Searching for Market Definition on the Internet: Antitrust in the 21st Century
Two weeks ago in Facebook’s Q2 earnings call, Mark Zuckerberg reiterated Facebook’s desire to become a more effective online “search” competitor, although in his description of the initiative it became clear that he was talking about becoming a more effective competitor in the “market for answers,” as the Wall Street Journal pointed out:
Facebook is trying to give people answers to what they’re looking for in hopes they’ll spend more time on the site or in the app, and in turn stealing searches away from Google or Microsoft’s Bing.
In fact, responding to an analyst’s question, Zuckerberg cited Facebook’s unique advantages in the answers market:
There is huge potential. There are a lot of questions that only Facebook can answer, that other services aren’t going to be able to answer for you. We’re really committed to investing in that and building out this unique service over the long-term. And I think at some point there is going to be an inflection where it starts to be useful for a lot of use cases. But that may still be years away. But we’re just committed to doing this investment and making this right.
Given our frequent musing here on the nature of competition online and its antitrust implications, Zuckerberg’s description of where Facebook is going was telling. Namely, that the narrow market definitions that rely on colloquialisms (“search engines” and “social media”) do not reflect the true nature of competition online.
A fundamental tenet of antitrust law is that in order to figure out if a company is monopolizing a product market, one has to define what that relevant product market is. Sounds simple enough, right? Well, maybe not, especially in the online world.
In economic terms, the “relevant market” consists of products and/or services that compete with one another from the perspective of a consumer. For a solid working definition of the concept, one could refer to the definition from the European Commission’s competition law:
[A] relevant product market comprises all those products and/or services which are regarded as interchangeable or substitutable by the consumer by reason of the products’ characteristics, their prices and their intended use[.]
When it comes to Internet commerce, it is common for tech commenters and the general public to group companies by how they go about their business (e.g. social media, mobile apps, search engine, cloud computing, etc.) instead of what they actually do (provide targeted information, serve ads, navigate through traffic, etc). What they actually do is important consideration for the definition of a relevant market.
It might be useful for a Wall Street analyst illustrating a macro trend to speak of the “mobile app market to almost double this year,” but it is wrong to think of “mobile apps” as a market from an economic perspective unless you contend that human beings have a generalized desire to have small, colorful buttons on their phones but do not care what they actually do. “WhatsApp is getting too expensive, so I’ll just download Angry Birds instead” is something you have never heard anyone say ever.
Unfortunately, when antitrust regulators substitute colloquialisms for actual market definition analysis, the purpose of competition law is turned on its head. Specifically, incumbents often use these incorrect market definitions in attempts to push back against disruptive competition that threatens their privileged economic positions in the marketplace. Here are some examples:
- European telecommunication companies, who make obscene amounts of money charging consumers egregiously high fees for the privilege of sending international text messages, have encouraged the European Commission to challenge the Facebook-WhatsApp merger on the grounds that this involves monopolizing the “instant messaging” market. For that claim to be true, then mobile messaging apps would not compete with the telcos’ own text messaging services, which seems to be the very reason that the European telecommunications companies are concerned about the merger in the first place.
- Despite story after story in major publications of how the big Internet brands all directly compete with one another (not to mention newer startups), the antitrust case against Google continues to drone on in Europe for what are apparently largely political, rather than economic, reasons. The case is predicated on the assumption that “general purpose search” is a relevant market for antitrust analysis. It is also of interest to note that a very diverse array of companies is pushing this case against Google, very few of which are actually in the “general purpose search” business. Although we have illustrated before that defining a relevant market for “general purpose search” is as useless as defining a market for “mobile apps”, the mere fact that a broad array of companies feels competitive pressure from Google gives anecdotal evidence that perhaps these market definitions are imprecise.
- Or, and perhaps the most farcical of my examples, there is Jim Appleton of the New Jersey Coalition of Automotive Retailers, who uses his contention that Tesla has a monopoly (in Teslas!) as justification for state regulations that prevent Tesla from selling its cars directly to consumers. The irony, of course, being that it is the independent auto dealers that have a legally mandated monopoly that prevents competition from other business models.
In reality, when antitrust regulators define a relevant market too narrowly and take action accordingly, they may be shielding some market participants from another competitor who is competing with the others in a slightly different — and often times disruptive — way. In the end, this is a net loss for consumers and innovation.