Lowenstein's View: How many cellular networks are necessary?

Mark LowensteinCuriously, the FCC has been preoccupied with concentration in the wireless industry, worried that consolidation has resulted in a less competitive mobile market. I'm not sure their reasoning adds up, given users' choice of as many as seven facilities-based providers in some markets, plus a growing selection of MVNOs and sub-brands. If we agree that data is going to be the major growth driver over the next several years, I would actually argue that we are headed toward a market populated by too many facilities-based network providers, which is resulting in redundant builds, compromises on quality and capital inefficiency. The result might be some high-profile market failures.

Let's look at the United States wireless landscape. Among the four national wireless operators, AT&T Mobility and Verizon Wireless are healthy, while Sprint Nextel and T-Mobile USA have been struggling. Between MetroPCS and Leap, there is near universal agreement that they need to merge or be acquired by another operator, given their lack of national footprint and the hyper-competitiveness of the pre/advance-pay market. Then there's Clearwire, which is having trouble raising the required capital to complete its build-out. And finally LightSquared, which is embarking on its highly speculative and still not adequately financed plan of building a wholesale 4G network. I'm not quite sure how investors can support LightSquared's proposition when the 4G buildout strategies of all but Verizon and AT&T remain unresolved and/or underfinanced.

If we are to assume that all of the major facilities-based operators believe that having a really good 4G network is necessary to be competitive over the next 7-10 years of wireless, a back-of-the-envelope calculation takes me to north of $300 billion capital outlay. This is assuming they all build to a minimum 75 percent of their licensed footprint and meet capacity demands, which will also require purchase of additional spectrum in many cases.

All of the sudden, folks on the landline side of the world look downright conservative. Both Verizon (FiOS) and AT&T (U-Verse) have slowed expansion of their fiber network builds, while cable firms such as Comcast are upgrading their networks at a fairly measured pace.

Now, let's look at the demand side of the equation. I don't dispute that numbers being tossed around by Cisco and Ericsson and other big infrastructure firms, showing X hundred percent increase in demand for wireless data over the next several years. However, I would throw a little nuance into this argument. First I do believe that individuals will require near-constant broadband network connectivity, especially for their primary mobile device--in most cases, a smartphone. That connectivity will be a combination of cellular and WiFi (where available and free). Note that in a relatively short amount of time, smartphone users have done a pretty good job of training themselves to toggle between the WAN and WLAN. But the next wave of mobile devices, which represents the underlying business case for 4G builds, will require a more occasional, infrequent need for WAN connectivity. Devices such as the iPad are not primarily mobile devices and are out of WiFi reach only some of the time. And among laptop users, outside the 15-20 percent classified as "road warriors," mobile broadband requirements are also occasional.

We are seeing these varying requirements in the new way of thinking about wireless data pricing. First, there's the usage-based model, where prices align more with consumption and demand. Then there are the increasingly popular pay-as-you-go price plans, such as Virgin Mobile Broadband2Go or Clear's recently announced Rover, which have flexible per-day/week/month plans, not requiring a contract. This gives the user a lot more flexibility. Why spend the extra $200 for a 3G iPad when you can spend circa $100 for a MiFi/iSpot/Overdrive type device, and buy mobile broadband service from the operator you want, use it when you want, on any device with a WiFi connection? Further, given the gradual, relative patchwork 4G network buildout, users will not want to be tied into a contract relationship with an operator for ancillary devices.

This brings me back to the discussion of how many wireless broadband networks are really needed. Even though we might be moving toward a more varied model of when and where people want connectivity, the ante for what represents a quality experience will increase. After all, these additional devices will be doing more video, connected gaming, mobile VoIP, tele-medicine, and so on. So we must ask the question: Do we want five or six 4G networks, some of which will be built over a lengthy period, to a relatively limited footprint, with lots of backfilling to 3G, or would we prefer three or four really good 4G networks? There's a disproportionate amount of capital involved in taking a wireless network from good to great--along the variables of density of footprint, in-building coverage, higher average throughput, and so on.

I personally believe we should be thinking in terms of three facilities-based networks, and perhaps one wholesale network. This means, in the current nomenclature, Verizon and AT&T. Out of the remaining facilities-based providers (Sprint, Clearwire, T-Mobile, Metro PCS), Leap, US Cellular, etc.), emerge, ultimately, two really good nationwide networks. This will, of course, require some consolidation and reshuffling of the current landscape. It will mean more in the way roaming agreements (such as between Leap and Sprint), wholesale arrangements (such as in the discussions between T-Mobile and Clearwire), and more network sharing. There could be some scenarios where multiple stakeholders contribute capital to the buildout.

This allows the U.S. to retain and extend what has become a leadership position with respect to the availability of mobile broadband, while recognizing the challenges of providing the capacity, quality, comprehensive coverage given the country's size and demographic patterns. In the case of mobile broadband, fewer just might mean better.

Mark Lowenstein, a leading industry analyst, consultant, and commentator, is Managing Director of Mobile Ecosystem. Click here to subscribe to his free Lens on Wireless monthly newsletter, or follow him on Twitter.

Read more on