How Big a Problem is Growing Mobile Data Demand?

Replacing declining legacy revenues with new sources is a key strategic challenge for service providers. So a key practical issue is whether new broadband access, video entertainment or managed services will grow fast enough to match voice revenue diminution.

In fact, Rupert Wood, Analysys Mason principal analyst, warns “there were already signs two years ago that the shifting balance between smart phone and mobile broadband would result in lower growth of mobile data traffic.” 

In other words, the problem is not "exploding growth," but not enough growth. 

Wood predicts mobile data revenue growth in Western Europe, for mobile service providers, will drop in 2012, for example. And though it might have seemed an improbable development, Wi-Fi has become a virtual default network for most smart phone and tablet traffic.

That doesn’t mean untethered mobile devices are best served by Wi-Fi when users are on the go, but the most data-intensive traffic will tend to get consumed indoors, because that’s the natural place to consume it, not because it is the only place with good enough connectivity, says Analysys Mason.

the problem is that demand will not grow enough. 

“Pipe” Services Will “Always” Drive Most Service Provider Revenue

Communications service providers dislike the phrase “dumb pipe” for obvious reasons, since it implies–often falsely–that a telecom supplier is “just” a provider of low-value, commodity access services. The notion is partly accurate, but has nothing to do with profit margin on “dumb pipe” services.

What, after all, is “best effort” Internet access but a “dumb pipe” service? The access is one thing, while nearly all the content and services are provided by third party suppliers. But profit margins on U.S. high-speed access are in the 40-percent range, hardly a low-margin, commodity service.

There are threats, but those threats are potential future threats, such as ever-increasing amounts of supplied bandwidth, for the same, or less, revenue. But service providers already are moving to tie consumption to revenue in a more-logical fashion, so the potential danger is unlikely to surface as a present danger.

Also, very few service providers are “just” Internet access providers. Video, voice, security, business services and managed services are applications with lucrative revenue streams that use the access network. 

The point is that pipe services always will drive most access provider revenue. 

Questions About Investment Return for IP Networks as PSTN Shutdown is Pondered

As policymakers and service providers ponder the implications of a future “shut down” of the public switched telephone network, they will face a new series of questions.

Since the robustness of investment in new networks will hinge on expectations about profits, many of the questions will flow directly from the more challenging business model. As the regulatory framework had to change in the shift from a monopoly model to a competitive model, so the framework arguably will have to change again with a shift to all-IP networks.

When service providers worry about the implications fo “dumb pipe” or “over the top” business models, they simply acknowledge a key change in the foundation of the networks business, namely that network access and applications now are logically distinct.

Whatever else might be said, it is obvious that “certainty” is not a key feature of the fixed network business anymore. That’s a problem since uncertainty is the enemy of investment. 

FCC Initiates Review of Mobile Spectrum Policies

The Federal Communications Commission (FCC) has adopted a Notice of Proposed Rulemaking (NPRM) to reexamine its mobile spectrum holding policies, last reviewed comprehensively more than a decade ago. 

The FCC review makes sense, many would argue, in light of the greater importance of terrestrial mobile broadband, compared to fixed or mobile satellite services, and the shifting of spectrum to terrestrial use. 

Also, the coming spectrum auctions for former broadcast TV frequencies necessarily will raise issues about the quality and quantity of competitor spectrum holdings, pro-competitive measures that might be taken and an overall need for clear and stable rules to stimulate investment.

As always, the setting of new rules necessarily will create concrete business advantage for some contestants, compared to others, likely creating new opportunities for at least some new or smaller contestants. 

In related moves, the FCC also is looking at its licensing rules for satellite services, with a new emphasis on interference issues. That has taken on new significance in light of repurposed low-power satellite frequencies that will operate as higher-power terrestrial Long Term Evolution 4G networks. 

The FCC also is preparing to auction spectrum formerly used by broadcasters, to support LTE networks. 

All of those moves show the heightened importance of spectrum policy, at the moment, for broadband progress overall. 

U.S. Needs “More Mobile Spectrum,” But When?

If there is any one "rule" that has "always" been right in the mobile business, it is that, over time, more bandwidth is consumed, and therefore more spectrum is needed. That isn't to deny the other ways any existing amount of bandwidth can be employed. There have been clear advances in coding and modulation that wring more usable bandwidth out of any amount of spectrum.

Traffic offloading using Wi-Fi also has become important, and networks can be redesigned using smaller cells, to effectively increase the amount of bandwidth any given amount of spectrum can provide. But the costs of shrinking cells grows as the cell radii shrink. 

At some point, there arguably is a point where the cost of tweaking existing bandwidth is more expensive than can be supported by retail service rates, and when allocating additional spectrum is the more affordable choice, from a national perspective of supplying lots more bandwidth at reasonable cost.

Some argue that there is no spectrum shortage. That might be correct, for the moment. Where argument increases is whether there will be potential spectrum shortages in the future, and what needs to be done to avert a problem.

One should be skeptical at times about claims that a specific country is "falling behind" on some measure of communications intensity. And that applies to claims of spectrum deficiency. 

Sometimes the "adoption" issue is simply that the value of a particular innovation is not immediately grasped. Once the value proposition is clear, adoption seems to occur rapidly enough.

In recent years it has been argued that the United States is behind in broadband adoption, either  fixed or mobile.   In earlier years it has been argued that the United States is “behind” Europe, for example, in mobile phone adoption.

Similar concerns have been expressed about use of text messaging (short message service) or U..S. use of mobile multimedia or mobile payments, compared to Japan, for example.

That might apply, in some ways, to claims the United States "has quickly fallen behind the world" in  auctioning off spectrum that can be used to support wireless communications.

It is argued that Germany and Spain have auctioned about 50 percent more spectrum for broadband than the United States has. It is said that France has auctioned about 40 percent percent more spctrum, while Italy and Japan have auctioned 30 percent more spectrum.

"Specifically, the U.S. has auctioned about 410 MHz, Germany about 615 MHz, Spain about 600 MHz, France about 560 MHz, Italy roughly 510 MHz, Japan an estimated 500 MHz, and the United Kingdom preparing to auction about 600 MHz, Precursor principal Scott Cleland says.

Some skeptics will argue that one would expect Cleland to take that view, as one virtually always will find Cleland taking positions that are "against" Google and "for" telcos. And there is little doubt that mobile service providers virtually always seem to be looking for more spectrum as they add more customers.

The new reality is that each of those new customers are starting to consume network bandwidth at unprecedented rates, compared to past usage of narrowband voice and messaging apps.

That isn't to deny that more spectrum will be needed, in most countries, as mobile broadband adoption increases. Nor are U.S. regulators unmindful of the need to clear unused former TV broadcast spectrum for mobile use. So the "auction gap," like many other past "gaps," will close over time.

Also, what isn't immediately so obvious is what other spectrum assets already exist that can be "re-purposed," as U.S. mobile service providers are decommissioning older 2G or iDEN spectrum for new use by fourth generation networks.

And then there is spectrum Clearwire already has deemed surplus, the potential Dish Network, LightSquared and Nextwave spectrum, for example.

Long term, most service providers will need more physical spectrum. What isn't so clear is that there really is a spectrum auction gap that means anything terribly important at the moment.

That isn’t to deny a future issue, though. The growth rate in global mobile  data traffic is projected to grow 60 percent annually from 2011 to 2017, which will result in a 15-fold increase in traffic by 2017, mainly due to video traffic.

“Such an explosion in data traffic requires more spectrum,” the International Telecommunications Union says.  In this regard, policy-makers and regulators can help to create a supportive environment and encourage investment and ensure sufficient availability of quality spectrum, the ITU broadband report  says.

Analysts at Deloitte concur with that assessment, and in fact urge more rapid clearing of spectrum for 4G mobile network use in the U.S. market.

U.S. investment in 4G networks could fall in the range of $25 billion to $53 billion during 2012-2016; conservatively, these investments could account for $73 billion to $151 billion in GDP growth and 371,000 to 771,000 new jobs, Deloitte argues.

The lower levels of investment and economic benefits are consistent with a “baseline” or “business as usual” scenario in which U.S. 4G deployment proceeds at a moderate pace and the transition from 3G to 4G stretches into the middle of the decade, Deloitte argues.

Under these conditions, U.S. fi rms would be vulnerable to incursions by foreign competitors capitalizing on aggressive efforts in their home markets to deploy 4G networks and develop
4G-based devices and services, Deloitte also warns, however.

In fact, Deloitte analysts believe that even an additional 500 MHz of allocated 4G spectrum would be insufficient to keep up with demand. “The Federal Communications Commission and Commerce Department are working tomake 500 megahertz of spectrum available for wireless
broadband in the next 10 years, but even if that goal is achieved, it could be difficult to keep U.S. commercial wireless spectrum supply and demand in balance as interest in new 4G offerings grows, Deloitte argues.

Past experience with 3G networks suggests the value of allocating more spectrum, and easing rules about its use, Deloitte says. From 1994 to 2000, FCC auctions tripled the amount of spectrum available for commercial mobile services.

“However, spectrum caps limited U.S. carriers to 55 MHz per market, while abroad most European and Asian carriers were allowed to own 80 to 90 MHz,” Deloitte says.  In 2003, the
spectrum caps were removed, while U.S. carriers also were permitted to buy and sell spectrum.

Those moves prompted a 250 percent increase in investment and a 300 percent increase in jobs in the mobile market, Deloitte argues.

But some might argue current Federal Communications Commission plans to auction off 500 MHz or so of new spectrum for Long Term Evolution 4G networks by about 2014 will happen soon enough to avert potential spectrum problems for the market as a whole. That doesn't mean particular competitors cannot have local issues. But that's a different matter.

Mobile Already is the Preferred Way of Accessing the Internet

It isn’t hard to argue that mobile is the preferred way most people around the globe want to use voice services.

Globally, there were  87 mobile phone subscriptions for every 100 people in 2011. There were in 2011 about 18 fixed line phone subscriptions for every 100 people.

What now is happening that mobile is becoming the way a majority of people access the Internet. A large part of the reason is that smart phones increasingly represent the way people use the Internet.

For the first time in history, the installed base of smart phones will exceed that of personal computers at the end of 2012.  
.

There were 589 million fixed broadband subscriptions by the end of 2011 (most of which were located in the developed world), but nearly twice as many mobile broadband subscriptions at 1.09 billion, the International Telecommunications Union broadband report says.

Of a stock of 5.97 billion mobile cellular subscriptions worldwide by the end of 2011, some
18.3 percent related to mobile broadband subscriptions..



According to Ericsson,mobile broadband subscriptions are growing by approximately 60 percent year over -year and could reach around 5 billion in 2017.

Worldwide, the total number of smart phones is expected to exceed 3 billion by 2017 (Ericsson, 201214), with the number of smartphones sold in Africa and the Middle East expected to increase four-fold from 29.7 million units sold in 2011 to 124.6 million by 2017 (Pyramid Research).

In Latin America, smart phones could represent half of all mobile phone sales by 2016.  Smart phone adoption is also gaining momentum rapidly in the Asia-Pacific region, where smart phones are projected to account for 33.2% of all handsets sold in 2012, with China alone representing 48.2 percent  of units sold.


But other mobile or untethered devices are important as well. Ericsson estimates that the total subscriptions of data-heavy devices (smart phones, mobile PCs and tablets) will grow from around 850 million at the end of 2011 to 3.8 billion by 2017.

To be sure, smart phone users do not  tend to consume as much  bandwidth as PC users do. But the number of smart phone devices will be so large that their overall impact will be about as large as the impact of PC Internet access, in terms of network demand.

Cisco estimates that adding one smart phone to a network is equivalent to adding 35 basic phones; adding one tablet is equivalent to 121 basic phones(or three smart phones); while adding a laptop or mobile PC is equivalent to 500 basic phones.

So total offered network load will be a combination of the quantity of devices and the bandwidth demand each type of device tends to create.

This leads Ericsson to conclude that, by about 2017, data traffic will be split fairly equally between smart phones, mobile PCs and tablets. 

How Big is SMB Revenue Opportunity?

While the U.S. cable operators in 2012 may generate over $7 billion in annual revenues providing telecommunications services to businesses, they "will be chasing a declining business telecom services segment" and face fierce competition from entrenched telco providers with very deep pockets ready to staunchly defend their existing base, according to a study from The Insight Research Corporation.

Cable operators will gain some market share, but "they will remain small players in a big industry with low margins and little cash flow," Insight Research argues. 

That the small business market is fragmented is uncontestable. That the market is "declining" is more contestable. 


By way of contrast, other analysts say services aimed at small and mid-sized businesses will be among the top-three fastest-growing communications services for fixed network providers, according to Atlantic-ACM. 

Machine-to-machine services, business Internet access and business VoIP all have double digit growth rates, according to Douglas Barnett, Atlantic-ACM senior analyst.

Between 2011 and 2017, M2M will have a 28 percent compound annual growth rate, small business Internet access will have a 24 percent CAGR and Business VoIP will have an 18 percent CAGR, Atlantic-ACM says. 

Of the 146 million U.S. wireline retail local telephone service connections in service in June 2011, about 38 percent were provided by incumbent local exchange carriers, about 26 percent were  ILEC business customers, while 20 percent of lines were supplied by non-ILEC residential service providers, while 16 percent were supplied by non-ILEC business service providers. 

In addition to the cable companies, local telcos also have emerged as significant suppliers in the CLEC business, especially in business customer segments. 

Revenues for U.S. CLECs were forecast to grow at a compound annual growth rate of 26.9 percent to reach $61.1 billion by 2006, Atlantic-ACM forecast in 2001. 

In 2003, The Brattle Group estimated that U.S. CLECs held more than seven percent of the U.S. business market, and nearly 10 percent of the U.S. consumer market. 

Neither of those figures has proven incorrect. A substantial amount of market share and revenue has indeed shifted to new providers. The Federal Communications Commission reported there were more than 206 million broadband access connections in service in mid-2011. 

Assume an average revenue for each of those connections of $40 each (a blended rate assuming $35 for a mobile connection and $50 for a fixed connection, and including both higher-priced business connections and consumer connections). About 81 percent of those connections were supplied by cable or wireless providers. For the sake of argument, assume that every wireless line is functionally a competitor to an incumbent broadband line. 

So 81 percent of 206 million connections would be 166.86 million accounts. At $40 a month, each of those lines might represent $480 a year worth of revenue. That would represent about $80 billion in annual revenue. 

To be more strict, assume only cable modem and a quarter of "ILEC" broadband accounts are counted as "CLEC" revenue, for purposes of estimating CLEC broadband access revenue, eliminating all wireless lines. 

That implies 27 percent of all fixed network broadband lines were supplied by "CLECs." 

Of the 206 million broadband connections, 23 percent are supplied by cable operators and 18 percent are supplied using DSL or fiber to home technologies. That implies 37 million CLEC lines using telco platforms and 47.4 million cable high speed lines. 

Assume 100 percent of the cable modem lines properly are counted as  "CLEC" revenue, at an average of $50 a month. Assume that 20 percent of the DSL or FTTH lines are sold by CLECs at $80 a month. 

That in turn suggests cable CLEC revenue of $28.4 billion and telco platform CLEC revenues of about $35.5 billion annually, for a total of about $35.5 billion in "CLEC" broadband access revenues. 

Assume that the 36 percent of fixed voice lines represent $45 a month in revenue (a conservative estimate including both consumer and business lines). That implies $540 a year in revenue for each line in service.

The FCC says there were 146 million fixed voice lines in service in mid-2011. That would imply 52.6 million "CLEC" lines in service, or $28.4 billion in end user revenues, not including access or other carrier revenues. 

So the "CLEC" revenue stream might be as little as $64 billion a year, or as much as $108.4 billion a year.