Should AT&T have to divest its broadband wireless holdings?

Comment With the regulatory review of the proposed AT&T/BellSouth merger looming, one issue is whether the combined company would be able to hold on to its 2.3GHz and 2.5GHz spectrum holdings.

Some lobbyists have picked on the wrong argument from the start, however, claiming that because broadband wireless in these bands would compete with the Bell operator's wireline services, AT&T would be likely to keep the spectrum dormant. In fact, although all the Bell companies have failed to use this spectrum for years – taking a swathe of valuable wireless real estate out of the market – now is the time they are most likely to dust it off in order to expand their triple play coverage into markets where they have no lines installed; to build a licensed band metrozone to try to see off companies like EarthLink; or to combine wireless with their DSL services to offer high value broadband bundles that include portable access.

Even without those commercial pressures, the FCC can certainly take the opportunity of the merger to insist, as it did with the Sprint Nextel merger, that AT&T uses its 2.3GHz assets or loses them.

The real issue is whether one company should be allowed to keep such a huge range of broadband network options – DSL, fiber, a global network of long lines, the Cingular cellular networks, and broadband wireless - or whether competition would be better served by sharing things around a bit more.

In many ways, with Verizon rejuvenated and the cablecos getting stronger, AT&T can argue it faces more effective competition in its markets than it has for years. And while the company looks absurdly powerful, certainly in terms of any smaller challenger having a chance of success against it, would the removal of the 2.3GHz holdings really make a difference?

Sprint, of course, was allowed to keep its 2.5GHz holdings post-merger, on condition it created a network and services by 2009. But Sprint has a far less impressive set of networks than AT&T, being effectively a wireless-only quadruple play operator now, especially once it spins off its local lines unit, Embarq. However, its broadband wireless assets are far more valuable than those of AT&T, partly because they cover a higher percentage of the US's markets and, in many areas, with large quantities of bandwidth, such that Sprint can create the ambitious strategy that it has to deliver a challenge to the Bells in association with the cablecos.

Weaknesses of the 2.3GHz band

Although BellSouth is rolling out a pre-WiMAX network in some territories using Navini kit in 2.3GHz, the potential of this spectrum for AT&T is far less than for Sprint because it has far less bandwidth.

The difference between 2.3GHz and 2.5GHz may not be important technologically, but in the US the two bands are a very different proposition. The main holders - Verizon, BellSouth, AT&T, and the owners of the assets of former operator Metricom - have let the licenses gather dust, partly through fears of cannibalising their other services and partly because the band is more difficult to exploit in America than elsewhere.

This is because there is limited space compared to other countries – 30MHz in two channels, from 2305MHz to 2320MHz and from 2345MHz to 2360MHz – and those two channels are separated by the DARS (Digital Audio Radio Service) band. This raises possible interference problems from DARS satellite radio terrestrial repeaters, which have high power limits (2,000 watts EIRP).

The satellite radio operators, Sirius and XM, having paid $80m each for their DARS frequencies, are highly protective of their signals, and even tried to limit 2.4GHz Wi-Fi. That move failed, but their aggression can cause problems for operators in 2.3GHz.

Technologies like WiMAX are able to make better use of 2.3GHz, through their spectral efficiency, than other equipment has in past times, but it is still not reasonable to compare AT&T's spectrum with Sprint's as an asset. Whether the merged Bellco keeps or loses these licenses will not have a huge impact on the broadband pattern of the US.

Where the issue could be more important is in the cellular side of the business. Increasingly, the 3G technologies are expanding into 2.5GHz, with the European regulators having set aside this band for 3G expansion, for instance. So technologies like TD-CDMA and even CDMA2000, plus the LTE next generation of W-CDMA, will cross the divide between current 3G bands below 2GHz and the broadband wireless spectrum.

This is another advantage Sprint has, since it will be able to run its CDMA EV-DO network and 2.5GHz systems in parallel, and may well choose a next generation network that will provide an upgrade to both systems at once (an option Qualcomm is pushing with the concept of using EV-DO and Flarion Flash-OFDM, with the promise of a common upgrade in future. IPWireless holds out a similar promise for a dual strategy based on W-CDMA and its own TD-CDMA, both then moving up to LTE in future). This could bring 2.3GHz into play as a valuable addition to the strategy of Cingular, the AT&T-BellSouth cellular joint venture.

And if the FCC decides that is too great an advantage for the cellco, there would be a golden opportunity for T-Mobile USA, which badly needs new spectrum, to acquire licenses to create its own next generation wireless network.

Another contender, of course, would be Clearwire, seeking to add to its own stockpile of 2.5GHz spectrum. This scenario argues for the divestment of AT&T's 2.3GHz – not because this would have a material effect on its core broadband business or the competition in that market, but because it could give a small but valuable leg-up to one of the mid-sized operators that represent one of the only hopes – apart from the Sprint-cable venture – of chipping into the dominance of what will become just two senior Bellcos.

Copyright © 2006, Wireless Watch

Wireless Watch is published by Rethink Research, a London-based IT publishing and consulting firm. This weekly newsletter delivers in-depth analysis and market research of mobile and wireless for business. Subscription details are here.