Actually, there are several ways AT&T’s attempted purchase of T-Mobile could be illegal, the most obvious of which is if the Department of Justice (DoJ) concludes that the deal is “substantially likely to lessen competition” in violation of the antitrust laws. The next most likely way would be for the FCC to find that transfer of the licenses would be contrary to “the public interest, convenience, and necessity” under Section 310(d).
But, as we pointed out in our comments to the FCC, it turns out there is a third way. AT&T’s acquisition of T-Mobile violates Section 314 of the Communications Act. This rather obscure and wordy section so rarely applies that, unless you are the kind of total telecom wonk without a life who actually reads through the entire Communications Act to see what’s in it, you’ve probably never heard of it. However, for reasons I shall explain below, I am fairly confident it ought to apply to this particular case and, if I am right, it creates an absolute prohibition to the FCC granting permission for Deutsche Telekom to transfer T-Mobile USA to AT&T.
If you read through the extremely long, wordy, and somewhat confusing statute, you might think that it generally tells the FCC not to transfer licenses where the effect would be to “substantially lessen competition” or “create unlawful monopoly in any line of commerce.” Indeed, I made an argument that the FCC should interpret the statute that broadly back in 2005, when I was at Media Access Project challenging the Adelphia/Comcast/Time Warner transaction. Needless to say, I lost. Perhaps unsurprisingly, the FCC has worked very hard over the last 75 years to narrow the scope of Section 314 as much as possible. According to the FCC decisions interpreting Section 314, it only applies when:
1) A company that owns a “cable or wire telegraph or telephone line or system” that sends/receives common carrier traffic between any place in the United States and any foreign country seeks to acquire –
2) A company that uses “radio” (meaning any wireless) to send/receive common carrier traffic between any place in the United States and any foreign country, AND
3) The transaction would “substantially lessen competition” or unlawfully create a monopoly in the exchange of common carrier traffic between any place in the United States and any foreign country (and possibly in related services, but the FCC just about never gets this far).
The FCC has also hinted, at various times, that the transfer of assets also has to involve a foreign company so that there is a transfer of assets “between any place [in the United States] and any foreign country,” not just an exchange of traffic between any place in the United States and any foreign country. Also, although the statute does not say that it has to be common carrier traffic as opposed to, say, broadband data, the FCC has interpreted it fairly consistently as meaning “common carrier” traffic, aka Title II, aka telecommunications.
So lets run through the checklist.
1) AT&T is clearly a company that is in the business of sending international common carrier traffic by “cable or wire telegraph or telephone line or system” between various places in the United States and foreign countries. Sure, they do a lot of other things as well. But, as AT&T pointed out in the filing, the company acquiring T-Mobile is AT&T, Inc., not simply its wireless subsidiary, AT&T Mobility. This factor would appear to check out.
2) Deutsche Telekom (DT) is based in Germany which, when last I checked, is a foreign country. They are also a telephone company, and use telephone systems to send common carrier traffic between the United States and foreign countries – like Germany.
3) T-Mobile USA is a wholly owned subsidiary of DT. They also provide common carrier traffic between the United States and foreign countries. Notably, they do so by “radio” rather than by wire.
4) AT&T and DT will exchange assets, after which AT&T will own T-Mobile USA. As noted, DT is in a foreign country, so to the extent it requires not just an exchange of assets, but an exchange of assets domestic and foreign, this requirement also appears to be met.
So we rather unequivocally have the right players. But does the merger impact the actual transmission of all this common carrier traffic flowing into and out of the United States? It is worth noting that when AT&T faced a challenge under Section 314 when it received a license to set up a “radio telephone” system back in 1957, the FCC decided that Section 314 did not apply because “it is clear it does not involve . . . any transmission or communication between the United States and any foreign country.” American Telephone and Telegraph Co., 22 FCC 1221, 1223. So maybe T-Mobile USA is just a “domestic” facility and doesn’t count for Section 314 purposes.
Even leaving aside the involvement of DT, however, we have an unambiguously international traffic/facilities issue present in the merger. As the record makes clear, one of the services offered by both AT&T and T-Mobile is something called “international roaming.” The Commission very helpfully (from my perspective) defined “international roaming” in this 2007 Order as a Title II telecommunication service in which U.S. carriers and foreign carriers agree to resell each other’s network capacity to make calls that terminate either in the U.S. or in foreign countries. So not only are AT&T, DT, and T-Mobile USA exactly the kind of entities described in the statute under the most limited application of the kind imaginable, but international roaming is exactly the kind of traffic to which Section 314 (according to the FCC) applies.
But wait! We still haven’t actually proven the case yet! How do we know that the transfer by foreign common carrier DT of the “stations” “apparatus” or other “system” for sending common carrier radio signals “from any place in any state to any foreign country” to AT&T, the domestic wireline common carrier, will actually “substantially lessen” competition for international roaming? After all, as the Supreme Court held in one of the few cases actually interpreting Section 314, the statute does not require the FCC to automatically give a competing license to a competitor if there is sufficient competition. So how do we know if there is still enough competition in international roaming to avoid triggering Section 314?
Again, we don’t need to take my word for it. Turns out a number of foreign carriers, like Vodafone and Japan Communication, as well as a New Zealand ministry, and a few others (such as the Rural Telecmmunications Group, MetroPCS/NTelos, and the International Users Group (INTUG)) have all squarely raised the international roaming question and protested that losing T-Mobile means going from 2 national GSM-based networks down to 1. Even if we adopt AT&T’s standard of only looking at local markets, ignoring national markets, and assuming all carriers are equal, you still have many local markets where you drop from 2 GSM-based carriers networks to just AT&T. And in the markets where you have another local GSM-based carrier, the number drops from 3 to 2.
Why does that matter? Because GSM is the standard for almost every other country in the world. To do international roaming with most of the world, at least until we have broader LTE deployment here and in other countries (where LTE deployment is even slower than here), you need a GSM-based partner. So the presence of Sprint, Verizon, or any other CDMA-based carrier is irrelevant to the impact on international roaming. Post merger, in most markets, it’s either AT&T or no one.
I would call going from 2 possible international roaming partners to only 1 possible international roaming partner a “substantial loss” of competition or creation of a monopoly. Nor am I alone. Pretty much every party filing on the subject made the same observation. They didn’t cite Section 314, mind (it is rather obscure), but they all stressed the same thing: after the merger the number of GSM-based international roaming partners goes from 2 to 1.
AT&T does not deny in its replies that in many markets the number of GSM-based networks available for international roaming will fall from 2 to 1 as a result of its gobbling up T-Mobile. Instead, AT&T argues that this doesn’t really matter because (a) the Commission should totally ignore any national market, so going from 2 national GSM carriers to 1 doesn’t matter; (b) in many markets it is already the monopoly provider; (c) LTE will get here eventually; and (d) AT&T would never have any incentive to exploit its GSM monopoly to raise international roaming rates. They also note that travelers that don’t like dealing with AT&T can always buy dual-use phones that can also work on CDMA – although that does not exactly help the foreign GSM-based carriers that need to negotiate rates with AT&T for the bulk of their customers who don’t want to incur the additional expense of buying a dual-use phone.
Leaving aside whether this is a good enough answer under the public interest standard (I don’t think so, but that’s up to the FCC), Section 314 does not care about any of this. Section 314 absolutely prohibits the transfer if it substantially lessens competition or creates monopoly “between any place in any state and any foreign country.” AT&T does not (and cannot) deny that the merger will – in a large number of markets – bring the number of possible roaming partners for international roaming down from 2 to 1. If Section 314 means anything, it surely covers a drop from 2 to 1, and should probably also cover a drop from 3 to 2 in those markets where a regional GSM provider remains. (It should also cover the loss of national providers going from 2 to 1, but I’m giving AT&T the benefit of its choice of market definition).
Hence my conclusion that the deal violates Section 314 – at a minimum for those markets where it drops the number of possible international roaming partners for 97% of the countries of the world from 2 to 1. And there are so many of these markets that AT&T will acquire a GSM monopoly that divestiture is pretty much out of the question. Besides, as noted above, Section 314 really ought to apply to the drop in the national market from 2 to 1, or, alternatively, regions where the number of international GSM-based roaming partners goes from 3-2. But even reading the statute as narrowly as possible, it still seems to me that AT&T is stuck and the deal is illegal.
AT&T’s Likely Counter-Argument
Needless to say, AT&T will not simply take my word for it. They have lots of good lawyers, and we all learn in law school that any case has a counter-argument and it is up to you as a lawyer to find that counter-argument and push it for all it’s worth. I anticipate a mix of ad hominem (especially from AT&T’s supporters) and non-substantive attacks (“Look, Harold brought this up before in 2005 in an entirely different merger and it didn’t work then!”) with some actual substantive replies. AT&T’s best argument is some old cases from the 1930s (called Mackay Radio) that said that the FCC could “lawfully” create monopoly and therefore issue a license for a transatlantic line to a pre-existing monopolist. Leaving aside the political problem of AT&T arguing that the FCC should use its authority to protect the “natural monopoly” of common carrier from “destructive competition,” Congress eliminated the ability of the FCC to insulate common carriers from antitrust and create “lawful” monopolies in 1996. (AT&T has the same problem with the aforementioned Supreme Court case, which in addition to finding that the FCC need only find there is “enough” competition also rested on this idea that there was no “national policy in favor of competition.”) Still, I expect AT&T to argue that Mackay shows the FCC has discretion to grant the merger anyway, despite the fact that the plain language of the statute says the FCC does not.
AT&T’s other argument is that the GSM monopoly shouldn’t count. But again, I don’t think that argument carriers the day. These are clearly facilities for international communication (among other things), and the transaction drops the number of potential international roaming partners for 97% of the world. That surely qualifies as “any foreign country.” But as I said, it is the purpose of lawyers to argue that plain language such as “between any place in any state and any foreign country” does not mean what it actually says.
In any event, we will see if poor little Section 314 will finally be enforced after 75 years of being narrowed, qualified, and generally ignored. It is easy for the cynical to believe, as the Mikado observed, that “it’s an unjust world, and virtue is triumphant only in theatrical performances.” Fortunately for me, this transaction already has a heck of a lot running against it. The fact that the acquisition is not merely a bad thing, but also violates the law, should (I hope) tip the balance in the decision.