The following originally appeared in the January 2021 issue of the CPI Antitrust Chronicle.
The communications industries range from plain old telephone service, through broadband access, whether fixed or mobile, to cable and satellite video distribution. These industries are not exactly passé compared to the online platforms such as the search engine “market” that Google has so admirably revolutionized and may (or may not) dominate today. After all, Google search queries need a pipe to travel from us, the hopeful searchers toiling on our computers, tablets or smartphones, to Google servers. Even more important, the pipe is necessary for the return trip back to us, bringing a cornucopia of audio and video that condenses real and imaginary worlds on one small screen in ways that would have been incredible to time travelers visiting us from the twentieth century.
And so, the communications industries and online platforms need each other: the first needs the second’s content. The second needs the first’s pipes. In that sense, they are contemporaries and codependents. But the communications industries have a history longer than that of the online platforms — after all, they go back to 1876, when Alexander Graham Bell summoned Mr. Watson on the phone. With the longer history comes longer experience with analyzing and resolving competitive issues. And the communications industries afford many useful teachings for a way forward in the Google search engine case, including a way past and around the impasses and conundrums presented by the complaint filed against Google by the Department of Justice (“DOJ”).
Here are two of these lessons: be careful not to throw out the baby of the competition you want with the bathwater of the exclusivity arrangements you do not like. And, to promote competition, nurture a competitor: find a white knight or two; and give them the resources to compete.
But first, a few words about what links legacy communications and online platforms together from the perspective of competition analysis. There is a lot that does. First, they are both networks prone to market power and its exercise: each additional user of the network does not merely add a proportionate unit to a provider’s market share; it takes out disproportionate quantities of oxygen away from would-be competing networks. Second, both industries are chains of intricately connected links. In the communications industry, we have electromagnetic spectrum, wireless towers, cable or satellite platforms, broadband access pipes, and online video products or cable networks that pass through these pipes. In the online platform space, we have search engines, browsers, and operating systems. And, of course, as mentioned, each industry is a link to the other’s chain in its own right: online platforms provide increasing amounts of content for the communications conduits, which in turn provide all of the transmission capacity for that content. Vertical integration over many links of the chain also gives companies with market power over one of them one more chance to leverage that power. At the same time, a presence in one link of the chain may endow a company with the wherewithal to become a credible competitor over another link.
- Don’t Throw Out the Baby With the Bathwater
The DOJ, joined by a group of eleven states, alleges that Google has a 90 percent share of all “general-search-engine queries” in the United States (95 percent on mobile devices), and that Google is able to maintain this position through contracts with browser providers (such as Apple’s Safari or Mozilla’s Firefox) or device manufacturers (such as Samsung) that make Google the default search engine in those browsers or on devices running Google’s Android operating system. The complaint takes aim at these exclusivity agreements as one of the chief methods that Google has employed to exercise its market power. In the complaint’s words, “Google has unlawfully maintained its monopolies by implementing and enforcing a series of exclusionary agreements with distributors over at least the last decade. Google’s exclusionary contracts cover almost 60 percent of U.S. search queries. Almost half the remaining searches are funneled through properties owned and operated directly by Google.” The complaint concludes, “Google has thus foreclosed competition for internet search. General search engine competitors are denied vital distribution, scale, and product recognition—ensuring they have no real chance to challenge Google.”
The question that arises is, what next? Let us assume for the moment that the DOJ is right in its market definition and market power assertions. Let us also assume (another big assumption) that the default search engine agreements reflect an exercise of market power on Google’s part. The complaint’s requested relief to address this behavior is vague. The complaint simply asks for an injunction preventing Google from engaging in anticompetitive practices and “enter structural relief as needed to cure any anticompetitive harm.” From this relative silence, it might be possible to infer that DOJ wants the court to strike at the default search engine agreements that, to hear the DOJ, are one of the roots of the competitive evil the complaint sets to uproot.
The inference goes like this: forbid Google to enter into such agreements, or require their unwinding, and the harm described in the complaint will be undone. There are two fundamental and related mistakes about this inference. First, there is no plausible connection between unwinding these agreements and promoting competition in the search engine market as defined by the DOJ. Assume away the default search engine provisions that have so exercised the DOJ. What then? Either other companies may be allowed to bid to receive these default rights instead of Google; or no one will be allowed to receive such rights. But none of this is likely to mean that a new vibrant search engine competitor will spring like a phoenix from the ashes of Google’s erstwhile contractual rights.
After all, the Bing search engine is owned by Microsoft, which also controls the Edge browser and has accumulated one of the largest concentrations of wealth the world has ever seen (with some $14 billion in profits in its most recent quarter alone). Another browser, Safari, is controlled by Apple, which matches or exceeds Microsoft in the annals of history’s wealthiest and most profitable companies. Both Apple and Microsoft could easily afford to outbid Google for these default rights (or in Apple’s case refuse to sell them to Google) if these rights were capable of catapulting them into an effective search engine competitor. Give me a place to stand, and I can move the earth, said Archimedes. Default search engine rights do not appear to provide this leverage. To move the earth, one may not stand on it. Neither Apple’s Safari 16 percent share nor Mozilla Firefox’s 7 percent share provide enough of such an extraterrestrial foothold. After all, Google’s Chrome browser would still enjoy 60 percent share of browser usage.
And suppose that DOJ had an answer for this: to propose that the court require the divestiture of Chrome. Here again, the effectiveness of such a remedy might be undone by the DOJ’s distaste for default search engine preferences. The main revenue source for an independent browser seems to be precisely the amounts paid by Google for these preferences. And that brings us to the second problem with the misguided knee-jerk inference — first, kill all exclusivity — that could be drawn from the DOJ’s complaint. If Chrome were divested but then prohibited from entering into such agreements with its former affiliate Google, one hand would be taking away what the other gives. If the idea is to help promote competition in the search engine space by making a divested Chrome browser into a lever for such competition, then a prohibition on default search engine agreements is like giving the lever but taking away the Archimedean ground from which to wield it. It would be self-defeating if a purported remedy to promote competition in the search engine market took away tools from those best equipped to become competitors.
Of course, antitrust law has developed a nuanced view of exclusionary agreements, recognizing their anticompetitive risks, as well as their possible efficiencies and procompetitive potential. Even exclusive dealing agreements reached by market dominant entities are not automatically considered illegal — they are instead analyzed under a case by case standard — the “rule of reason.” Under this standard, exclusive dealing can only be unlawful if there is foreclosure of a “substantial share” of the market in question. To determine whether foreclosure is “substantial,” the Supreme Court has directed courts to examine “the relative strength of the parties, the proportionate volume of commerce involved in relation to the total volume of commerce in the relevant market area, and the probable immediate and future effects which pre-emption of that share of the market might have on effective competition therein.” Tampa Elec. Co. v. Nashville Coal Co., 365 U.S. 320, 328 (1961).
Communications law, too, has some useful teachings in store. Witness the treatment of exclusive deals in Federal Communications Commission (“FCC”) program carriage cases. Independent programmers face enormous difficulty in obtaining carriage on the distribution platforms of vertically integrated conglomerates, whose holdings extend to distribution and content, such as AT&T (cable, satellite, broadband access, Time Warner media) or Comcast (cable, broadband access, NBC Universal).
Let us now assume that the only way for a small independent programmer to gain carriage is to grant some exclusivity rights to the vertically integrated distributor. Does outlawing exclusivity help promote competition in the market for cable programming? Or does it set back the case of competition by eliminating the only hope of that independent programmer to have a shot at improving its ratings and to become an effective competitor in that space? The FCC’s program carriage rules resolve this conundrum by stopping short of a per se prohibition on exclusivity. On the one hand, distributors are prohibited from imposing exclusivity as a condition on carriage. But it is up to the independent programmer to invoke this prohibition by filing a program carriage complaint. The same goes for the related rule that vertically integrated distributors may not discriminate in favor of their programming affiliates and against independent programmers. Nothing keeps the independent programmer from accepting a discriminatory agreement under protest and then complaining about discrimination.
None of it is to say that all is well in the programming market. In fact, the enforcement of the program carriage rules by the FCC (or the lack of such enforcement) has done little to mitigate the market power enjoyed by the large, vertically integrated programmers. But the point is that the program carriage rules do not worship in the shrine of an abstract principle — no anticompetitive exclusives, and do not elevate such a principle above the practicalities of what it takes to promote competition.
2. Find Some White Knights
Back in Ithaca a few thousand years ago, Penelope had more than one hundred suitors vying for her favors. None of them succeeded. The problem for markets dominated by one company, as the search engine market (if it indeed is a market) seems to be, is that the suitors are fewer, and their likelihood of success is not much better. Demanding that Google stop doing anticompetitive things, or even structural divestitures, is all well and good, but it is not clear which firms would be empowered by such steps to take Google on in the search engine space. The complaint establishes no link between the hinted-at narrow cure for the practices it describes and the broad cure for an uncompetitive market that is necessary.
The best way to promote competition is to nurture one or more competitors. Here, too, the experience of the communications industry is instructive. In particular, both the court and the DOJ could take a page from the remedy that the DOJ itself helped forge to cure the anticompetitive effects that the merger of T-Mobile and Sprint would otherwise produce. Granted, that case involved a remedy for a merger, not for monopolization claims. But it is relevant nevertheless here for a number of reasons. Most important, it suggests that the solution of nurturing a competitor depends on three key ingredients: first, identifying the entity or entities that already possess key inputs necessary to make them a competitive threat; second, ensuring that these inputs do not come with anticompetitive baggage of these entities’ own; and third, endowing these entities with additional tools that will enable them to compete, including the temporary use of the incumbent’s network, appropriate divestitures, and a revenue stream that will facilitate construction of their own network.
Let us look more closely at the remedy approved by two agencies (the DOJ itself and the FCC) and two courts in the T-Mobile/Sprint case, and then explore its implications for the Google case. The white knight in that case was DISH Network, which the remedy teed up as the new nationwide fourth wireless carrier to cure the effects of what would otherwise have been a four-to-three merger.
Here are some of the remedy’s relevant features:
- T-Mobile was required to divest its prepaid wireless business operating under the Boost brand to DISH Network. T-Mobile is also required to lease part of its wireless spectrum to DISH.
- T-Mobile had to make other assets available to DISH, including decommissioned cell phone sites and retail locations.
Creation of New Competitor:
- In turn, DISH is required to offer retail mobile wireless service nationwide within a year of acquiring Boost.
- Under conditions between DISH and the FCC, and adopted into the DOJ’s consent decree, DISH is also required to build out a nationwide 5G network by June of 2023.
- As DISH works to build out its own wireless network, it has full access to T-Mobile’s network for seven years.
The DISH remedy had all three of the aforementioned key ingredients: DISH had almost as much spectrum holdings as Verizon, and so had the required wherewithal — enough spectrum to become the nation’s fourth wireless carrier; DISH has no market power in any relevant market, and so the remedy did not risk creating other unwanted competitive repercussions; and DISH was empowered to become a fourth carrier through the right to use T-Mobile’s network while DISH builds out its own network, and through revenue streams from the acquisition of the Boost service and from a temporary lease of some spectrum to T-Mobile.
Here, too, the white knights, which may extend to a variety of stakeholders without market power, must be provided with a full chain-mail suit of armor, shield, and spear, including an enhanced revenue stream and contractual rights. And of course, it would also be helpful if no Odysseus shows up to chase away the suitors: limitations on the conduct of dominant incumbents at all links of the online chain may need to round out the remedy package.