by Chiara Farronato, Andrey Fradkin, Andrei Hagiu, Dionne Lomax
January 11, 2024
Scott Olson/Getty Images
Summary. Regulators in the United States and Europe have been taking on Big Tech, challenging what they say are the companies’ anti-competitive and predatory strategies that harm consumers and third-party users of their platforms. This article examines the FTC’s case against Amazon and explores the economic arguments in favor of and against each of the agency’s key claims. They highlight fundamental trade-offs between the benefits of aggregating economic activity on a single marketplace and the costs of that marketplace setting the rules for everybody relying on it.
Large tech platforms, such as Amazon or Google, have been drawing increasing criticism over the last few years. According to the allegations, they have been using anticompetitive and predatory strategies to illegally exploit their market power to the detriment of consumers and third-party providers that depend on these platforms.
Amazon is a case in point: The company is facing lawsuits and new antitrust regulation on both sides of the Atlantic. In September, the U.S. Federal Trade Commission (FTC) and 17 U.S. states filed a lawsuit denouncing the company’s abuse of its dominant position. The 172-page complaint alleges that Amazon has engaged in conduct that kept prices high and quality low, limiting fair competition and stifling innovation.
While the earliest the case is expected to go to trial is 2026 — and, of course, there is always a chance that it could be settled before then — the issues are worth considering given that regulators’ challenges to Big Tech’s power in the United States and other countries are likely to persist. Drawing from our expertise, we will discuss in this article the best economic arguments in favor of and against each of the key claims in the FTC’s case. They highlight fundamental trade-offs between the benefits of aggregating economic activity on a single marketplace and the costs of that marketplace setting the rules for everybody relying on it.
Bundling of “Fulfillment by Amazon” with Prime Eligibility
One of the two key pillars of the FTC complaint is that Amazon bundles its Amazon’s Prime badge and its Fulfillment by Amazon (FBA) service to sellers, effectively forcing third-party sellers who wish to obtain the Prime badge to also use FBA.
It is first useful to explain how the Prime badge and FBA work today, especially given some recent developments that would seem to weaken the FTC’s bundling claim. To obtain the Prime badge, sellers must be able to guarantee fast and reliable shipping (one- and two-day shipping options, less than 0.5% order cancellations), as well as return policies matching standards set by Amazon. Because the badge signals high-quality service, consumers prefer products with a Prime badge, which, in turn, means the badge is highly coveted by third-party sellers.
As of this writing, the Prime badge can be obtained in one of two ways. First, sellers can enroll in Seller Fulfilled Prime (SFP) and obtain the badge after a 30-day trial period. Under this option, Amazon charges no fee and sellers can use their own fulfillment services. Second, sellers can get the badge by enrolling in Fulfillment by Amazon (FBA), which gives sellers access to Amazon’s warehousing and logistics infrastructure in exchange for a fee.
It is noteworthy that the SFP option was initially launched in 2015, discontinued in 2019, and then made available again in October 2023. Given that today Amazon sellers can freely choose between obtaining the Prime badge via the SFP option and obtaining it via the FBA option, this FTC complaint seems moot. Furthermore, while it is in principle possible that Amazon may have at some point seen bundling of FBA and Prime as a way to discourage sellers from selling through other channels, that no longer seems to be the case. Indeed, in recent months, Amazon has made the combination of FBA and Prime badge available to sellers on their own websites, including those powered by Shopify.
Given how recent all these developments are and that the FTC lawsuit was filed in September 2023, one might speculate that the threat of regulatory oversight may have pushed Amazon to expand the Prime badge to include third-party fulfillment and to open FBA to third-party ecommerce websites. Whether this is true or not, it does create an interesting situation where the potentially offending behavior claimed in the lawsuit (bundling of Prime and FBA) is no longer in place, possibly because of the lawsuit.
Nevertheless, it is still useful to discuss whether the bundling of Prime and FBA was indeed anti-competitive during the period it was in place (between 2019 and 2023). The FTC claims that such bundling is anti-competitive because it limits sellers’ ability to use competing fulfillment services when selling on Amazon, and to use the same fulfillment solution when selling across multiple channels.
To counter this claim, Amazon can (and does) point out that there is a reputational rationale for bundling FBA with Prime. The Prime badge is directly associated with the Amazon brand in the mind of consumers and is perceived as a guarantee of fast and reliable shipping. Any poor shipping experience with a seller under the Prime badge may have negative spillovers on the brand itself, undermining the value that consumers place on Prime — for Amazon and every other seller carrying the badge. Then, since Amazon can better monitor shipping quality when handling fulfillment directly (FBA) instead of allowing it to be outsourced to a third party, bundling can help mitigate the negative spillovers.
Price Parity
The second pillar of the complaint focuses on price parity. When a marketplace like Amazon.com imposes price parity rules on its sellers, this means that sellers are not allowed to sell the same products at lower prices on competing sales channels (typically, under penalty of being excluded from the marketplace). While Amazon did indeed have explicit price parity rules until several years ago, it has not imposed these rules in the United States since 2019 as a result of regulatory pressure. However, the FTC claims that these rules are still implicitly enforced: It contends that sellers who have lower prices on other ecommerce sites find themselves punished by seeing their products demoted in search results on Amazon.com.
Why might price parity rules be harmful? Consider a seller on Amazon who also has the option to advertise on Facebook and direct buyers to its own website. If advertising costs on Facebook are low relative to Amazon’s transaction fees, the seller can pass through some of the savings to the consumer and keep the rest of the savings. By requiring that prices are the same on the seller’s website and on Amazon’s, price parity rules prevent this from happening. This has the effect of removing the seller’s incentives to advertise on Facebook and sell outside Amazon at lower prices. Thus, Amazon’s market power combined with price rules may lead to artificially high prices across all sales channels. This mechanism has been well articulated by economists.
On the other hand, price parity clauses may be justified because they let Amazon prevent “showrooming,” whereby consumers find products through Amazon and then buy them from a different website. If that were to happen, Amazon would not get compensated for the search and recommendation services it provides, which may lead to under-investment in those services. That said, the risk of showrooming may in fact push Amazon to create even more value along the transaction funnel to convince consumers to stay on its marketplace — for example, by increasing delivery quality and allowing for flexible return policies. We don’t have the data to validate these concerns.
Advertising
Another set of tactics highlighted in the lawsuit has to do with the priority given to different products. The complaint alleges that Amazon displays too many sponsored (as opposed to organic) search results and that this is a way to extract even higher fees from third-party sellers (advertising fees in addition to transaction fees). This also places third-party sellers who compete with Amazon’s own products at an even greater cost disadvantage. When one considers the nature of competition in online marketplaces, the allegation that this is anti-competitive is not so obvious.
Consider a new seller that joins Amazon or an existing seller seeking to introduce a new product. Despite Amazon having a lot of information about existing products frequently sold on its marketplace, identifying which new products are worth showing to consumers is a challenging task. Indeed, today there are more than 2 million third-party sellers on Amazon’s marketplace and more than 600 million products sold on it. This makes it hard for any given seller to stand out from the crowd. In this context, advertising (in the form of sponsored search results) introduces a price-based mechanism to identify which products should be shown to consumers. After all, advertising is ubiquitous online and offline, with advertising expenditures exceeding $300 billion in the United States.
However, (excessive) advertising may also hurt consumers if relatively worse products end up winning the sponsored search result slots and consumers do not scroll past them. Advertising may also increase sellers’ costs, which are then passed on to the consumers in the form of higher prices. What is the net effect of advertising in the case of Amazon? Once again, it is hard to know for sure without data that is very hard to come by, given that both the benefits and costs of advertising are very real.
Self-Preferencing
Out of all products sold on Amazon’s platform, third-party sellers account for roughly 60% and Amazon’s first-party sales (products sold by Amazon or carrying an Amazon brand such as Amazon Basics) account for roughly 40%. It is worth noting that the percentage of third party sales has been steadily increasing over time.
The FTC’s complaint alleges that Amazon oftentimes gives undue priority to first-party products over those from third parties, which creates unfair competition and leads to consumers purchasing inferior products. If this were the case, it would qualify as abuse of a dominant position.
It is true that there are many instruments that Amazon can use to tilt the playing field on its marketplace to favor certain products over others: the ranking algorithm that determines which products appear higher in search results, badges (such as the “Best Seller” badge), product recommendations (such as “Frequently Bought Together”), and the choice of default seller when a user presses “buy” on a product page.
The challenge, however, is proving that self-preferencing is taking place. Amazon can argue that customers prefer its own brands for reasons that are not as objective or easily observable as, say, price. While one should not take Amazon’s claims at face value, it is important to recognize that the measurement of self-preferencing is much more complex than it might appear at first glance. Amazon may be using signals of product quality, such as return rates, which are simply not available to researchers and policymakers wishing to measure self-preferencing. Without such data and the corresponding ranking algorithms, or at a minimum data on customer demand, it’s hard to know for sure what explains the position of each product on an Amazon page.
Remedies
Suppose that the FTC does prove at least some of its allegations. What then should be the remedies?
We tend to be skeptical of heavy-handed structural remedies, such as breaking up Amazon’s product brands, fulfillment services, and marketplace into separate business entities. This is a very blunt instrument: While it may preclude some of the potentially harmful behaviors mentioned above, it would also deprive both consumers and third-party sellers of significant benefits such as scale and scope economies and one-stop shop convenience.
Behavioral remedies are more reasonable: Any conduct proven to be illegal should be banned. If, for example, the FTC proves that Amazon’s enforcement of price parity is illegal, the FTC would be justified in preventing this practice and in determining monetary damages that reflect consumer harm. The same goes for self-preferencing.
A key challenge, however, is how to monitor whether Amazon does, in fact, engage in such practices. One approach would be to require Amazon to allow approved third parties (regulators, researchers) to audit its ranking algorithm and search results pages. In fact, we have already advocated for this in the past, and the European Union is implementing a similar mandate in the Digital Services Act.
Other conducts would be harder to limit. For example, suppose that the FTC proves that Amazon abuses its dominant position through excessive advertising. A possible remedy would compel Amazon to reduce the amount of advertising on the platform, but how much less? Banning advertising altogether is an extreme and inefficient solution, which would remove sellers’ ability to signal their willingness to pay for preferential product placement. And regulating an optimal level of advertising relies on the assumption that such a level can be better and easily identified by regulators.
The outcome of this high-stakes litigation will undoubtedly shape the future landscape of ecommerce regulation for years to come. While regulating the high-tech industry presents significant challenges, striking the right balance is critical. If the FTC prevails, the remedies pursued ideally should preserve fair competition without sacrificing the inherent benefits of scale and convenience for consumers and third-party sellers that Amazon offers.
