Friday, November 4, 2011

Google Doesn't Want to be a Service Provider


From a return on invested capital perspective, the difference between Google’s current business model and that of a facilities-based wireline service provider like Verizon could not be starker,” say Sanford Bernstein analysts Craig Moffett and Carlos Kirjner.

“In 2011, we expect Google to post an ROIC of 56 percent, or 38 percent when including goodwill,” they say. “In 2010, Verizon’s wireline segment (which includes FiOS) sported an ROIC excluding goodwill and ‘one-time items’ of  just 1.6 percent.”

“Including goodwill and similar intangible, and smoothed one-timers, it was minus one percent,” the analysts say.

Those are good reasons why Google will not want to become a service provider, even as it considers the virtual necessity of offering entertainment video and voice services in addition to broadband access on its 1-Gbps test networks in Kansas City, Kan. and Kansas City, Mo.

Wireline networks have the weakest returns on invested capital with a 1.5 percent gain over the last decade, Moffat says Wireless networks had a meager return of 0.3 percent. Cable garnered a 2.5 percent return. Low returns from invested capital
Satellite networks had the best return on invested capital at 5.5 percent. It’s no wonder that DirecTV shares have trounced other companies in 8-year returns. Others stocks—AT&T, Comcast, Dish, Sprint and Verizon—have negative returns.



Google, reports the Wall Street Journal, is looking to add video entertainment services, and possibly voice, for customers of its 1-Gbps fiber to home network in Kansas City, Mo., and Kansas City, Kan. The moves would be logical.

Many observers have wondered how such a network, delivering only 1-Gbps Internet access service, at prices "comparable" to existing services provided by telcos and cable companies, could possibly generate enough revenue even to break even.

As it turns out, Google has no magic rabbit to pull out of its hat. The costs of its network are not dissimilar from the costs any other service provider would incur. And few service providers would contemplate building a fiber-to-home network with a single revenue stream, namely Internet access.

Of course, Google could have chosen to operate as a "wholesale only" provider of bandwidth to other service providers. It could still do so. But the few U.S. examples of access network providers who attempt to operate "wholesale only" have not proven highly viable, most would probably conclude.

The only way to approach break-even apparently is to operate the network the way all other such networks are operated, namely providing retail triple-play services to consumers.

Nobody expects Google to become a "service provider" with its own facilities, on a wider scale. But that isn't the point. To some small extent, Google might become a distributor of voice and video services, not just a broadband access provider. But once it secures distribution rights, there are other possibilities.

So far, it seems unlikely Google would get licensing rights that will immediately save consumers money. In fact, any video rights will likely include the normal clauses that require Google to pay as much as other video distributors. But if Google were to focus its services only on "over the top" delivery, it might still have a clear price advantage, compared to other service providers who must build and operate access facilities, of course.

Google might also find it only can get content rights if it agrees to bundle channels in the typical way cable, satellite and telco TV providers do, which would limit the amount of innovation Google could attempt. Also, until Google got serious volume, the prices it pays for content rights will not allow significant retail price discounts.

But any move by Google into the triple-play services market would be a bit of a shock, even if nobody thinks Google wants to become a traditional service provider. The broader issue is that if Google can get what essentially amounts to "streaming rights" to most of the standard TV channels, it would have a bit of room to challenge not only the telco, satellite and cable providers, but over time might gain some leverage to package those channels differently.

In the near term, we should anticipate little change, as the content providers will act in ways to protect the existing distribution model. Longer term, if Google should get traction, matters will change. Google Ponders Pay-TV Business

Android Device Returns Cost Operators $2 Billion Per Year

http://www.wds.co/docs/controlling-the-android.pdf
The return and repair of Android smartphone devices is costing mobile operators as much as $2 billion per year as they try to evolve their customer service strategies to keep pace with the rapidly growing ecosystem, says WDS. 


The study analyzed over 600,000 technical support calls that the WDS teams around the world have handled in the last 12 months.


The study finds that fragmentation has led to a higher than average propensity for hardware failure on Android-based devices: 14 percent of technical support calls on Android relate to hardware, versus 11 percent for Windows Phone, seven percent for iOS and six percent for BlackBerry. Android Device Returns Cost Operators $2 Billion Per Year



Judge allows Sprint suit against AT&T/T-Mobile deal

A U.S. judge has ruled that Sprint and C Spire Wireless can pursue part of their antitrust lawsuit against AT&T Inc's proposed $39 billion acquisition of T-Mobile USA. Though the judge dismissed large parts of the suit, U.S. District Judge Ellen Huvelle agreed to allow the competitors to pursue their injury claims about the effect the deal would have on the market for wireless devices.


AT&T and T-Mobile, a unit of Deutsche Telekom AG , had sought to dismiss the lawsuit, but U.S. District Judge Ellen Huvelle agreed to allow the competitors to pursue their injury claims about the effect the deal would have on the market for wireless devices.


The decision may complicate matters for AT&T and T-Mobile USA because they will now have to simultaneously fight the government's attempt to block the deal and argue against claims by the two competitors. Sprint suit against AT&T/T-Mobile deal


Though the outcome remains in doubt, some observers might argue that the fearsome, well-oiled AT&T regulatory affairs team has had an unexpectedly difficult time making the argument in favor of the acquisition. 


Some might argue that this indicates not some sudden loss of effectiveness on the part of AT&T's persuasion machine, but merely that the case is a tough one to present. 


Perhaps most difficult in that regard is the argument that there is a spectrum shortage that will be alleviated, in substantial ways, if the acquisition moves forward. Others have pointed out that it would cost less for AT&T simply to light up spectrum it already owns, for example. 


The other problem, from an antitrust perspective, is that market concentration in the U.S. mobile industry already exceeds the normal tests antitrust attorneys normally use to determine the extent of market concentration. HHI index

Google Looking at Triple Play Services

Google, reports the Wall Street Journal, is looking to add video entertainment services, and possibly voice, for customers of its 1-Gbps fiber to home network in Kansas City, Mo., and Kansas City, Kan. The moves would be logical. 


Many observers have wondered how such a network, delivering only 1-Gbps Internet access service, at prices "comparable" to existing services provided by telcos and cable companies, could possibly generate enough revenue even to break even.

As it turns out, Google has no magic rabbit to pull out of its hat. The costs of its network are not dissimilar from the costs any other service provider would incur. And few service providers would contemplate building a fiber-to-home network with a single revenue stream, namely Internet access.

Of course, Google could have chosen to operate as a "wholesale only" provider of bandwidth to other service providers. It could still do so. But the few U.S. examples of access network providers who attempt to operate "wholesale only" have not proven highly viable, most would probably conclude.

The only way to approach break-even apparently is to operate the network the way all other such networks are operated, namely providing retail triple-play services to consumers.

Nobody expects Google to become a "service provider" with its own facilities, on a wider scale. But that isn't the point. To some small extent, Google might become a distributor of voice and video services, not just a broadband access provider. But once it secures distribution rights, there are other possibilities.

So far, it seems unlikely Google would get licensing rights that will immediately save consumers money. In fact, any video rights will likely include the normal clauses that require Google to pay as much as other video distributors. But if Google were to focus its services only on "over the top" delivery, it might still have a clear price advantage, compared to other service providers who must build and operate access facilities, of course.

Google might also find it only can get content rights if it agrees to bundle channels in the typical way cable, satellite and telco TV providers do, which would limit the amount of innovation Google could attempt. Also, until Google got serious volume, the prices it pays for content rights will not allow significant retail price discounts.

But any move by Google into the triple-play services market would be a bit of a shock, even if nobody thinks Google wants to become a traditional service provider. The broader issue is that if Google can get what essentially amounts to "streaming rights" to most of the standard TV channels, it would have a bit of room to challenge not only the telco, satellite and cable providers, but over time might gain some leverage to package those channels differently.

In the near term, we should anticipate little change, as the content providers will act in ways to protect the existing distribution model. Longer term, if Google should get traction, matters will change.

Thursday, November 3, 2011

Republic Wireless Readies $20/Month Unlimited Mobile Service

Republic Wireless is a new mobile phone service from Bandwidth.com that will be launching on November 8, with some features some users will find compelling. Republic Wireless teaser


Among those features is the price, said to be $20 per month, with unlimited talk, texting and data. The plan is also touted as truly “unlimited” with no bandwidth caps. $20 a month unlimited smart phone service?


Republic is a mobile virtual network operator, but is integrating Wi-Fi connectivity and software that will switch connections between the mobile network and the Wi-Fi network automatically.

The hybrid service essentially will default to Wi-Fi whenever possible, and use the mobile network only when Wi-Fi is not available. 


That should translate into lower calling costs for end users, since Bandwidth.com will be able to use fixed network VoIP much of the time.
Public Wi-Fi usage

In all likelihood, pricing will set at levels much lower than what users typically expect from mobile service plans, and without need for a contract. Republic Wireless


Users will have to buy a new device specially created by Republic Wireless, which undoubtedly will prove a barrier to some users. But $20 a month, with unlimited data access, texting and calling, is going to be attractive to lots of people.



Smart phone owners are using Wi-Fi hot spots in increasing numbers, both in the home and out in public.  A recent study by WeFi shows an uptick in smart phone data consumption, much of it by users on the Android platform, for example. That might explain why the Republic Wireless service will use Android devices.  Smart phone Wi-Fi usage

"Zero" Effect of Government Stimulus Spending over 50 Years


Nothing is more important than understanding whether particular sets of government spending actually "solve" the problems they are supposed to fix. The reasons are hugely significant, such as whether certain government policies can "help" get an economy out of recession, and "merely" important as matters of good public policy, including the charge not to waste huge amounts of taxpayer-supplied revenue.

The problem is that there is arguably much more disagreement among economists about macro-economic policies than about micro-economic policies. And though not everyone will agree, a study of 50 years worth of government spending by economists at the Phoenix Center for Advanced Legal and Economic Public Policy Studies finds that such spending has “zero” effect on private-sector job creation.

“During periods of economic sluggishness, we find that government spending has zero effect on private-sector job creation,” the report suggests. “This result is consistent with the apparent impotence of huge federal government spending increases  aimed at reducing unemployment.”

In contrast, when it comes to job growth,  expansions in private investment are
effective in both regimes, but its efficacy is greatest during economic stagnation, the study authors say.

By implication, policies that discourage private investment may have severe job-killing effects during economic downturns, since it is during the low growth periods that private investment is most effective at creating jobs. Spending has no impact

The effects on (various measures of) economic activity of government spending and private investment have been studied by economists for decades, and while the economic stimulus effects of private investment are unquestioned, the effects of government spending on labor markets remains an open empirical question, the study says. That might come as a surprise to some who believe the impact of massive government spending during recessions are effective.

In our earlier work on the influence of regulation on job creation, titled Regulatory Expenditures, Economic Growth and Jobs: An Empirical Study, we found that even a small across the board five percent reduction in the operating budgets for all federal regulatory agencies (about $2.8 billion) would result in an increase in employment by 1.2 million private-sector jobs annually, growing private-sector GDP by about $75 billion each year, the authors say.

“Put differently, eliminating the job of a single regulator grows the American economy
by $6.2 million and creates nearly 100 private-sector jobs annually,” the study says.

Mobile Data Revenue Drivers Continue to Change

Most mobile service providers in developed markets have moved past the point where text messaging revenue is the primary “data service” augmenting declining voice revenues. That was not the case in the late 1990s or most of the first decade of the 21st century. These days it is mobile data plans for smart phones that have taken that role. Messaging drove revenue in mid-2000s
Mobile data revenue trends

In 2007, for example,  more than $28 billion was generated by mobile short message service (SMS), multimedia message service (MMS), and instant messaging (IM) services
between Western Europe and the United States alone. 

If email is also included, more
than half of all mobile data revenue in these regions was derived from messaging. It was true that mobile messaging was the foundation of most data strategies.

These days, it is smart phone data plans, augmented by data services for other connected devices, which is more strategic. Worldwide mobile connections will reach 5.6 billion in 2011, up 11 percent from five billion connections in 2010, according to Gartner. 


Mobile data services revenue will total $314.7 billion in 2011, a 22.5 percent increase from 2010 revenue of $257 billion. “Mobile data traffic will increase significantly as more people will have access to mobile data networks, there is a migration toward smart phones and an increase in sales of media tablets,” said Jessica Ekholm, principal research analyst at Gartner. 

Early in 2011, for example, Vodafone Group hit a milestone. Vodafone's latest quarter data revenue exceeded messaging revenue for the first time. 

Mobile data forecast
Vodafone's emphasis on sales of smart phones and associated data plans seems to have been the driver. Mobile data tops SMS

Right behind that strategy is expected revenue from connected devices that will include sensor applications of various types. Connected device revenue forecast

Machina Research, for example, estimates that connected devices will grow from nine billion in 2011 to 24 billion in 2020. The lion’s share of the growth will come from machine-to-machine connections, which will grow from two billion at the end of 2011 to 12 billion at the end of 2020.

That doesn’t directly translate into mobile service connections, though. The majority of those devices are expected to be connected using  Wi-Fi, which really is an untethered use of a fixed broadband connection. Machina Research expects 2.3 billion of those device connections will use the mobile network in 2020, accounting for 19 percent of all cellular connections.

That implies M2M revenue will grow to EUR714 billion ($979 billion) in 2020.
PC and laptop mobile broadband will grow dramatically, from 215 million connections at the end of 2011 to 1.5 billion in 2020. By 2020 most PC/laptop broadband connections globally will be mobile, the firm suggests.

Wireless wide-area connected tablets and e-readers will grow from 66 million in 2011 to 230 million in 2020, as well.

Growth in handset data users will also be significant, with 3G+ devices set to grow from two billion at the end of 2011 to nine billion by 2020.

Machina Research forecasts that global mobile data traffic will increase from four
exabytes in 2011 to 42 exabytes in 2020, with 60 percent coming from PC/laptop connections and 37 percent from handsets.

Machina Research expects mobile network  operator data revenue to grow from EUR130 billion ($178 billion) in 2011 to almost EUR500 billion ($685 billion) in 2020.

The basic evolution is from data revenues based on text and email services to mobile data plans to support smart phones, to be followed by connected devices. Initially, connected devices will be tablets and similar devices such as e-readers, but sensor applications will grow over the longer term.

Will Generative AI Follow Development Path of the Internet?

In many ways, the development of the internet provides a model for understanding how artificial intelligence will develop and create value. ...