Netlfix ISP Rankings Show Local Access, by Itself, Doesn't Help Much

Though we normally, and rightly, spend lots of time thinking about, demanding or complaining about local access speeds, Netflix data on how various ISPs handle Netflix streaming video continue to show that end user experience of the Internet depends much more on the architecture of the entire Internet ecosystem, than it does on any tail circuit.

The January 2013 Netflix rankings for US Internet service providers (ISPs) continue to show that local access bandwidth doesn't help much.

To be sure, user experience is contingent on lots of elements other than raw access speed at the end user location.

But the rankings also show that end user access networks have a modest impact on Netflix delivery speeds. 

Google Fiber's 1-Gbps access connection does deliver the highest performance. At about 3 Mbps on a sustained basis, Google Fiber is not that much faster, when it comes to delivering Netflix streams, than Verizon's FiOS, at about 2 Mbps, Time Warner, Cox, AT&T or Cox access services, all of which Netflix says operated at about 2 Mbps. 

Mobile networks run slower, as you would probably expect. The thing to watch is what happens as 4G Long Term Evolution networks become more common. 

The rankings from November 2012 suggest mobile streamingl  is as much as four to six times slower than a fixed network connection. 

Chromebook "Pixel" Apparently is Coming

With the caveat that some think the Chromebook Pixel is a chimera, there is growing thinking that the product is real, and is coming. Derided in some quarters as a hobby, or as a product that missed its window, Pixel might indicate Google doesn't think Chromebooks are a hobby. 

HP has joined Lenovo and Samsung in offering Chromebooks. Granted, for most people smart phones and tablets are more interesting personal devices. But content creation still has to happen. For some, Chromebooks are a better way to do what we once called "netbooks," especially when users are traveling. 

Lots of people say they only travel with a Galaxy Note II, as people once said that about their Blackberries. More people can do just fine with a tablet. Some of us have to take a notebook. Skype support and handling of .pdf documents have proven to be the two issues that I personally find are drawbacks of using Chromebooks. 

Telecom Business is Fragmenting

The separation of access from applications that is a fundamental feature of Internet Protocol will probably have an apparently contradictory impact on the structure of the global telecom business.

On one hand, slower growth will drive a new wave of consolidation among service providers, both domestic and internationally.

That will mean a smaller number of larger suppliers. Just five percent of global telcos control about 62 percent of industry revenue, according to analysts at Booz and Company. One might argue those figures show a high degree of industry concentration. But Booz say there is significantly more to come.

On the other hand, end user application markets might be fragmenting, essentially creating smaller islands that run counter to the growing trend to scale on the networks side of the business.

Increasingly, the emphasis is on local services, local content and local development
grounded in the community needs of local markets, according to the  International Telecommunications Union.

That trend to “localism” should lead to business models adapted for the specificities of geography, demography, generation, and economic and social environment in each country.

You might call that “personalization.” But personalization also means fragmentation of formerly homogenous markets. And that runs counter to the consolidation trend among access providers.

And that will complicate retail packaging, pricing and offer development across markets. In some cases, that will encourage access providers to focus “mainly on access.” Other suppliers might try to play more on the personalized applications end of the business.

So, as has been the case for a couple of decades, service providers will adopt different business models. And though it might not be a popular observation, access providers might well find they don’t have as many natural assets as they would hope, in the applications area. For any number of reasons, carriers simply cannot develop on the fast-paced,consumer-based model.

Access provider time horizons for investment and product life cycles simply are too slow, compared to the consumer application providers.

In fact, access to the network is now the basic product, rather than voice - which is increasingly fragmented and embedded as a feature or function of an application
such as gaming or messaging, rather than being a stand alone service, the ITU suggests.

The fragmentation of the industry  into closed, proprietary systems operating in silos of
communication could potentially threaten the interconnectivity upon which the global telephony system has been built, the ITU says.

So there is a growing disconnect between the diversity and localism or personalized nature of applications and the scale and need for interoperability of access operations globally.

In some key ways, the way software and applications now get created--independently of access--creates the potential for distinct revenue models. Generally speaking, the service provider “customer” generally is the actual end user of an access service.

The application customer generally is an advertiser or retailer, not the end user of an application. That means service providers generally sell to a different customer than an application provider, even if a “user” is a focus of the operating business.

And that will be a growing issue. Can access providers balance the challenges of selling products to consumer end users and other business partners (advertisers, merchants, enterprises)?

In the applications area, access providers will compete with dozens to scores of other competitors, most of whom capable of innovating faster.

In the access area, they will compete with a handful of other providers, most of whom cannot move significantly faster than each other. Which model makes most sense? Which models will regulators allow?

And which is the easier path? One might reasonably argue that the quickest path to revenue growth is to simply acquire other assets, essentially postponing a bigger strategic choice.

In fact, “the telecom sector remains relatively fragmented,” Booz consultants say. In the healthcare and oil and gas sectors, companies in the top five percent generated 79 percent of total 2009 industry revenue.

International revenue provides another indication that further consolidation is possible. In the telecom business, international revenue accounted for only 25 percent of telecom industry revenue in 2008. That is well below the 38 percent level that all industries averaged, and only half that of some industries, Booz and Company consultants say.

Regulators might make the key decisions easier, especially where they mandate a "separated" industry structure where "access" is a wholesale product, and all retailers compete on a non-facilities-based basis.

That wouldn't directly enable an access provider to consider a bigger move into "applications." But removing most "network operations" overhead from the organizational culture would make easier a shift of resources into applications.

Why U.S. Consumers Pay "More" for Internet Access

U.S. consumers tend to pay more for broadband, high speed Internet access than some people in other countries, on a nominal basis. What does that actually mean? Not much, in one sense. 

Ignore just for the moment the growing number of entrepreneurs who think they can change that state of affairs in the U.S. market. It is true that in nominal terms, U.S. high speed access prices are "higher" than in many other countries. 

On the other hand, average incomes and average costs of living also differ significantly between countries. So "nominal" costs don't tell us as much as some think. 

But even on a nominal basis, across countries globally, U.S. Internet access rates are somewhere in the middle. Critics say lack of competition, spectrum scarcity or regulatory capture by the tier one telcos accounts for the pricing. It isn't hard to find evidence of the higher costs. 

But nominal retail prices are not the issue, anymore than the typical cost of just about anything, across countries, is a terribly useful way of understanding broadband costs. 

There are other fundamental reasons why retail prices vary from country to country. Communications spending is related to income. Communications spending is related to the average level of retail prices for all other goods and services, as well. 

Communications prices therefore reflect the backrop of overall income and prices in each country.

So one way of trying to filter for those facts is to look at what high speed access, or any other commodity, costs as a percentage of typical personal income (mobile services sold to people) or fixed broadband (sold to locations). 

As a percentage of income, U.S. broadband is no more expensive than it is in other developed countries. What matters is what consumers pay as a percentage of income. 

"Google Now" is Google's "Siri"

Google Now is a natural language query app inevitably to be compared to Apple's "Siri."  It was first included in Android 4.1 ("Jelly Bean") and was first supported on the Galaxy Nexus. 

One Way Device Share Creates Revenue Share

Between them, Apple and Samsung earn nearly all the profits in the global smart phone device industry. Because other major vendors lost money in smart phones last quarter, the combined profits from Apple and Samsung in the category were greater than the total industry's Q4 profits.

Because other suppliers lost money, Apple and Samsung captured 101 percent of smart phone profits in the fourth quarter of 2012 and 103 percent for the full year, according to Canaccord Genuity analyst Michael Walkley.

Apple and Samsung's share of smartphone industry profits was greater than 100 percent every quarter of 2012. 

That sort of share also creates other revenue opportunities, though. Application providers, such as Google, often will pay a device supplier for placement as the default app on all shipped devices.

That might represent $700 million to $900 million in current annual revenue for Apple. 

Morgan Stanle

Has "Peak SMS" Been Reached?

Spain is in many ways an exemplar of what is happening to text messaging revenues in the European and some other markets.

After peaking at the end of 2008 at about €450/quarter, Spanish text messaging revenues have fallen by six percent to about €171 million in the third quarter of 2012.

As some would note, text messaging represents nearly 100 percent operating profit for mobile operators, so losing volume in such a high-margin revenue stream is a particular problem.

The reason for the declines is substitution by users of IP-based messaging for text messaging.

According to asymco, 97 percent of Spanish smart phone users have Whatsapp installed, allowing users to send free instant messages to other users.

That is cannibalizing text messaging. Use of SMS was down about 30 percent in the third quarter of 2012, for example.


Globally SMS traffic is still rising, but Informa Telecoms & Media forecasts that mobile operators will still generate a total of US$722.7 billion in revenues from SMS between 2011 and 2016.

But text messaging share of global mobile messaging traffic will fall, from this point forward, analysts suggest.


By some estimates, we are close to the point where over the top message volume should exceed that of text messaging.

In the U.S. market, for example, text messaging revenues and volume fell for the first time ever in the third quarter of 2012.

The only surprising fact is that text messaging revenue has fallen, even in the U.S. market, which has been relatively more protected from such losses, to this point.

What Long Run Mobile Broadband Growth Rate?

In 2012, global bandwidth growth slowed to about 40 percent, from about 70 percent annual growth in 2008. But keep that in perspective. Growth rates always slow when any organism, business or trend reaches an adult stage, garners a much larger installed base or achieves high penetration. In other words, there is a “law of large numbers” at work.

And that seems the case for Internet bandwidth growth as well. While the pace of growth is slowing, international Internet bandwidth continues to grow rapidly, more than doubling between 2010 and 2012, to 77 Tbps, according to TeleGeography.

Average international Internet traffic grew 35 percent in 2012, down from 39 percent in 2011, and peak traffic grew 33 percent, well below the 57 percent increase recorded in 2011, TeleGeography says.

International Internet traffic and capacity growth rates are declining due to slowing broadband subscriber growth in mature markets, and the expansion of content delivery networks (CDNs) and local caching technologies, which reduce the need for new long-haul capacity by storing popular content closer to the end-users.

Some think the same sort of trend ultimately will characterize mobile broadband bandwidth growth rates as well, In fact, there is little reason to doubt that future trend, given historical precedents.

In March 2011, for example, AT&T projected that data bandwidth growth would be on the order of eight to 10 times over then-current levels between the end of 2010 and the end of 2015.

That forecast appears to be based on an expectation that volumes would roughly double in 2011 and then increase by a further 65 percent in 2012.

Instead, AT&T seems to be seeing something like 40 percent annual growth. To be sure, 40 percent annual growth is significant. It means bandwidth consumption doubles about every two to three years.

Cisco estimates mobile broadband grew about 70 percent in 2012, and will grow at a compound annual growth rate of 66 percent from 2012 to 2017.

Some believe Wi-Fi offload will slow the rate of mobile broadband growth. On the other hand, even such offloading, at high rates of perhaps 80 percent, would slow the rate of growth by about 50 percent.

Utilities Look to "Field Area Networks"

Add “field area network” (FAN) to the list of acronyms we use to describe communications networks of various functions and coverage areas, ranging from wide area network to metropolitan area network to local area network.

By 2020, annual global shipments of wireless communications nodes to support FANs will reach 14.3 million units, according to Pike Research.

The market for private utility FANs--which generally will be private networks--will be led by North America, which represented about 82 percent of world shipments in 2012.  

That share will decline steadily to 2020, but North America will still account for 44 percent of world shipments by 2020.  

The fastest growth in the decade will come in Latin America, where shipments will increase at a compound annual growth rate (CAGR) of 48 percent.

FANs will use a variety of wireless networks, ranging from radio frequency mesh and Wi-Fi to WiMAX or LTE technology.

A FAN is a network used to connect various devices located in a utility’s “field” of operations, which can include smart meters, concentrators, distribution assets, control and protection equipment, and substation equipment, according to Pike Research.

Some field area networks will use the 802.15.4 standard for fixed terrestrial radio networks. Others might use Long Term Evolution, Some also predict systems using the 802.15.4 standard, with point-to-multipoint fixed wireless, will dominate the networks used to support FANs.

But LTE networks operated by leading mobile companies could be a growing factor in the FAN market.

ABB, GE, S&C and Eaton are some of the established firms selling systems for the field communications market. 

Google Nexus 1 to Control Satellite, XBox Kincet Next, Seriously

A standard issue Nexus One smart phone will be controlling a new satellite to be launched by the end of February 2012. Granted, it is a "nano satellite" just  30 cm long and weighing 4.3 kg.

Still, using a consumer smart phone as a satellite controller is novel. 

A Strand-2 satellite is under development. For that generation of satellites, two cubesats will use the motion-sensing technology in Microsoft's XBox Kinect devices to locate each other in space and dock together.

Huawei, ZTE, Lenovo Among Top 5 Smart Phone Suppliers in 4Q 2012

Huawei, ZTE and Lenovo ranked among the top-five suppliers of smart phones in the fourth quarter of 2012. You might expect Samsung and Apple to top the list, and they do.

In the fourth quarter of 2012, Samsung provided 63 percent of smart phone shipments while Apple had 48 percent. Huawei shipped about 11.5 percent of all smart phones globally in the fourth quarter of 2012, while ZTE shipped 10 percent. Lenovo shipped about 9.5 percent of smart phones in the quarter.

For some of us, the surprise continues to be that BlackBerry and Nokia do not appear among the top five. We sometimes become myopic and assume that Nokia and BlackBerry are fighting Samsung and Apple. In one sense, that is true. Nokia and BlackBerry will have to fight for a spot among the “high end” providers in the market.

In a larger sense, it appears the situation is that Nokia and BlackBerry have to catch Huawei, ZTE and Lenovo, all of which seem to be gobbling up the low end that Nokia once dominated.

Mobile Broadband Grows 18% in OECD Region

Mobile broadband subscriptions have reached nearly 700 million in OECD countries, the Organization for Economic Cooperation and Development reports.

Mobile broadband has grown at about an 18 percent rate from June 2011 to June 2012, largely driven by continuing strong demand for tablets and smart phones.

The average broadband penetration in the OECD area is 56.6 subscriptions per 100 inhabitants, OECD says.

Korea (104.2) and Sweden (101.8) are the only two countries with more mobile broadband subscriptions than inhabitants.
‌‌‌
Fixed wired broadband subscriptions reached 321 million in the OECD area in June 2012, for an average penetration of 26 subscriptions per 100 inhabitants, a 1.7 percent increase over the previous six months. Keep in mind that penetration per home depends on your assumptions about housing density in each country.

As a rough assumption, you might assume average households of two people each, so fixed broadband penetration might be 52 percent of homes.

Digital subscriber line  subscriptions are being replaced by fiber connections, though slowly.  The share of fiber subscriptions in fixed broadband has increased to 14.2 percent, while DSL represented 54.7 percent of the total fixed wired broadband subscription.

Since DSL and fiber represent 69 percent of connections, That could mean roughly 30 percent of connections.

Will LTE Be Complementary to Fixed Broadband, or a Substitute?

Will Long Term Evolution remain a complementary form of broadband access, or will it become a functional substitute? And, if so, for which sets of consumers will this be a logical choice?

AT&T and Verizon Wireless certainly believe there are some significant percentage of customers for whom broadband access supplied by mobile networks, though in a “fixed” basis, will be a viable substitute for fixed network broadband access.

Up to this point, the amount of such mobile substitution for broadband access has been fairly limited, and nothing like the substitution of mobile voice for fixed voice. But fourth generation networks, offering high speeds, which some liken to “digital subscriber line” speeds, should provide greater potential for substitution.

Still, some do not believe mobile broadband is a substitute for fixed broadband. Some of us doubt that. There might be many ways to infer meaning from the fact that fixed network broadband adoption has not changed much since 2009.

Some might say that is because later users value broadband access less, or because tough economic conditions are forcing consumers to make choices, or simply that people who use the Internet mostly already buy broadband access. It is hard to disagree with the logic.

But some of us would argue that there is a growing trend for some users to substitute mobile broadband for fixed broadband because most of what those users do can be done on a smart phone.

Some might argue, with reason, that such preferences remain a single digits kind of development, and that probably is quite true. On the other hand, a couple of important market drivers are operating, and should grow in importance.

As tablets have shown, most people, most of the time, consume content, instead of creating it. This is what is contributing to the “post-PC” era we seem to be entering.

But that very fact means a greater number of users might conclude that they can get by with smart phone broadband, and really do not need a fixed broadband connection. That might be more true for single person households and households of younger and unrelated persons.

On the other hand, the other angle is that a fixed connection might increasingly have the most value as a way of offloading traffic from smart phones and tablets, not so much to enable use of PCs. That might account for the finding that 83 percent of smart phone users also have access to fixed broadband at home.

In other words, despite growing “smart phone only” access, the vast majority of smart phone users also have fixed access services. But the market can change. If fixed broadband tariffs start to rise, and if LTE 4G tariffs start to fall, many more users could opt for a different mix of spending than they have shown in the past.

Up to this point, 4G prices have been higher than 3G . But that could already be changing.

Comparing retail prices between the second quarter of 2012 and the fourth quarter of 2012, service providers in 73 percent of countries have reduced the “effective cost” of their 4G tariffs  to a significant degree, according to ABI Research.

LTE Prices Dropping, Globally

Up to this point, 4G prices have been higher than 3G . But that could already be changing.

Comparing retail prices between the second quarter of 2012 and the fourth quarter of 2012, service providers in 73 percent of countries have reduced the “effective cost” of their 4G tariffs  to a significant degree, according to ABI Research.

The effective cost in terms of “dollars per gigabyte” has dropped by 30 percent, overall.  In United States, service providers kept fees the same but offered larger data quotas.

In Australia, Sweden, Japan, Singapore and Saudi Arabia the operators lowered the monthly fee but have data quotas unchanged.

​India currently offers the lowest priced plan, ABI Research says. India’s lowest priced mobile data plans decreased 29.4 percent year-over-year in the fourth quarter of 2012.

And some mobile service providers are pricing 4G at a discount to 3G. “In Norway, Telenor has introduced 4G tariffs that are cheaper than 3G,” ABI Research says.

Mobile broadband services using Long Term Evolution will experience a substantial 60-percent price drop  in retail prices between 201 1and 2016, according to Tariff Consultancy, on the way to gaining 250 million users worldwide by the end of 2016.


Average monthly user data allowances for LTE mobile broadband services in 2011 were about 22 GBytes per month, in some cases ranging as high as 80 GB per month in the case of Tele2 Sweden.

The average mobile broadband price globally for a top of the range LTE mobile broadband service was 50 Euros per month, for two-year postpaid contract service.

The study showed that average LTE broadband prices ranged from 0.5 Euro (Tele2 Sweden) up to 9.9 Euro (Omnitel Lithuania) per GByte of data mobile usage.

But retail prices started dropping at least by 2011. . Telstra (Australia) “BigPond” offered an 8 GByte monthly data user allowance for the equivalent of 30 Euro per month,

BigPond had launched offering 4 Gbytes for 38 Euro. In Singapore M1 (Mobile One) was offering customers a 40 percent discount off the monthly list price.

The price trends might not be so obvious in all markets. U.S. service providers, so far, have generally tried to price at a premium for LTE services.

Vodafone Earnings Woes Spur Talk of Sale of Verizon Wireless Assets

Vodafone’s latest quarterly financial report illustrates the reasons why some analysts, and executives at Verizon, might be weighing some action to change the current ownership status of Verizon Wireless, majority owned by Verizon, but with a big minority intBut erest held by Vodafone.

To be sure, Verizon says no such talks are underway. Well, to be specific, Verizon says no talks about a full purchase of the 45 percent Vodafone stake in Verizon are underway. That would still leave some room for less complicate measures, such as a gradual purchase by Verizon of Vodafone shares.

In fact, most observers have noted for a decade that Verizon would prefer to own all of Verizon Wireless. But talk about a Verizon purchase of Vodafone has heated (again) recently. Earnings weakness at Vodafone could be a factor.

Vodafone posted a worse than expected drop in group revenue for the last three months of 2012. The biggest declines came from Europe operations.

Vodafone service revenue dropped by 2.6 percent to £10.37 billion in the fourth quarter of 2012.

Southern Europe services were hit the hardest, dropping nearly 12 percent to £2.3 billion.

In Italy service revenue tumbled 13.8 percent, partly caused by lower mobile termination rates.

Spain revenue was down 11.3 percent. Significantly, that hit was caused at least in part by the ending of handset subsidies.

Africa, Middle East and Asia Pacific service revenue grew 2.7 percent to £3.14 billion.

Verizon Wireless service revenue grew 8.7 percent. Verizon Wireless also paid Vodafone £2.4 billion in dividends,

So the issue for Vodafone is whether the value of the dividend stream is more important than the monetization of Verizon Wireless equity value.