"Keep Copper Network and TDM," Some Argue

Creating good public policy in the communications realm is never easy, as regulators constantly are balancing. But it does seem as though “balancing” has become more precarious recently. Consider the whole matter of what to do about the legacy copper network, the “transition to Internet Protocol networks” and support for legacy services.

To be sure, the Federal Communications Commission has multiple goals. For example, the FCC says its mission includes “promoting competition, innovation and investment in broadband services and facilities, and “supporting the nation's economy by ensuring an appropriate competitive framework for the unfolding of the communications revolution.”

To some extent, of course, the goals of promoting competition and promoting investment are contradictory. As both U.S. and European regulators have discovered, promoting competition by expanding wholesale access to incumbent facilities often succeeds quite well.

But that same success also discourages further investment in facilities, since the owner of the scarce access network automatically enables robust competition when it invests in next generation facilities.

Likewise, the transition to the next generation of broadband networks would seem to require creating and maintaining incentives for facilities investment, balanced with the goal of promoting competition.

Sometimes that translates into rules that specifically require maintaining legacy facilities and services, even if that conflicts with the goal of supporting next generation network investment.

Ironically, we now see support both for expanded optical fiber access to support gigabit networks, and talk of preserving the life of copper networks to preserve competition.

The problem, of course, is that the copper network and legacy services are serving fewer and fewer customers, meaning the costs of serving each remaining customer are growing. at the same time the legacy services are being replaced by next generation services.

It isn’t easy, all might agree. But neither are matters helped when waffling occurs, as tough as firm policies might be. Granted, there are constituencies for keeping the copper network, and legacy services, alive.

But few seem to think it would be better if Internet service providers slowed their migration to fiber access networks, slowed the rate of speed increases or put obstacles in the way of faster Ethernet and IP services.

It is a balancing act, to be sure. But some methods of protecting competition actually are harmful to the goals of expedited investment in next generation infrastructure.

Making matters worse are asymmetrical regulatory frameworks that do not treat all providers of access and other services the same way.

The biggest U.S. high speed service providers (and Internet access is the strategic service)  are AT&T, Verizon, Comcast and Time Warner Cable. But they play under different rules.

In fact, in terms of subscriber share, Comcast is the biggest, followed by AT&T, then Time Warner Cable, then Verizon, according to Leichtman Research Group.  

High Speed Internet Access Subscribers	Subscribers 2Q 2014	Net Adds 2Q 2014
Cable Companies
Comcast	21,271,000	203,000
Time Warner	11,965,000	86,000
Charter	4,850,000	62,000
Cablevision	2,779,000	(9,000)
Suddenlink	1,103,300	200
Mediacom	987,000	3,000
WOW (WideOpenWest)	769,600	12,900
Cable ONE	482,725	(1,443)
Other Major Private Cable Companies	6,475,000	25,000
Total Top Cable	50,682,625	381,657
Telephone Companies
AT&T	16,448,000	(55,000)
Verizon	9,077,000	46,000
CenturyLink	6,055,000	(2,000)
Frontier	1,900,500	27,500
Windstream	1,153,800	(16,600)
FairPoint	333,421	1,883
Cincinnati Bell	270,300	300
Total Top Telephone Companies	35,238,021	2,083
Total Broadband	85,920,646	383,740

source:  Leichtman Research Group

True, Comcast and Time Warner Cable are not yet in the mobile business. But that will come, meaning all the largest telcos and cable companies will compete across the full range of anchor products and customer segments.

And one might also argue that asymmetrical financial returns--that underpin investment--now flow to app providers and device providers in the ecosystem, not to access providers.

Granted, it is not the business of the FCC to oversee the financial health of the device and app industries that all agree contribute to the nation’s economy. On the other hand, if there were obvious shifts in business model that directly affected the health of all contestants in the access business, one would think that would inform decision making.

In Europe, communications regulators have discovered that decades of successful promotion of competition have also lead to decades of lessened investment, and that the “pro-competition” policies are directly related to those outcomes.

It’s a balance; a tough balance. But both investment and competition must be supported.

Is Wi-Fi a ¨Power Shift¨ in Telecommunications?

There has been a long-simmering debate about whether public Wi-Fi hotspots can become a competitor to mobile networks. As with many such debates, actual practice suggests the choice is false.

These days, mobile operators consider Wi-Fi access an integral part of the overall access resources picture. And even mobile operators that might like to base all their access operations on Wi-Fi acknowledge that a combination works best.

EE has argued LTE users use Wi-Fi less. But those users do not stop using Wi-Fi. Still, the questions might grow as the footprint offered by public Wi-Fi hotspots grows.

Maravedis Rethink estimates there will be 47.7 million public Wi-Fi hotspots deployed worldwide by the end of 2014.

By 2018, the number of global hotspots will grow to over 340 million, Maravedis Rethink estimates.

France, the United States and China will have the biggest public Wi-Fi footprints. Already, China has five times more commercial Wi-Fi hotspots than any other country.

And many of those locations are networked so that roaming is possible. Some 22.7 million Wi-Fi hotspots are enabled for roaming between different provider networks in 2014, and will grow to 289.3 million in 2018.

Some questions might remain about how prevalent ¨homespot” networks will become. Homespots leverage private end user Wi-Fi locations to create public Wi-Fi hotspots.

There are perhaps 40 million such homespots in operation in 2014, growing to over 325 million in 2018. The big question is potential demand for homespot connections in suburban and less-dense areas.

Executives at iPass argue (perhaps understandably) that Wi-Fi is changing the power structures of the telecommunications industry.

Perhaps 50 percent of all commercial hotspots are controlled by brands whose core business isn't telecommunications, argues Evan Kaplan, iPass CEO. ¨We are witnessing a power shift from traditional telcos to business owners, such as cafes, hoteliers and retailers who are all getting in on the Wi-Fi game.¨

There is plenty of room for debate about the future role of Wi-Fi as an access medium, and in most cases, the debate will center on how much each network--mobile and Wi-Fi) is used, where, when, by whom, for what devices and application scenarios, and what the business models might be.

But it might already be clear that pricing of mobile Internet access plays a role.

In France, for example, despite high levels of smartphone ownership (60 percent, according to a recent survey by Deloitte, up from about 53 percent in 2013 ), only 11 percent of respondents had a 4G phone.

Many respondents apparently were put off by the high costs of both 4G handsets and data plans, although there are indications this is a misperception, as perhaps 81 percent also report they pay no additional fees for 4G access.

On the other hand, Wi-Fi hotspot availability might have something to do with lagging interest, as France has the most public hotspots of any nation, according to iPass. "Most of the devices we use are Wi-Fi only and even on the most advanced 4G handsets, 78 percent of data goes over Wi-Fi,” said Kaplan.

LTE service pricing tends to reflect a desire by most operators to price Long Term Evolution as a premium service, a tack that generally has not been taken by U.S. mobile operators, which might explain high adoption rates in the U.S. market.

To be sure, as mobile revenue shifts from voice to Internet access, it is understandable that service providers would look to price at a premium.

Mobile data has emerged as the single most important driver of telecom revenue growth, according to Pyramid Research, which forecasts that mobile data revenue will reach $633 billion globally in 2018, increasing from 40 percent of overall mobile revenue in 2013 to 52 percent in 2018.

Asia-Pacific, the world's most populous region, which accounted for nearly 38 percent of the world's mobile data revenue in 2013, should lead the growth.

Pyramid Research expects the global 4G subscription base to grow at a 52-percent compound annual growth rate, from 211 million users in 2013 to 1,750 million users in 2018.

But that growth will still be conditioned by device costs and retail service pricing.

And then there is Wi-Fi. It remains unclear how much reliance users will continue to place on Wi-Fi access even when they have Long Term Evolution access.

WiFi is the connectivity of choice among LTE subscribers. According to a study by Mobidia Technology, a provider of mobile analytics, Wi-Fi accounted for an enormous 75 percent to 90 percent of all mobile data consumed in “leading LTE markets,”according to a study by Mobidia.

Is there a danger that Wi-Fi cannibalizes some mobile data revenue? Probably. But is Wi-Fi also necessary to keep customers happy by allowing them to use lots of data without stressing trhe mobile network? That also likely is necessary.

So pricing will play a huge role as consumers make decisions about which networks to use, and whether LTE makes good sense. Price LTE too high, and adoption will fall; price it too low and the mobile networks could crash.

Mandatory U.K. Roaming Would Eliminate “Not Spots” and Differentiation

“Not spots” are a major mobile customer irritant, therefore a major problem for most mobile service providers, as well as a problem for capital budgets, as “not spots,” places where signal coverage is weak to non-existent, typically exist because the cost of installing new infrastructure does not allow for recovery of costs.

Usually, mobile operators try to remedy those sorts of problems by striking roaming agreements with other networks that do have coverage in the areas of weak to non-existent signal. But that doesn’t always seem to work.

The U.K. government has asked U.K. carriers to come up with a voluntary and universal plan, and apparently they have failed to do so.

So mandatory network access now might be on the agenda in the United Kingdom.

The idea is that a mobile operator whose own network fails, in a specific area, would have mandatory access rights on other networks serving that area that can provide service. In essence, that creates a mandatory and enforceable roaming agreement.

It might not be hard to figure out why the leading mobile service providers have rejected the idea so far. Competition on the basis of network coverage often can be a major marketing weapon.

Mandatory roaming eliminates a source of differentiation in the market.

If coverage, normally a key component of “quality” (“we have the best network, with the best coverage”) goes away, what are the key parameters on which operators can differentiate?

What mostly is left is the variable they dislike the most: price (more precisely, value related to price).

Under a mandatory roaming plan, incentives for building new infrastructure would change.

Where there is universally poor coverage, such as in a rural area, there is an incentive for at least one mobile operator to build, and then claim the “best coverage” in that area.

If mandatory roaming is required, then the firm that made the investment incurs cost with no advantage: the rivals will be able to use the network as well. That makes the investment unwise.

Up to this point, suppliers have done so on a limited basis. Vodafone and O2, for example, allow roaming across their networks, while EE and Three likewise allow roaming across their networks.

Under some circumstances, the existence of “not spots” might call out for some sort of voluntary sharing of infrastructure, which mobile operators have agreed to in other instances globally (sharing towers and sometimes radios).

Under the arrangements already in place, however, each consortium can claim advantage in some areas, if not all. All that would go away in the event of mandatory sharing across all networks.

Differentation might be eliminated in other ways, as well. Some studies show Vodafone has the best voice coverage in some areas, while EE has the best 4G Long Term Evolution coverage in some areas. All of that would be leveled, with no advantage for any provider, in the event of mandatory roaming.

The issue illustrates how hard it can be to both protect the public interest by ensuring universally good service, and yet also provide incentives for mobile operators to invest in providing that service. 

Mexico's LTE Wholesale Network: Opportunity and Risk

Mexico is creating a big Long Term Evolution wholesale network using the entire 90 MHz spectrum in the digital dividend (700 MHz band), and hopes to have the network activated by 2018, and soon will begin taking financing bids.

The new network might cost $10 billion, and require construction of 8,000 to 15,000 cell sites.

The wholesale Long Term Evolution network is viewed as a way to bring the benefits of more competition to the Mexican mobile market.

Existing mobile service providers have not been entirely sure they want to operate under such a structure.

Others say the incumbents may boycott the network. There is good reason to believe Telcel, the largest mobile provider, which has its own network, will simply continue to use its own facilities.

Telefónica likely also might believe it has enough scale to justify its own network. Telcel (America Móvil) has 69 percent market share, but will divest assets to get its share down below 50 percent.

Movistar (Telefónica) has 19 percent market share. Assuming the divested Telcel assets go to a third party, not to Telefónica, but that the new buyer acquires the cell tower networks in its serving areas, it is conceivable that providers of about 12 percent of Mexican mobile service are the primary candidates to buy service from the wholesale network.

That might not be sufficient volume to justify building and operating the new network. In a more-optimistic scenario, Telefónica would eventually switch some of its leased access to the new network, and the owner of the divested Telcel assets might do so as well.

That could create a potential opportunity representing 30 percent or more of the Mexican market, eventually.

Incumbent service providers will be able to buy capacity on the wholesale network, with one key trade-off. If they do so, such incumbents also must open up their existing networks to third party wholesale as well, on conditions similar to wholesale access terms on the new 700-MHz wholesale network.

That is another reason either Telcel or Telefónica might not want to source capacity from the new wholesale network.

Others think there are additional risks. The business model is a concern, given that the Mexican government has promised lower prices and mobile communications “as a human right.”

All that means the government will be under pressure to keep prices on its network low. So regulated prices that are too low could endanger the wholesale network’s viability, or create a need for continuing subsidies.

Wholesale prices too low might mean the wholesale network is not profitable. But if prices are too high, potential customers will conclude there is not a viable business case for their retail operations, and they will not buy.

The other hard to assess issue is whether the existence of a state-subsidized network would discourage private investment because other networks can simply buy from the state network, rather than building their own facilities.

Worse, some competitors might simply decide not to compete in the market.

The wholesale-only network will sell capacity to retailers, according to Ernesto Flores-Roux, Associate Researcher, Centro de Investigación y Docencia Económicas - CIDE, Mexico.

That is similar to the situation in Rwanda, where 4G spectrum was donated--not auctioned--to a an entity charged with building a national LTE network. In Rwanda, KT Corp. was selected by the Rwandan government to build a national LTE network, known as olleh Rwanda Networks, (oRn).

The new infrastructure company oRn will operate exclusively in a wholesale capacity, providing services to retail service providers. And it appears that as many as nine other African nations are considering doing something similar.

Nigeria also appears to have taken the wholesale-only LTE approach.

Kenya likely also is interested in a wholesale LTE network approach.

The decision to build a wholesale-only network was driven by the belief that this is the best way to assure lowest-possible cost for consumers, said Flores-Roux.

It remains unclear how investment in the wholesale network will be made. At the moment, “any conceivable structure can be used for the ownership and financing of the network,” said Flores-Roux.

For Next Generation Emergency Calling, "Punishment" Must Fit the Crime

If you have been in the telecom business for a long a time, you might have encountered, in discussions about emergency calling issues, quips that go something like “we spend 70 percent of our time dealing with an issue that is a cost of doing business, not the business.”

That’s an exaggeration for most people whose responsibilities do not directly concern emergency calling operations.

But the gist remains: an essential part of operating a communications business entails dealing with lots of other issues that are a necessary part of the business, but actually cost money and take lots of time.

And 911 emergency calling is that sort of issue. What is different now is that emergency calling now occurs in a context where fewer people actually use the fixed network for calling, where most people rely on mobiles for voice, and where revenue actually is driven by all sorts of services other than “calling.”

Without minimizing the importance of issues related to emergency calling, all effort there is expended on a shrinking, soon to be “very small” revenue source.

Also, at least for the moment, that effort is expended on multiple networks, one of which--the copper access network--is supporting fewer and fewer customers all the time.

Voice--and emergency calling--remains a key function for a communications network, if increasingly not its revenue underpinning.

And that raises a larger, tough to solve issue: how much capital and time should be spent supporting even vital functions on multiple networks that will not be sustainable for much longer?

Yes, the essential emergency calling feature has to be supported on new IP and optical networks. But that also occurs in a context where most calling happens on the mobile networks, and revenue is not driven by voice on any of the future networks.

The point is that “proportionality” is important. Proportionality is a concept dealing with fairness and justice under law. As applied to criminal law, proportionality is the idea that the punishment of an offender should fit the crime.

Under international law, proportionality is the concept that the legal use of force in an armed conflict should be bounded.

In European Union law, for example, proportionality requires that there must be a legitimate aim for a measure.

The measure also must be suitable to achieve the aim. The measure must be necessary to achieve the aim, that there cannot be any less onerous way of doing it.

The measure must be reasonable.

Dealing with next-generation emergency calling must be “proportional,” one might argue. It is a legitimate objective, but the measures to achieve the objective must be suitable, not onerous, reasonable and effective.

AT&T, Verizon Boost Data Allowances for Some Plans

AT&T Mobility has boosted data allowances on $40 a month and $70 a month Value Share Value plans.

The new plans boost the data allowance from 2 GB to 3GB for the $40 a month plan. The $70 a month plan used to feature 4 GB, and now offers 6GB more data.

Verizon Wireless, for its part, has boosted allowances for customers (new or existing) as well. Customers on $80 a month plans get 10 GB data allowances, up from 6 GB. Customers on $100 a month plans receive 15 GB data allowances, up from 10 GB.

Those moves reflect the continuing mobile marketing wars in the U.S. market, and represent a non-price form of competition that helps both firms maintain prices while providing more value, a rather typical form of competitive packaging in the Internet access business.

Over time, as has been the case for PCs and other computing devices, retail prices tend to decline over time, even as processor speed, memory or other attributes tend to improve.

That is an analogous process to what is happening in the U.S. mobile market, as Sprint and T-Mobile US attack retail packaging, both in terms of retail price and data allowances. AT&T and Verizon now are forced to react.

For some of us, the more surprising development is not that Moore's Law continues to operate, but that in the realm of Internet access, despite nearly continual criticism, Internet access speeds in the U.S. market have continued to improve nearly as fast as Moore’s Law suggests processing will improve.

That is somewhat shocking, as Internet access is a civil engineering exercise, while Moore’s Law operates at the device level. And it is far easier to replace devices than access infrastructure.

But Internet access bandwidth in fact does advance nearly as fast as Moore’s Law would suggest for devices.

Consumer Internet access bandwidth has grown about as fast as Moore’s Law would suggest, according to Jakob Nielsen, Professor Rod Tucker and Phil Edholm, former Nortel's CTO.

That is shocking, but historically accurate. That is why some of us are sanguine about prospects for U.S. Internet access speed advances, providing blockages (regulatory intervention, massive economic disruption, re-monopolization) do not develop.

source: Bureau of Labor Statistics

Consider a 2004 prediction (remember that in 2000 most U.S. Internet users were on dial-up connections): “Edholm's Law says that in about five years (that would have been 2009) 3G (third-generation) wireless will routinely deliver 1 Mbps, Wi-Fi will bring nomadic access to 10 Mbps, and office desktops will connect at a standard of 1 gigabit per second.”

History has shown that prediction to be about right for mobile, possibly too conservative for Wi-Fi, while too optimistic about desktop connections.