Are U.S. Mobile Carriers Customer Bases Differentiated?

It is a commonplace observation that the largest four U.S. mobile service providers are differentiated on the dimensions of churn and average revenue per account. Basically, Verizon Wireless and AT&T Mobility customers churn at half the rate of Sprint and T-Mobile US customers. 

But there might also be a significant differentiation based on why customers choose their service providers. 

When Cowen and Company analysts asked customers why they chose their service provider, AT&T and Verizon chosen for "network coverage and quality," 

Sprint was chosen for "unlimited data plan and better price," while T-Mobile US likewise was chosen for "better price and unlimited data plan.

The distinctions are clear: customers who value coverage and quality tend to buy AT&T and Verizon. Customers who value unlimited data and price chose Sprint and T-Mobile US.

So the question, assuming you believe a big marketing war will escalate, is how big each of those customer segments are. For that might limit the gains either Sprint or T-Mobile US can gain and hold, over the long term. 

It will be easier for AT&T and Verizon to match price offers than for T-Mobile US and Sprint to dramatically expand their footprints. 

But all that assumes no major change in market structure. With the possibility that something happens with T-Mobile US (merger with another provider), and if Dish Network does enter the market, along with activation of assets from one or two of the mobile satellite firms that want to repurpose their mobile satellite spectrum, tomorrow's market could look different. 

But that will occur within a context where it appears customers fall into two broad camps: buyers who value coverage and quality, even at higher price; and customers who value unlimited usage and want lower price. 

Coverage might not matter for the latter, while lower price, while helpful, still is not why the former customers make their fundamental choices. 

How much the contestants can structure their operations to attract the "other" type of customer will become a bigger issue.

Spectrum Management Heading for a Historic Change

Spectrum not only is the foundation for all wireless and mobile services, it is a foundational matter for would-be service providers. And big changes are coming.

New spectrum formerly used for TV broadcasting is being reallocated to mobile communications. And at least some of that reallocation process will involve methods of adjusting the behavior of networks and devices dynamically, based on interference issues.

New thinking is happening about sharing existing spectrum as well. It simply is too expensive and time-consuming to conduct widespread "clear and auction" operations across much of the communications-capable spectrum below 3 GHz.

So much attention now is focused on how existing licensed users can be persuaded to share their frequencies with new commercial users as well. Incentives will play a key role: existing licensees will have to see clear financial benefits for doing so, and since so much licensed spectrum below 3 GHz is licensed to government and military users, such incentives will be tricky.

And the ways mobile operators use mobile licensed spectrum, unlicensed spectrum and fixed network assets is changing. Already, even without formal business relationships, perhaps 60 percent to 80 percent of mobile device Internet access occurs on Wi-Fi connections.

Some would say that reveals a key strategic weakness for mobile operators, whose costs of providing Internet bandwidth are too high to survive massive adoption of mobile video consumption.

And though there is more attention paid to global standards, the U.S. market will in some ways remain a bit different.

In some ways, the continental-sized U.S. market has allowed the U.S. communications business to develop based at times on local standards not shared with most of the rest of the world. The primary past example is use of CDMA air interfaces alongside GSM, where in most of the rest of the world GSM was the sole standard.

Frequency plans within the United States will be the salient example of this in the next phase of the mobile business, as most of the rest of the world tries to create a common 700-MHz band for mobile communications across Europe, Africa, Asia and Latin America.

Once again, because of past spectrum allocation decisions, the U.S. will remain a market without full frequency harmonization with most of the rest of the world. How important that might be is not so clear, though.

Global frequency coordination is helpful for manufacturers, as it allows more scale when developing handsets. It is helpful for international travelers who then can roam almost at will when traveling (assuming they don’t mind the international tariffs).

But advances in radio agility are important. In principle, it is possible to equip a device for frequency agility that can compensate for different frequencies used in different countries. And since all mobile carriers have agreed on Long Term Evolution as the common air interface standard (even if there are differences in modulation), frequency might ultimately be less an issue.

The issue for European regulators and industry concerns is whether the 2012

World Radiocommunication Conference of 2012, which decided to open up the 700MHz band across the EMEA region for mobile communications, will be able to come up with enough consensus to allow a further decision at the 2015 WARC meeting.

Of course, that process will be contentious, as TV broadcasters will have to be induced to surrender their use of spectrum to mobile interests. That has not proven easy, wheneve it has had to happen.

But such decisions now are among the growing range of ways national regulators are playing a fundamental role in setting the stage for the next phase of mobile communications growth. Though observers might disagree on how much additional spectrum is required, nearly everybody believes more spectrum will be needed.

And much of the focus will be on ways to find new ways to use spectrum already licensed for some other purpose, to some other users, as very little of the spectrum most useful for communications actually is unclaimed. That means a historically new look at ways to enable sharing of licensed spectrum, more use of frequency-agile networks and devices, and a new business role for unlicensed spectrum.

Erosion of Subscription Video is About 1% a Year

The strategic problem faced by traditional video subscription services is not that massive customer desertion is happening now. In fact, attrition, though real, is rather low at the moment.

According to The Diffusion Group, nearly 88 percent of all adult broadband Internet access users in the United States subscribe to a cable, satellite, or telco-TV service.

“The notion that we’re on the edge of a ‘mass exodus’ from incumbent pay-TV services to online substitutes is not supported by the data,” says Michael Greeson, co-founder of TDG and director of research.

The longer term problem is that younger consumers do not seem to buy the product as heavily as older consumers. And unless that changes, trouble lies ahead.

For example, TV subscription rates among those 25 to 34 are 82 percent, and 85 percent among those 18 to 24.

Note that the TDG metric is adoption of subscription TV among “broadband households.”

Of course, since household adoption of broadband Internet access is itself a percentage of all occupied U.S. homes, the adoption of video entertainment services as a percentage of all occupied U.S. homes might be  lower than 88 percent.

U.S. broadband penetration is estimated at 78 percent of U.S. homes. If 88 percent of those homes buy a video service, then video penetration hypothetically (some homes buy video service but not broadband or dial-up access) could be as low as 69 percent.

Nobody believes video subscription rates are that low, estimating that more than 102 million U.S. homes actually buy a subscription. That compares to about 115.6 million TV households altogether, including homes that only watch over the air TV.

If there are about 130 million occupied homes, that implies video subsription penetration of about 78 percent, oddly enough the same penetration rate as broadband.

60% of Surveyed Rural Telcos Offer Mobile, Fixed Wireless Services

Some 60 percent of about 31 rural telcos surveyed by NTCA:The Rural Broadband Association report they are providing “wireless service” to their customers, including both mobile and fixed wireless services, using a mixture of owned spectrum and facilities as well as resale or agency agreements with mobile service providers.

Some 82 percent sell fixed broadband Internet access (29 percent say they also sell fixed voice using wireless), while 49 percent sell  mobile services.

Of the respondents not currently offering wireless or mobile service, some 30 percent are considering doing so.

About 60 percent of independent and smaller rural U.S. telcos surveyed by NTCA:The Rural Broadband Association report they own at least one wireless license in the frequency range

below 2.3 GHz, according to NTCA.

Some 70 percent of carriers owning a license below 2.3 GHz have a 700 MHz license, 47 percent have an AWS license, 47 percent a PCS license, 13 percent some other license

(such as microwave), 11 percent cellular, six percent paging, and two percent SMR. For the most part, those are “access” frequencies.

About 21 percent of the 31 reporting companies hold at least one license above 2.3 GHz. About half of service providers holding a license above 2.3 GHz have a BRS license, 25 percent an LMDS license, 17 percent a 3.65 GHz, 17 percent a license at 11 GHz, and 17 percent a microwave license. While BRS and LMDS are “access” frequencies, the others likely are used for trunking.

The average cumulative investment in wireless facilities, excluding spectrum, is

$6.2 million, while average cumulative investment in spectrum totaled $646 thousand.

About 46 percent of respondents are using unlicensed spectrum to provide some

wireless services.

About 28 percent of those respondents offering mobile service  resell another carrier’s service under their own brand, and 21 percent do so under a national brand. So 39 percent “resell” or have an agency agreement.

Sprint Can Get Financing for T-Mobile Bid

Sprint reportedly has gotten assurances from at least 

o banks about the feasibility of raising enough money to finance a takeover of T-Mobile US. 

The proposals envision a total enterprise value (value of outstanding equity and debt) of about $50 billion for the deal.

It might cost $31 billion to buy all T-Mobile US shares and then also raise about $20 billion to cover the cost of assuming existing T-Mobile debt. 

But financing is likely to be the least of concerns about any such combination. The bigger issues involve antitrust clearance and approval by the Federal Communications Commission, since the U.S. Department of Justice already has said it considers the U.S. mobile market too concentrated already. 

Sprint to Enable Free Wi-Fi Calling?

Sprint reportedly is going to enable " no incremental charge" calling when users of at least a couple of Sprint phones are in a Wi-Fi zone, without counting against a customer's voice usage cap.

As reported, a customer will simply need a compatible device (possibly Samsung S4 Mini or Galaxy Mega), enable the feature through a web interface and begin using it. 

No monthly charge will be assessed for turning on Wi-fi Calling, apparently. 

Offering such features ("free calling") often makes sense for a service provider when the risk of cannibalizing revenue is slight and when the network is lightly loaded enough that a significant increase in usage will not stress the network and when there are other business reasons for encouraging usage. 

In this instance, the happy set of circumstances seems to include the fact that the Sprint 3G network will start to see less demand as traffic shifts to 4G (apparently the Wi-Fi calling feature requires presence of a 3G signal). 

The Wi-Fi calls are domestic only, and of course all the leading national providers have found that voice usage keeps dropping anyhow, freeing up voice network capacity.

In this case the 3G network resource primarily used might only be the signaling network, as most of the time only the signaling network actually will be consuming network resources. Only in case of an emergency call being placed would actual bearer traffic be imposed on the 3G network.

The new feature also might create some distinctiveness for 3G network access, the supported handsets and Sprint's access services in general. 

Spectrum Policy is Approaching a Revolution

Ever since the publishing of the U.S. National Broadband Plan in 2010, innovations in spectrum management have been at the forefront of thinking about the future of communications in the U.S. market.

In the United States, about half of all spectrum most suitable for communications, fixed and mobile, is licensed to various Federal government agencies, and, as you might well expect, much of that spectrum arguably is inefficiently managed.

Much of the new thinking centers on fundamental changes in the process whereby spectrum is made available for communications uses.

As a practical matter, though there are several ways to wring more effective use out of a finite spectrum resource, the absolute amount of spectrum useful for communications is limited.

As demand continues to grow, we bump up against physical constraints, even if demand shaping (Wi-Fi offload, tariffs), better technology (signal compression, better algorithms, agile frequency hopping) and network design (small cells, carrier Wi-Fi) can have meaningful impact.

Still, making better use of existing spectrum is among the tools policymakers can wield.

The President’s Council of Advisors on Science and Technology (PCAST) report makes a couple of nearly-revolutionary statements. Among the observations is the impractical method traditionally used to reclaim and then commercialize spectrum (“clear and auction”), simply because the costs of transition are so high.

The other really-novel insight is the observation that assigning spectrum in slivers, each with an assigned application profile, is inefficient, in light of modern technology.

In the wireless and mobile communications business, spectrum exhaust is a perennial problem. But there is new thinking that perhaps such scarcity is perhaps partly a result of allocation policies, not an absolute “shortage” of spectrum as such.