Tuesday, August 30, 2022

Whether Amazon instigates or draws antitrust scrutiny, it's always about avoiding price competition: abstract parallel between cloud software licensing and most-favored nation clauses

This is the promised follow-up to yesterday's post, New Microsoft software licensing terms to take effect on October 1: revisions designed to strengthen smaller cloud solution providers--and to address Amazon-orchestrated EU antitrust complaint.

There has been a steady trend in recent years for this blog to look at more and more tech industry antitrust issues. In the end, they always involve intellectual property in one form or another. At some point I stumbled upon Professor Frédéric Jenny's "analysis of potentially anti-competitive practices" with respect to cloud infrastructure services, a study commissioned by CISPE. In the previous post I commented on CISPE: it's striking that an organization seeking to promote European cloud sovereignty is primarily backed by Amazon.

Without a doubt, Professor Jenny--an economist--is a prominent figure in the French antitrust community. However, that report he authored for CISPE last year is just a piece of rather unscientific advocacy. What one would normally expect from a competition economist is a clear causal chain and, especially, numbers. Instead, the "evidence" adduced in that paper is just anecdotal. It's like "one customer said" and "oh wait, another customer also said."

That reminded me of a passage from Qualcomm's reply brief in support of its Ninth Circuit appeal of the district court's FTC decision:

"See United States v. AT&T Inc., 310 F. Supp. 3d 161, 211 (D.D.C. 2018) (in weighing evidence of competitive harm, 'competition authorities and courts . . . refus[e] to take the views expressed by customers at face value and insist[] that customer testimony be combined with economic evidence providing objective support for those views'), aff’d, 916 F.3d 1029 (D.C. Cir. 2019)."

In other words, it's old news that customers prefer to get more and pay less. For the avoidance of doubt, the relevant passage is found in a court ruling, but is a quote from a treatise: Ken Heyer, Predicting the Competitive Effects of Mergers by Listening to Customers.

Professor Jenny did the very opposite of what that passage proposes: he took--presumably selective--customer quotes at face value and did not provide any objective support.

The objective of that study isn't totally clear. In no small part it's actually legislative advocacy, suggesting that the EU's Digital Markets Act should also treat software companies like Microsoft, Oracle, and SAP as "gatekeepers." The DMA is huge, but it can't be all things to all people. The reason why some companies must be subjected to special gatekeper rules is their control over platforms, not their ownership of software copyrights. An open letter that CISPE--together with other organizations--addressed to EU competition chief Magrethe Vestager in February urged at a late stage of the legislative process (as the letter even concedes) a "clarification" that would result in the designation of Microsoft, Oracle, and SAP as "monopoly software gatekeepers":

"We cannot wait for a revision of the DMA in five years, nor for a pyrrhic victory in antitrust litigations in 10 years or more when the competitiveness of the market will not be recoverable.

That's what they wrote in February, but by now what they want is a formal antitrust investigation of those companies, initially Microsoft. After yesterday's announcement of new licensing terms that pay heed to the valid ones of the concerns raised by European cloud service providers, regulatory intervention doesn't appear necessary, much less does it seem urgent.

There are major issues to be addressed with respect to mobile app ecosystems: the app tax; the app review tyranny; App Tracking Transparency; access to NFC functionality (for payment systems and other important applications). There's Google's search monopoly and (apart from iOS) superdominant market position in browsers. And Amazon's own conduct.

Amazon would benefit in two ways if CISPE's antitrust initiative resulted in full-blown investigations: by harming a competitor and by defocusing the Commission from other issues that include what Amazon itself has been doing.

The most well-known issue surrounding Amazon's business is a most-favored nation clause: third-party sellers using Amazon's platform are prohibited from "[s]etting a price on a product or service that is significantly higher than recent prices offered on or off Amazon." This is called a "most-favored nation" (MFN) clause and means that vendors cannot offer lower prices elsewhere, be it through direct distribution or on other platforms. The District of Columbia filed an antitrust lawsuit (PDF) over this in May 2021. In 2017, the European Commission accepted commmitments from Amazon on e-books that also involved the MFN topic. As the Commission noted, "[t]he clauses may have led to less choice, less innovation and higher prices for consumers due to less overall competition [...] in e-book distribution" (emphasis added).

Interestingly, Professor Jenny's study discusses the potential competitive effect of cloud service providers who also make software, such as Microsoft, offering customers particularly attractive terms if they buy cloud services as well as software licenses. As I wrote further above, there aren't really hard facts and numbers in that study. But let's assume--just for the sake of the argument--that this is right. It means a company like Amazon with its AWS cloud service could compete, but it would have to charge less for its own services so the total cost of ownership (TCO) for the customer won't be too high.

If a competition authority actually barred Microsoft and others from offering attractive prices for the combination of cloud services and software licenses, the net effect would be that AWS gets to charge customers more than otherwise.

With Microsoft having fleshed out the implementation of its European cloud principles, all of CISPE's members but one--Amazon--have nothing left to complain about that could reasonably give rise to antitrust investigations. The customers of small European cloud service providers will be just fine with the licenses they already have secured from Microsoft (or with new ones that they can optionally obtain through those CSPs). Amazon is obviously free to file an EU antitrust complaint of its own. But to do that, Amazon would have to argue that it can't compete, which is simply not credible based on market share.

As things stand, regulatory intervention doesn't appear imminent. But presumably Amazon won't give up anytime soon. It has the resources to keep on trying.