|
While the US was enjoying the fireworks and other celebrations on July 4, Bloomberg popped a cracker by reporting that the US Department of Justice (DOJ) is on the cusp of approving the controversial $26.5 billion proposed merger of Sprint and T-Mobile. This, despite 10 state attorneys general filing a lawsuit in March to block the merger.
Undoubtedly, DOJ approval was made more likely by a deal revealed two days earlier between T-Mobile and satellite provider Dish Network. In the deal, T-Mobile would divest wireless spectrum to Dish and provide Dish access to its (merged) network for a number of years, to enable Dish to operate first as a Mobile Virtual Network Operator (MVNO) with a view to ultimately become a full-fledged fourth mobile network operator. Similar undertakings have been significant sweeteners in smoothing the path toward approval of several so-called “four-to-three” network mergers in Europe. Notable cases are H3G and Orange Austria, Telefónica Deutschland and E-Plus in Germany, and Hutchison 3G UK and Telefónica Ireland (O2).
SCF and fast-moving technologies
At the crux of these cases is the increasingly dominant view in antitrust academic scholarship that the historical Structure-Conduct-Performance (SCF) measures are not particularly informative in technologically fast-moving industries where there is considerable uncertainty as to how the industry will develop in even the short-to-medium-term future. These include measures such as the Herfindahl-Hirschman Index (HHI) of industry concentration and the Lerner Index (measuring markups) before and after a proposed merger, as well as evidence of alleged price increases following similar past mergers, which have served antitrust analyses well in the past and from which legal precedents have been derived.
These are precisely the circumstances facing the mobile telecommunications industry in which the T-Mobile-Sprint merger is positioned. The industry is on the cusp of deploying very high-capacity 5G networks capable of supporting a ubiquitous Internet of Things environment. These networks differ from legacy networks because of the need for comparatively larger investments in infrastructure deeper into the network (requiring a larger number of masts and towers covering much smaller “cell areas”) and in blurring many distinctions between core network (data transportation) and data processing functions in last- and middle-mile segments. These features pose significant challenges to the established economics of mobile telecommunications network operators. And they have led to many different proposed scenarios about how even established network operators will respond, let alone potential new entrant operators or existing MVNOs.
4 to 3 to 1?
For example, in many countries (including the US) it has been proposed that network operators should engage in much more extensive network sharing arrangements than have been observed to date to facilitate faster and wider deployment than if each firm deploys its own infrastructure. Some proposals (e.g., Chorus in New Zealand and Rivada Networks in the US) go so far as to suggest that one firm alone should build the 5G network and function as a wholesale-only operator selling services to other operators to resell in MVNO fashion. Even if this option is not selected, it appears likely that network operators will seek to find new ways of sharing 5G network assets in the same way that they have collaborated (at times via third parties) in sharing passive infrastructure such as towers in their 3G and LTE networks. To prevent a 4-to-3 merger in an industry that may well be destined for even greater levels of concentration in many different parts of the supply chain through innovative new contractual agreements appears futile.
This leads to the question of how the DOJ or the judicial panel hearing the case brought by the 10 state attorneys general will select a counterfactual depicting what is expected to occur if the merger is blocked, against which to compare the proposed effects — either good or bad — anticipated if the merger proceeds. At a time when even the industry itself is struggling to understand how deployment of the new 5G networks will evolve, it is far from clear how antitrust analysts should respond. To choose any counterfactual at all invokes a risk of selecting the wrong one. However, the odds seem to be stacked against the status quo ante of four independent mobile network operators competing for the traditional business of residential and business voice and data service customers. This is due to long-heralded machine-to-machine communications emerging as a significant new market segment or perhaps even the dominant use of new 5G infrastructure.
When in doubt, do nothing
So how should the relevant authorities proceed? As a general rule in decision-making under uncertainty, the best approach is to make no decision but rather wait for more information to be revealed. This goes a long way toward avoiding the “bad news consequence” of taking action when subsequent events prove that doing nothing would have been better, because the consequences cannot be easily undone. For regulatory and judicial authorities in the case of a 4-to-3 merger such as T-Mobile and Sprint, the best thing to do at the current time is probably not intervening in the market by preventing the merger.
If subsequently the industry evolves in such a way that parties with market power can and do act anticompetitively in ways that harm long-term welfare, then other tools in the antitrust tool kit can be invoked to deal with the consequences.
When conditions are highly uncertain, the best thing to do is generally to wait for more information to come to light. Regulators should take that to heart in approaching the Sprint and T-Mobile merger.
|
