Will AT&T Use Mexico Wholesale Mobile Network?

Mexican regulators have approved AT&T’s Iusacell acquisition, which now sets up an interesting issue: how does AT&T expand geographic coverage nationwide, especially for the new Long Term Evolution network AT&T believes represents a major opportunity? 

To be sure, Iusacell now reaches about 70 percent of Mexico’s potential mobile consumers and has perhaps eight percent to nine percent market share. But those networks use CDMA or GSM 3G air interfaces. The LTE network remains to be built. Nor is the spectrum to support the new network readily available.

“Expanding and enhancing Iusacell’s mobile network to cover millions of additional consumers and businesses is our top priority,” according to Randall Stephenson, AT&T chairman and CEO.

Consistent with AT&T thinking about the market opportunity Long Term Evolution adoption represents. Smartphone penetration in Mexico is about half that of the United States.

“AT&T sees a significant opportunity to increase smartphone adoption and mobile Internet usage in Mexico,” AT&T said. How AT&T will do so, and when, is not yet clear.

In the immediate future, it is more likely AT&T will create new cross-border calling services, as that will leverage existing customers--fixed and mobile--on both sides of the border.

China Telecom, for example, is creating a consortium to win a contract to build a national wholesale mobile network in Mexico. Others bidders are expected to emerge as well.

Whether AT&T would use such a network is one question. Whether it would use a network built by China Telecom that also uses Huawei gear is another question.

But the new network, which must be built, according to the Mexican constitution, might be valuable for mobile virtual network operators as well as mobile operators with at least some owned facilities, as the network would create a seamless national infrastructure, presumably also offering Long Term Evolution services, as soon as 2018.

Presumably, the wholesale network would allow contestants to use spectrum without specifically acquiring spectrum of their own. But sourcing fron the wholesale network also means each contestant would have the same features, coverage and quality as all others on the wholesale network.

That might leave retail price as the key variable, within some clear limits, unless contestants were able to bundle with other products and services.

So a reasonable person might argue AT&T will not want to rely on the wholesale network, burt rather build its own facilities.

Ting, the U.S. MVNO, Launches 2nd ISP Operation

Ting, the mobile service provider owned by Tucows, now has made its second investment in an Internet Service Provider business. In an earlier move, Tucows had invested in a small Charlottesville, Va. ISP, with new plans to provide gigabit access.

In the latest move, Ting will be the retail operator of a fiber to home network owned by the municipality in Westminster, Md. It isn’t yet clear what bandwidths will be offered.

Tucows, an Internet domain name registrar, in 2012 launched Ting, a mobile virtual network operator.

It doesn’t take much insight to note that lots of smaller towns and smaller ISPs might find the prospect of building or operating gigabit access networks a reasonable business proposition.

Decades ago, we used to call this sort of thing "overbuilding." That was a scenario where a third provider built a network and competed directly with a local cable TV operator and telco for video and other services. It always was a tough business, and few ever occurred.

These days, the focus is high speed Internet access, with or without video entertainment. That's the Google Fiber model. What Ting is trying is based strictly on a pure-play ISP business model.

What Will Drive 5G Apps? History Suggests We Don't Really Know

Nobody can say for sure what future fifth generation network (5G) mobile standards will look like. More significantly, nobody yet can predict how “value” will occur. Some emphasize new applications, as always is claimed for the next generation of networks. Others say it is universal access, across multiple networks, that will be a defining feature.

“Speed” often is positioned almost as an afterthought. But it would be reasonable to assume that speed will be among the most-obvious advantages for consumers, and that 5G-specific applications will develop later.

Ofcom, the U.K. communications regulator, believes 5G mobile will offer extremely fast data speeds of perhaps 10 Gbps to 50 Gbps, compared with today’s average 4G download speed of 15 Mbps (not considering LTE-Advanced or other developments in channel bonding that will boost speeds into the hundreds of megabits per second).

In South Korea, 300 Mbps already is commercially feasible over LTE networks.

For some, the new use of very high frequency spectrum above 6 GHz--not traditionally used for end user communications--could support financial trading, entertainment, gaming or holographic projections, Ofcom suggests.

Much more prevalent apps, at least initially, will be simple Internet access in high-density, high-traffic areas. Traditional “capacity,” in other words, is likely to be the immediate value.

The University of Surrey’s 5G Innovation Centre says the visioin for 5G is user experience that gives the user “the impression of infinite capacity.”

The International Telecommunication Union suggests peak 5G data rates will be in the range from 10 Gbps to  50 Gbps and latency of one millisecond.

At the moment, it is difficult to predict what the salient value of 5G--beyond capacity, speed and latency--will be. But that has been true for every succeeding generation of digital mobile networks.

Supporters often tout “new apps.” For service providers and consumers, the value almost always is found elsewhere--speed--initially. Eventually, new apps do emerge. You might argue text messaging was the singular legacy for 2G. You might suggest mobile email and Internet access was the legacy of 3G. Video entertainment might be developing as the singular new app that defines 4G.

Many believe apps related to the Internet of Things might characterize 5G. That sounds logical enough. But forecasters were wrong about new apps for 3G, which stubbornly refused to emerge. Most gave up on predicting new apps for 4G, arguing “there is no killer app.”

It is safe to argue 5G will support IoT apps. History might not confirm those predictions, at least in terms of magnitude. We rarely get it right, where it comes to next generation networks.

Content Rights Remain the Key Obstacle for Streaming

Big new business platforms typically require that a new ecosystem be built. That is the case for streaming TV as a successor to linear TV.

Content suppliers must agree to license distribution of the content in that form. Internet access must be widespread and of reasonable quality. Scores of millions of devices or adaptors typically have to be put into place.

PC and mobile device screens are widely populated; TV-capable screens less so. That means robust markets both for media adapters such as Chromecast, Fire TV and Roku, as well as higher sales of smart TVs.

On that score, about 26 percent of U.S. households own a streaming media device and 34 percent own a smart TV, according to Parks Associates.

That might help hexplain why linear video viewing is dipping and streaming is growing. About 40 percent of U.S. households subscribe to a paid digital video subscription service, with Netflix being the leader (32 percent), followed by Amazon Prime Instant Video (19 percent) and Hulu Plus (9 percent), according to the The Diffusion Group.

Across all of these services, Millennials have significantly higher subscription penetration, with nearly half buying Netflix.

Among Netflix subscribers, the preferred method of watching (44 percent) is through Internet-connected TV devices such as Apple TV and Google Chromecast.

Computers (27 percent) and gaming consoles or Blu-Ray players (21 percent) also are extensively used by the Netflix subscriber base.

The average household also spends over $6 per month on subscription Internet video services such as Netflix or Amazon Prime.

The streaming ecosystem is well on its way to completion. The last remaining big hurdle is the licensing of rights to distributors on a nearly ubiquitous basis. The other big questions are the terms and conditions of such licensing.

Business models contestants can try will be shaped by contracts that limit ways the content can be sold to customers. Up to this point, standard linear video contracts stipulate that channels must be placed on the retail tiers that are most often purchased.

Streaming sometimes is allowed when the user already has purchased a given channel as part of a linear package.

Streaming licenses can vary. In some cases, whole TV series and movies can be licensed, and viewed, a la carte. In other cases the access is possible as part of a streaming package created by a distributor such as Netflix or Amazon Prime.

Netflix introduced the concept of binge viewing, where all episodes of a given series could be watched, immediately.

Among the bigger questions for a future streaming business is whether bundling will reemerge in new forms. Some would argue it is cumbersome for consumers to navigate a content landscape where every network is a separate destination. Channels and networks aggregate content for a reason: it is user friendly.

Likewise, distributors aggregate multiple channels and sources for the same reason. In a future streaming business, one issue is the dominant or preferred packaging model. And many would bet that some form of bundling will reemerge. 

The linear video business will be transformed. But the role of distributors is likely to remain. 

UK Regulator: BT Cannot Set Wholesale or Retail High Speed Access Rates "Too Low"

Facilities-based competition might be the best alternative, where it is possible, but competition based on wholesale access often is the only available choice to regulators. But there are some obvious limits to the degree of innovation possible when the range of features is fixed, by definition. Since all contestants use the same platform, the services and features also will be the same.

The other problem is the range of pricing options. In such wholesale-based markets, the typical marketing tack is “same service, lower price.” By definition, price is the main form of potential differentiation (though suppliers are free to try and create value in other ways).

Oddly enough, wholesale prices can be set too low! In other words, the magnitude of difference between wholesale cost and BT retail pricing, U.K. regulators maintain, must be large enough to allow a competitor to buy wholesale service and then at least match BT’s retail price, while still allowing the wholesale customer a reasonable profit margin.

The new pricing rule means that BT must maintain a sufficient margin between its wholesale and retail high speed access prices that other providers can match BT prices and make a profit.

Essentially, the rule means BT will not be able to set retail prices for high speed access,or wholesale prices, that prevent competitors from making a profit when matching BT’s own retail prices for the same services.

How much that will help is always a debatable matter. In the U.S. market, a brief period of experimentation with wholesale-based competition showed that many competitors were possible when the spread between wholesale and retail price offered by the competitors was in the 40-percent range.

That competition collapsed when a rule change reduced the spread between wholesale cost and competitor retail cost to about 20 percent.

That will require complex monitoring of estimated BT costs, as a way of maintaining wholesale discount rates. The other complicating factor is that Virgin already has more share than BT in the 25-Mbps and faster portion of the market.

In December 2013 Virgin had 57 percent market share of all connections operating at 25 Mbps or faster. Overall, BT had the largest Internet access market share at 32 percent, followed by Sky with 23 percent and Virgin having 20 percent share.

OneWeb Plans 648-Satellite Fleet to Provide Internet Access to Underserved

WorldVu Satellites Ltd. has raised funding from Virgin Group and Qualcomm for a proposed global satellite internet company focusing on potential users in developing countries that cannot be reached by fixed or mobile networks, as well as to supply Internet access to flying aircraft.

The proposed network will cost between $1.5 billion and $2 billion, backers believe.

WorldVu Satellites founder Greg Wyler said. Wyler, formerly of satellite firm O3b, has been touting this idea for some time.

Virgin Group and Qualcomm are investors in the WorldVu “OneWeb Ltd.” service, which hopes to launch a constellation of 648 satellites. Investor Richard Branson thinks the total could eventually be higher than that. Branson also says voice service will be part of the core service.

Virgin is a sprawling conglomerate that now wants to build a new global satellite fleet to deliver Internet services to users who today do not have access by any other means.

Some might note that others have tried in the past, without much success, to create such networks, and few have tried to reach billions of people who today cannot afford to connect to the Internet.

But Virgin believes it has other assets that will help achieve OneWeb attain the goal of service at far lower costs.

Virgin Galactic’s “LauncherOne” program will be used to make frequent satellite launches at lower cost. Other launch partners. might be added, the press release announcing the venture hints.

Branson suggested that Virgin Galactic will be launching most of the OneWeb spacecraft, but not all of them.

“We have the biggest order ever for putting satellites into space,” Branson said. “By the time our second constellation is developed, the company will have launched more satellites than there currently are in the sky.”

The first launches are supposed to happen in early 2017. OneWeb controls a block of radio spectrum that it will use for the Internet service, but has to begin deploying the network to retain use of the frequencies, a typical requirement for spectrum grantees.

OneWeb’s satellites will weigh about 285 pounds and operate in a low-earth orbit about 750 miles above the planet’s surface. That has significant positive implications for potential bandwidth and latency performance, allowing much-lower latency than possible for geosynchronous satellites.

The deployment challenges will be significant for such a large fleet, but backers hope lower satellite and launch costs will help the venture provide consumer Internet access at far-lower prices than possible in the past.

What is the Killer App for Connected Car?

What is the killer app for connected car services? That’s another way of asking where the obvious value lies, for end users, car owners or fleet managers. So far, the answer is not clear.

Nor does it help much to assert that there is “no killer app.” That is a fuzzy answer essentially tantamount to saying “we don’t know,” or that we haven’t thought it through.

ABI Research forecasts the penetration of “embedded telematics” in new vehicles will increase from 13.4 percent in 2014 to 52 percent in 2020.

Nonetheless, a GSMA-commissioned report suggests that by 2018 such connected car markets will triple from 2012 levels, and generate €24.5 billion from in-vehicle services such as traffic information, call center support and web-based entertainment.

Some €6.9 billion will be generated by the sale of hardware such as telematics control units. An additional €4.5 billion will be earned by suppliers of telematics services such as customer relationship management.

The access connections portion of the business might represent €4.1 billion. But it always is possible such forecasts will prove too optimistic.

“While penetration levels of embedded connectivity in vehicles continue to grow steadily, it remains challenging for car OEMs to convince users to pay for built-in connected car services,” says Dominique Bonte, ABI Research director VP.

That is just another way of saying we haven’t discovered a connected car killer app, yet, and that consumers deem the value-price relationship yet to be strong enough to encourage high take rates.

So “full penetration is unlikely to be achieved solely through consumer-led drivers,” says Bonte. That likewise is another way of saying supplier push is going to be the key factor for wider adoption.

The “perceived value of safety services or in-car Wi-Fi hotspots simply are not big enough,” says Bonte. So something else will have to happen to spur adoption.

ABI Research suggests car sharing could be one such driver, enabling the remote finding and unlocking of “shared” vehicles.

Also, as the percentage of electric-only vehicles grows, remote monitoring of vehicle battery status could emerge as another value driver.

Or, ABI Research suggests, autonomous vehicles could provide the value.

In the meantime, growth will be fueled by other “non-consumer” apps such as eCall and stolen vehicle mandates in Europe, Russia, and Brazil, or manufacturer diagnostics, prognostics and preventive maintenance.