What's Good for AT&T Is Good for America AT&T

A talking point is a carefully constructed edifice, and no one builds an edifice without placing something behind it. On Mar. 21, the day after AT&T (T) announced its intent to buy T-Mobile, three AT&T executives spoke to analysts on a conference call and converged on a set of terms. "Infrastructure," they said, and "investment," and "the President's wireless broadband goals." And they all agreed: A merger would bring high-speed LTE, or 4G, wireless access to 95 percent of the U.S. population.

AT&T's purchase would offer unambiguous benefits to its shareholders and to those of Deutsche Telekom, the German former state monopoly that had been unable to fix T-Mobile, its underperforming U.S. wireless subsidiary. T-Mobile and AT&T use compatible technology, and the latter can more efficiently use the spectrum that would have soon proved inadequate for T-Mobile alone. It's a good fit, which means the real sport in this merger has never been in AT&T's hometown of Dallas or in Bonn, where Deutsche Telekom is based, but in Washington, where the Justice Dept. and the Federal Communications Commission must approve the merger. When the FCC looked at Comcast's (CMCSA) purchase of NBCUniversal, the negotiations swung on network neutrality and conflicts of interest. This time around, look for one word: infrastructure.

The U.S., unique among developed economies, relies almost exclusively on private capital to fund its telecommunications infrastructure. Cables, ditches, and cell towers cost a lot of money, which creates tension between carriers and the federal government. Better infrastructure gooses the economy; the government, since it's not paying, always wants more of it. The FCC decides whether a merger is in the public interest, and it tends to use this power to extract commitments to add capacity to the network. And the carriers feel pressure to make promises whenever they request any kind of regulatory or tax leniency. They seldom feel pressure to keep them. "We keep getting told that if we only give them x, y, or z, they'll build infrastructure," says Susan P. Crawford, who teaches at Yeshiva University's Cardozo School of Law in New York and has advised the White House on technology. "It's like an endless Lucy-with-the-football story, over and over again."

Bruce Kushnick, who runs the New Networks Institute, an advocacy group highly critical of telecom carriers, has compiled a list of mergers and infrastructure promises. In the 1990s, Pacific Bell in California, SNET in Connecticut, NYNEX in New York, and Bell Atlantic all secured state legislation that would ensure greater profits. In exchange, those carriers were to roll out high-speed wire-line Internet access. All are now part of either AT&T or Verizon Communications (VZ) and, according to Kushnick, all projects have either languished or been abandoned. (To be fair, in the past decade, Verizon has invested $23 billion on FiOS, its fiber-optic network in selected markets.) Similar patterns emerged after the mergers of SBC Communications and Ameritech (now part of AT&T), and Bell Atlantic and GTE (both now with Verizon), as well as the acquisitions of MCI, SBC, and Bell South in the past decade. Kushnick's simple advice for the FCC on mergers: Don't allow them.

Christian Sandvig, a telecom researcher at Harvard University's Berkman Center for Internet & Society, suggests looking at the differences between how a company presents a merger to the public and what it shows to investors. In AT&T's conference call with analysts, Chief Financial Officer Richard G. Lindner said the merger would allow AT&T to reduce future capital outlays. Other executives pointed out that, with T-Mobile in the fold, AT&T could reduce "churn"—or customers' frustrating tendency to seek other providers—and increase average revenue per customer. This is not to paint AT&T as rapacious but as capitalist. When carriers invest, they're thinking about shareholders, not America.

Sandvig points to the National Telecommunications and Information Administration's broadband map, completed this year. Originally, the project was to display—by street address—broadband availability, technology, speed, infrastructure, and revenue per customer. It was a relatively anodyne move, since the carriers have access to this data through private services. Yet, as detailed in a 2009 letter to the NTIA signed by most carriers (including AT&T and Verizon), the telecom industry collaborated to replace the speed experienced with the speed advertised, display connections by street segments rather than by home address, and remove completely the requirements to share infrastructure data and revenue per user.

AT&T says it's confident the merger will be approved; it gave what it described as a "heads-up" call to the FCC before the announcement, and has offered a "breakup fee" of $3 billion and some spectrum to T-Mobile should the deal fall through. And the company's new refrain—LTE wireless Internet access to 95 percent of the U.S. population—is phrased to sync with the goals laid out in the national wireless plan announced by the Obama Administration in February. The offer, in essence: Give us the merger and you can declare victory on your wireless plan.

Before agreeing to anything, FCC Chairman Julius Genachowski should ask some pointed questions about this 95 percent. Where will it be rolled out? When? What average speed can users expect in each market? And he should get the answers in writing.

The bottom line: To get merger approvals from Washington, telcos promise big infrastructure investments. Data show they often don't come true.

    Before it's here, it's on the Bloomberg Terminal.