Monday, March 7, 2016

Sharing Will Help Fix Spectrum Scarcity Problem

India faces a variety of institutional and business barriers to more-rapid deployment of Internet access services, some of them related to business models and the cost of building networks to serve customers with average spending of about US$2 a month. But regulations, taxes and zoning arguably play a significant role.


One might argue that making lots more spectrum available, free or cheap, would dramatically change the cost side of business models. That tends not to happen in India, since the economic rent from leasing spectrum is seen as so lucrative for units of government.


Spectrum sharing might be one way to reconcile both objectives: better business models for Internet service providers; faster deployment and lower prices for consumers and yet incremental revenue for governments.


The broad movement is to free up licensed but unused spectrum in an efficient way (enabling sharing), as the traditional methods (clearing users and then relicensing) take too long and cost too much.

The reality is that most licensed spectrum is not used, most of the time. Some licensed spectrum seems never to be used at all. That is quite inefficient, at a time when useful communications spectrum is required for all manner of Internet apps.

Sample Spectrum Occupancy Report: Each band averaged over six locations. National Science Foundation: M.A.McHenry Shared Spectrum Co - Click for full sized image

Saturday, March 5, 2016

U.S. Moving Ahead on 28 GHz for Mobile Uses

Most spectrum actually does not get used, most of the time. Hence the new emphasis on sharing of existing spectrum. That takes a variety of forms.

New methods for allowing mobile networks native access to Wi-Fi, as well as sharing of other spectrum bands between licensed and commercial users, provide examples of the trend.

In other cases, sharing between commercial users is the task.

Even if the recent World Radiocommunications Conference did not allocate 28 GHz for 5G mobile use, the United States Federal Communications Commission is moving ahead to allow use of that band for mobile networks.


As with some other bands of spectrum, there are sharing and interference protection issues, as satellite operators use portions of the 28 GHz band. Methods for enabling sharing of other bands, including 5 GHz (mobile and Wi-Fi) and 3.5 GHz (licensed and commercial users) also are being developed.


The broad movement is to free up licensed but unused spectrum in an efficient way (enabling sharing), as the traditional methods (clearing users and then relicensing) take too long and cost too much.

Friday, March 4, 2016

Google Fiber Vdeo Take Rates Modest, So Far

Not that it matters too much at the moment, but Google Fiber does not seem to be getting many video subscribers, according to Craig Moffett, MoffettNathanson analyst. The primary impact is to limit Google Fiber revenue per account, as high speed access only tends to generate about $70 a month, where video plus Internet tends to generate $130 a month gross revenue.


Over the long term, that difference in average revenue per account could be quite important, for obvious reasons: getting nearly double the revenue per account generally matters quite a lot for any ISP or triple play provider.


According to Moffett, Google Fiber ended 2015 with about 53,000 video subs.


At the end of 2014, Google Fiber had 12,659 video subscribers.


In Google Fiber’s most-mature market, Kansas City, Kan., video adoption is about 23 percent, a reasonable enough rate of adoption.


In Kansas City, Mo., video adoption is about 17 percent.


In Provo, Utah, a newer market, Google Fiber has achieved eight percent.


More important, arguably, is high speed access adoption, which one suspects is quite a lot higher than video adoption. In 2014, Google Fiber Internet access adoption had hit levels as high as 75 percent, for example.

Another survey showed 20 percent adoption after a year’s worth of marketing, figures that compare well with Verizon’s take rates for FiOS after a year’s worth of marketing. On that glide path, it would be reasonable to estimate that, over a few years, Google Fiber would get 30 percent to 40 percent adoption. Some might argue Google Fiber could do even better than that.  

CenturyLink to Launch Own Video Streaming Service

CenturyLink is testing it own streaming video service in four markets. As planned, the service will allow CenturyLink to expand the number of locations within its fixed network footprint where it can offer entertainment TV services.

At the moment, Prism TV, the CenturyLink linear video service, has limited reach. CenturyLink Prism TV passed 3.2 million homes in 2015.

Prism Stream will work on high speed Internet access lines supporting just 10 Mbps, instead of the 25 Mbps otherwise required for Prism TV.

CenturyLink provides high speed access to well in excess of six million homes across its footprint, and passes more than 22.2 million homes in total.

Prism Stream might mean an additional 12 percent of potential buyers will be able to use Prism Stream, where only about 7.6 percent can buy Prism TV, at least according to 2014 statistics.

CenturyLink Download Speed
Percent of People Able to Buy
Down >768k Up >200k
15.6%
Down >3M Up >768k
14.5%
Download > 768k
15.6%
Download > 1.5M
15.4%
Download > 3M
14.6%
Download > 6M
13.5%
Download > 10M
12.3%
Download > 25M
7.6%
Download > 50M
0.1%
Download > 100M
0.1%
Download > 1G
0.1%

source: Broadband Map

Will Netherlands Businesses See Higher Prices After Ziggo-Vodafone Merger?

The Netherlands communications market will see an interesting real-world test of what happens when a market consolidates, particularly in the areas of prices for business services and churn rates, as Viggo and Vodafone merger their Netherlands operations.


Coursr Research, for example, predicts prices will rise as a direct result of the combination, as competition lessens.


Duopolies, though, have not always--perhaps almost never in the Internet era, lead to that result, for voice or data communications in the business segment, where prices have fallen for decades.


To be sure, prices often fall in hyper-competitive markets as well, for a time. When there are too many competitors, though, that often becomes unsustainable, and the number of firms that can operate in the market long term will fall.


The general expectation is that prices rise, when such consolidation happens. The issue is whether that actually happens. There still is evidence that even a duopoly can produce pro-consumer results (lower prices, better services, better service). The current fixed network market in the United States provides one example.

At least in terms of Netherlands fixed network high speed access, the market is starting to look like a duopoly.
source: Telecompaper

Whether that also is true in the mobile market is an outstanding issue. In most markets, there are three to four leading providers, and regulators and policy advocates disagree about the minimum number of providers to produce effective competition.


The point is that there is not clear evidence even a fixed network duopoly necessarily produces higher prices and fewer other benefits, compared to a more-competitive market.


Also, increasingly, even the notion that a duopoly exists in the U.S. market is being challenged by Google Fiber, other independent Internet service providers and municipal operations of various types.


In a growing number of U.S. markets, there are two or even three providers of 1 Gbps Internet access. How far that can continue, and whether it is sustainable, is the issue.


The point, though, is that we cannot automatically conclude that a smaller number of providers in any market necessarily leads to higher prices for business or consumer customers. With the exception of content services, prices for data and voice have tended to fall, even in U.S. consumer “duopoly” markets.


Business services arguably have not been a duopoly for some time, as most U.S. markets have multiple providers of business data access and voice, far exceeding two providers in number, and at least a handful that are facilities based.


A combination of forces, ranging from new competitors to Moore’s Law (cheaper processing and storage), better radio technology, better air interfaces and efficient modulation techniques, optical fiber, new architectures, new spectrum and even regulation (allowing network builds in only some areas of a city) have combined to increase capacity and lower price-per-bit metrics.


Nominal price increases, where they have occurred, have to be seen in the context of customers buying services that are faster and offer more capacity, or provide unlimited domestic calling or messaging, for a flat fee. Even when customers spend more for some services, the quality and quantity of usage has changed dramatically.


Also, though one might expect quite a lot more customer churn in the wake of such a merger, where the new entity has greater ability to bundle a wider range of services (fixed and mobile), a
survey of 343 business customers by Coursr Research seems to be inconclusive on that score.


Only about one percent of respondents who do not use Vodafone or Ziggo say they will leave their current providers as a result of the merger. About five percent of respondents who use neither Ziggo nor Vodafone suggested they might switch to either Ziggo or Vodafone for data or mobile services.


With the caveat that people often do not behave as they say they will, bundling could work. Some 21 percent of Ziggo customers suggested they would add Vodafone mobile services, while 17 percent of Vodafone mobile customers suggested they might add Ziggo data communications.



In any given year, in the U.S. market, perhaps 12 percent to 18 percent of customers actually change providers, arguably fewer in the business segment, however.


On the other hand, Coursr data also suggests the amount of definite churn would initially be quite low.


Very few would leave their current provider if the combined operation enters the market -- none at all out of Vodafone's business customer base, and just 3 per cent of Ziggo's customers. The joint venture would probably also pick up customers from other players such as T-Mobile and Tele2.


After the merger of Ziggo and Vodafone, the new firm will have a much-stronger position in the customer segment of businesses with 50 to 500 customers or more employees, at least in terms of installed base.


The installed base of customers in the communications market in the Netherlands shows strong KPN presence, among organizations of every size.




In the data services market, the combined Ziggo and Vodafone will be even stronger with small business customers, but with less installed base in the large enterprise market.


The issue is whether the new combination actually will lead to higher prices for business voice and data. Some think it will do so. Classic economic thinking would suggest it could, or even should do so.


But some real-world evidence suggests even a duopoly, and that will not exist in the Netherlands market, can produce competitive market benefits for buyers.


The new combined Ziggo and Vodafone arguably will have a stronger position where it comes to fixed network business data services.




The new entity will be most competitive in the mobile segment of the business.

Thursday, March 3, 2016

Wi-Fi Offload Not Simply a "Developed Market" Trend

Most observers would now agree that Wi-Fi offload plays a significant role in supporting mobile device Internet access. The only real issue is the extent of the reliance. According to Juniper Research, phone data is delivered to phones using the mobile network 65 percent of the time.

Most assume that is true in developed markets where fixed access infrastructure is well developed, with more questions about the value or possibility of Wi-Fi offload in developing regions.

Wi-Fi offload might be important even there. Vodafone is using the technique in India, and other Indian mobile operators seem to be looking at doing so as well. Other carriers in Southeast Asia, including Vietnam and Indonesia,  also are doing so.  

Wi-Fi networks deliver about 35 percent of phone data.

Other estimates of Wi-Fi offload traffic suggest might be higher than 35 percent.


That might suggest Wi-Fi cannot be a full substitute for the mobile network, though it can be an important method for reducing the overall cost of operating a mobile access network, especially by reducing capital investment to increase capacity.

That undoubtedly will be even more true as methods for ensuring carrier-grade access on Wi-Fi networks becomes commercially available.

Mobile Internet access traffic offloaded from mobile networks to Wi-Fi represented 45 percent of total mobile Internet traffic in 2013, and will grow to 52 percent of total mobile Internet access traffic by 2018, according to Cisco’s Visual Networking Index.

Likewise, the amount of traffic offloaded from smartphones will be 51 percent by 2018, and the amount of traffic offloaded from tablets will be 69 percent by 2018.


Ting to Sell Gigabit Internet Access in Sandpoint, Idaho

Ting says it will bring gigabit Internet access service to the greater Sandpoint, Idaho market. Sandpoint is a town of about 7,400 people, located in the northern part of the state, located on the shores of Lake Pend Oreille.

Ting already serves Holly Springs, N.C.; Charlottesville, Va. and Westminster, Md. and appears willing to launch high speed Internet access service in other towns as well.

Ting sells symmetrical 1-Gbps Internet access service for about $90 a month. In Sandpoint, Ting will be competing with Northland Communications, which offers Internet access at 24 Mbps,  and CenturyLink, which sells service at up to 40 Mbps.

Will Video Content Industry Survive AI?

Virtually nobody in business ever wants to say that an industry or firm transition from an older business model to a newer model is doomed t...