Streetwise Professor

Federal Trade Commission chair Lina Khan has had her undies in a bunch about Amazon since she wrote an overly long and overly praised Harvard Law Review article about it. Now as a capo in the administrative state, Khan has the power to execute her vendetta, here in the form of a massive antitrust lawsuit against the e-commerce giant.

The case is a farrago of economic ignorance and idiocy. It brings to mind Coase’s remark (which I paraphrase) to the effect that a certain kind of mind sees a non-textbook contracting practice and immediately concludes it is an evil monopolizing practice.

Khan fanboi Matt Stoller provides a pithy summary of the FTC case, which simultaneously reveals its absurdity: “The overall point is that Amazon is degrading the shopping experience, raising prices, and yet somehow still gaining market share.”

Quite the mystery! I mean, maybe it’s possible but the “somehow” part seems rather miraculous. Ask the 1970s-80s Big Three about how degrading quality raising prices usually works out, market share-wise.

This theory brings to mind the Underpants Gnomes:

So yeah I’d really like some explanation how that works, exactly. Neither Stoller or the FTC provide it.

The specifics of the FTC case are hardly more persuasive. I’ll look at the ones they highlight in their press release.

Leading with their best?

Anti-discounting measures that 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.

If so, God save them.

As background, Amazon used to have a contractual term that precluded such differential pricing across platforms. The Euros no likey, so Amazon did it by algorithm (which twists Matt Stoller’s undies).

Contract, algo, whatever. However implemented, this is basically a form of resale price maintenance (RPM). My thesis advisor and mentor, the late, great Lester Telser, wrote the seminal article on the subject when I was one (so maybe the FTC should have caught up by now). Telser argued that RPM is efficient when some retailers provide costly “special services”–notably information–to their customers. This creates a potential free riding problem. Sellers can free ride off the information/special services provided by some retailers/platforms and sell at a discount at stores/platforms that don’t offer the costly service.

Amazon provides a lot of information on products, including product reviews. Offering a product on Amazon, a seller’s potential customers can learn about the product there, and in the absence of the restriction that the FTC is attacking, can then go and buy it from a platform that doesn’t incur these costs, thereby free riding on Amazon’s information provision.

Next up is a real doozy:

Conditioning sellers’ ability to obtain “Prime” eligibility for their products—a virtual necessity for doing business on Amazon—on sellers using Amazon’s costly fulfillment service, which has made it substantially more expensive for sellers on Amazon to also offer their products on other platforms. This unlawful coercion has in turn limited competitors’ ability to effectively compete against Amazon.

For one thing, I don’t see how requiring use of Amazon fulfillment on sales via Amazon raises the cost of selling on other platforms. But even putting that head-scratcher aside, where is the “coercion” here? Amazon offers a bundle. They don’t put a gun to your head forcing you to take it.

Further, there is a very good efficiency reason for this bundling. Amazon offers a delivery guarantee. If it vertically integrates into shipping, it has considerable control over whether it performs according to the guarantee. If a seller uses another shipper, Amazon has no such assurance.

Again, there is a free rider problem. Seller gets sales from customers expecting speedy delivery. Seller uses a cheap but less reliable carrier. Goods arrive late, or in bad condition. Whom is the customer mad at? To whom does the customer look for recourse? Amazon.

Vertically integrating into shipping reduces the cost of providing the guarantee that buyers like.

The FTC also apparently doesn’t understand the basic point that if alternative carriers are cheaper than doing it in house, by forcing customers to use a higher cost service Amazon reduces the derived demand for its services and its platform. Since–as the FTC also whines about–Amazon apparently prices to extract large sums from sellers, in the absence of efficiency benefits like those just described, it would be able to, and find it more profitable to let sellers choose their putatively cheaper carrier and charge the sellers a higher price to sell on Amazon.

Next!

Degrading the customer experience by replacing relevant, organic [but are they vegan????] search results with paid advertisements—and deliberately increasing junk ads that worsen search quality and frustrate both shoppers seeking products and sellers who are promised a return on their advertising purchase.

This is basically the Stoller argument. I’ve already discussed its facial absurdity. But it’s also absurd when you dig deeper.

Search ranking is a scarce resource, and an extremely valuable one. Scarce resources have to be allocated somehow. Usually best way to allocate a scarce resource is to price it. (This is done all the time with online advertising, by the way.) So what is the best way to price this resource?

An auction? Possible–Amazon has no doubt considered that. But maybe pricing through advertising is superior. I don’t know, but given that advertising can serve a reputational branding role, and can provide information, it may well be the case that for selling information-intensive goods, or goods of uncertain quality, advertising is better than auction to prioritize search results.

What I do know is that Amazon internalizes the costs and benefits here. Meaning that they have the incentive to choose the best alternative. Lina Khan–not so much.

And I can guarantee that if Amazon allocated search ranking by auction or some other pricing mechanism, that would cause the FTC to have a meltdown too. They apparently believe that “organic” rankings (WTF are those?) sans pricing of any form are best.

Fine. Start your own platform, Lina.

And next:

Biasing Amazon’s search results to preference Amazon’s own products over ones that Amazon knows are of better quality. 

Preferencing in the absence of some other efficiency advantage is hard to rationalize–and certainly the FTC provides no such rationale. Again, since Amazon could extract the alleged quality advantage (and perhaps cost advantage) of the third party sellers by raising its prices to them, why offer lower quality products?

Here’s one reason: by offering its own products, and competing on price and perhaps visibility, Amazon is entering the market, creating competition, and inducing third party sellers to reduce prices. Which helps consumers and drives them to use Amazon. That is, by creating competition preferencing is pro consumer.

And finally:

Charging costly fees on the hundreds of thousands of sellers that currently have no choice but to rely on Amazon to stay in business. These fees range from a monthly fee sellers must pay for each item sold, to advertising fees that have become virtually necessary for sellers to do business. Combined, all of these fees force many sellers to pay close to 50% of their total revenues to Amazon. These fees harm not only sellers but also shoppers, who pay increased prices for thousands of products sold on or off Amazon.  

Oi. Amazon charges fees. Who knew? Amazon has market power that allows it to charge high fees. Also hardly a revelation.

But this would apparently be a revelation to the FTC: it is not an antitrust violation to exercise market power gained by “superior skill, foresight and industry.” Only nefarious acts to monopolize fall afoul of the antitrust laws.

This one should be a slam dunk for Amazon’s lawyers.

One thing that jumped out at me in the complaint that isn’t in the press release: Amazon’s pricing and policies make it difficult for “single-category ecommerce companies to achieve the scale they need to succeed.”

So, who says they should succeed? Amazon’s success and the success of other platforms (e.g., Walmart) suggests that there are strong economies of scope here, which would make “single-category” platforms inefficient. We don’t want inefficient platforms to survive. If the FTC were to get its way, and somehow handicap Amazon so that inefficient single-category platforms that can’t exploit economies of scope survive, we are worse off.

One interpretation of the anti-Amazon arguments are that it is essentially using its scale and scope economies to price below the cost of alternative platforms, so those platforms don’t enter. This is analogous to Demsetz’s theory of natural monopoly, or the behavior of a dominant firm in a contestable industry (a la Baumol, Panzar, Willig). In these theories, there is no entry, but that’s a good thing because entry would be duplicative (given the scale and scope economies). Moreover, the potential for entry disciplines the incumbent’s pricing, mitigating traditional monopoly triangle losses.

The FTC’s lament about the high prices charged to sellers on Amazon’s platform highlights a major lacuna in its attack: the failure to take into account that Amazon is a two-sided platform. Pricing in two-sided platforms is quite complex, and it is impossible to evaluate Amazon’s pricing justly without taking these complexities into account.

One thing missing from the complaint is one of Khan’s law review criticisms–that Amazon uses the information generated from its operation of the platform to identify profitable entry opportunities, i.e., to identify products that Amazon can produce (or procure) and sell in competition with third party sellers.

Entry is pro-competitive, dontcha know. It tends to lower prices. I’m sure that’s news to you.

Companies perform market research all the time to identify which markets they can enter profitably because they can make products better or cheaper than incumbents. If Amazon can perform this research low cost due to the information it sweeps from the platform–great! Let them use it. (I note that an early theory of vertical integration advanced by Ken Arrow was that such integration produces information that the upstream firm can use to make better decisions.)

In sum, the FTC case against Amazon is pathetic. I would characterize it as economics free. Or maybe more accurately anti-economics.

It is also entirely untethered to any credible theory of how Amazon’s actions harm consumers, or more importantly, how eliminating the practices that the FTC decries would benefit consumers. And as my foregoing arguments show, consumers would almost certainly be harmed if the FTC gets its way.

But this is not a surprise. The Khan FTC has explicitly rejected the Consumer Welfare standard that has guided antitrust since Robert Bork first mooted it in the 1970s.

Now I have my issues with the way courts have interpreted the standard–which, alas, is largely due to how Bork framed it. A close reading suggests that Bork interpreted “consumer welfare” to mean efficiency-enhancing. The courts have typically read it to mean “lower prices to consumers.” Now in many contexts, the courts’ Cliff’s Note version leads to the right results–but not all.

That is only a reason to refine the standard, not to reject it altogether. But reject it the Biden FTC and DOJ have.

Which is why, as a general matter, and in the specifics of the Amazon case (and other cases Khan’s FTC has filed), it is more than fair to say that “FTC” really means “F the consumer.”