FTC argues that Amazon coerced sellers into using Fulfilled by Amazon / FBA (Amazon’s fulfillment service). FTC says that sellers would prefer to “multi-home”, i.e. have a single fulfillment partner that can fulfill orders across multiple marketplaces. But Amazon forces sellers to use Fulfilled by Amazon by essentially tying it in as a requirement to be Prime-eligible.
In other words, as a seller, you could refuse to use Fulfilled by Amazon. This would make you not eligible for Prime, and in turn less likely to show up in the “Buy Box”, and therefore lose your business on Amazon. So you don’t have to use Fulfilled by Amazon but you kind of do. This mechanism, called a “tie-in”, is another classic from the monopolist’s playbook - this is exactly what the DOJ is accusing Google did by tying in essential Android services to things that majorly benefitted Google.
I think there are a couple of strong counter-arguments to this:
First is the “consumers love it” argument. Amazon argues that consumers love Prime because it’s such an incredible consumer value. This is enabled by FBA which provides best in class fulfillment, and sellers can also choose to use other fulfillment services and still be Prime-eligible.
Second is the “Amazon can’t be forced to take the risk of poor 3rd party fulfillment” argument. Ben Thompson of Stratechery argues that it’s unreasonable for a regulator to demand that Amazon give out access to the Prime label and bear the reputation risk when 3rd party services don’t meet the standards of the Prime delivery guarantee.
I generally agree that the consumer welfare standard doesn’t hold up very well here - Prime has been a great deal for consumers and Amazon deserves credit for achieving the incredible feat of reliable 1-2 day delivery. However, in today’s environment, it’s important to look at anti-trust not just from a short-term consumer welfare standard angle, but also from the long-term angle of how this impacts competitiveness in markets.
I do think that Amazon stifled competition in the independent fulfillment providers space, and consequently made it difficult for sellers to sell on rival platforms. In 2015, Amazon allowed 3rd party fulfillment providers through a program called “Seller-Fulfilled Prime” and it was well-received by both consumers and sellers, but the FTC argues that this program was shut down and is back up partially today with an unreasonably high quality bar for fulfillment providers.
Overall, consumers did get a good deal in the short-term but that came at the cost of long-term competitiveness - if you’ve used Amazon for many years now, you’ve anecdotally seen that Amazon is not really “cheap” anymore, just reliable. Playing this out five years, with lack of competition, there is nothing stopping Amazon from continuing to increase prices. The FBA / Amazon Prime tie-in, while less cleaner to prove as an anti-trust violation, is a key anti-competitive mechanism, and my hope is that the FTC is coming with enough ammunition to prove it.
Non-arguments that are being misunderstood as arguments
I’ll briefly cover some non-arguments that are being misconstrued as FTC’s arguments:
Prime subscriptions make users want to order more on Amazon, and this is bad - this is not the FTC’s argument, and it’s a piece of context they include to describe how critical Prime is for Amazon
Amazon aggressively promoted the “Buy Box” and “Prime” filters in search, and that is bad - again, not the FTC’s argument. It’s a piece of context for how important both of those mechanisms are
Advertising is bad and degrades the consumer experience - FTC does make this argument, and I agree with FTC’s critics that the argument is both trivial and distracting (re. FTC needs a better PR person). However, this is not the crux of their case. It’s mostly included as context to show Amazon’s increasing take rate, which along with anti-discounting measures is increasing prices consumers pay across marketplaces.
What lies ahead
The case is still in its early days unlike DOJ v. Google which is already in trial. So, it will be sometime before we see how this plays out. And to clarify, this is by no means an “in defense of the FTC” article. I have been explicitly critical of the FTC’s missteps in the past and their brute force / “everything big is bad” approach.
But especially given my past skepticism, I’m really encouraged to see the FTC embrace a more clinical approach that targets specific anti-competitive mechanisms, and I’m optimistic about where their efforts will land.
Rest of “Tech Policy September” on Unpacked:
I feel FTC's narrow market definition unjustly targets Amazon, ignoring broader retail competition that could dilute its alleged monopolistic status.