Why IoT is Directly Related to Rural Internet Access

It might not be self evident that there is a direct relationship between Internet of Things revenues and profits earned by mobile and other service providers, and rural Internet access services. Simply, one day IoT revenues will be necessary to support rural Internet and other communications services.

The reason is simple enough. At some point, after voice, texting and mobile data revenue growth has ended, some big new sources, likely including IoT, will be needed to generate financial surplus that can be tapped to provide rural Internet access.

How to subsidize rural communication networks is a problem for service providers, regulators and policymakers in most countries, except for some small city-states that essentially have no rural areas. .

In developed or developing nations, mobile or communications network rural coverage is an issue. In the United Kingdom, for example, two percent of road miles have zero 2G network coverage, while 12 percent of road miles have only partial 2G coverage, the RAC Foundation says.

On six percent of road mileage, there is no 3G coverage, while 45 percent of road miles have only partial 3G coverage.

Some 56 percent of road miles have no 4G coverage, while 27 percent of road miles have only partial 4G coverage.

	Miles (%) of road in Britain with…
	Full network coverage	Partial network coverage	No network coverage
2G	211,753 (86%)	28,975 (12%)	4,561 (2%)
3G	119,057 (48%)	111,679 (45%)	14,554 (6%)
4G	43,070 (18%)	65,950 (27%)	136,271 (56%)

source: RAC Foundation

As a general rule, networks serving such areas are not sustainable on their own, meaning that potential revenues never will justify building facilities. Instead, such investments must be subsidized either from surplus generated elsewhere on the network, or by direct government subsidies.

In the past, rural service has been subsidized both by profits earned from customers in urban areas, by business users and by government support programs. That will continue to be necessary.

But profits from sales of voice services or text messaging services are becoming more difficult. Mobile data services are the current driver of surpluses in most developed nations, and will become the key drivers in most developing nations, eventually.

But what happens when mobile data revenues have reached saturation? What comes next?

Internet of Things is the big answer many believe will develop. In other words, revenues and profits earned from IoT will drive supplier business models, generating enough surplus to keep extending service to rural areas.

That is one reason why IoT is a key focus of the Spectrum Futures conference this year. Here’s a  fact sheet and Spectrum Futures schedule.

At some point, after mobile data access has reached saturation, IoT services and revenues will be required to sustain the mobile business model in general, and pay for money-losing services in rural areas.

That is why work on new platforms, more affordable and therefore potentially able to close the business model gap between revenues and cost, are so important.

Put simply, costs must drop much more, consumers must perceive the value of Internet services and be able to pay for such services, or alternative methods of subsidizing end users must be created.

Some of us would argue all of those developments are required. That is why developments such as Telecom Infra Project, Free Basics, Project Loon, unmanned aerial vehicles, unlicensed and shared spectrum, use of millimeter wave radio and creation of new apps and services that drive value are so important.

Without serious innovation in the access networks business, it might literally not be possible for many service providers to continue operating.

Spectrum Fees, High Incremental Capex, Lower Value in Ecosystem Mean Historic Changes Might be Necessary

Mobile operators in India pay fees for use of spectrum. For spectrum acquired in the upcoming 2016 auctions, the spectrum usage charge will be three percent of of service provider adjusted gross revenue (AGR) on spectrum acquired in forthcoming auction in 700, 800, 900, 1800, 2100, 2300 and 2500 MHz band.

Bharti Airtel, India’s largest telecom operator will pay 3.8 percent. Reliance Jio Infocomm will pay about 3.05 percent.

Those charges might be seen as a small part of a much larger problem: high capital investment and regulatory risk that makes the business model unsustainable.

And at least some now question whether there is truly enough potential revenue in at least some emerging markets to sustain robust investment in traditional mobile network bandwidth. An open question: whether a dramatic reduction in access network costs is feasible.

By the same token, one might question whether there is enough demand to support gigabit Internet access in many countries without key subsidies from government.

Nearly 40 percent of South Korea’s broadband network build between 2005-2014 was provided by the government, for example.

Few countries are able to support--or even consider--that level of support to create high-performing Internet access facilities for their citizens.

It gets worse.

To a large extent, one might argue, access providers are capturing a smaller share of overall ecosystem value. Bluntly, one might say, access providers really are becoming commodity dumb pipe providers while value and revenue shift to app providers, device suppliers and others.

“Even after accounting for Wi-Fi and new technologies and alternate business models, there will be still significant global wireless data demand that is not economically possible to serve,” says James Sullivan, J.P. Morgan head of Asia equity research. .

Declining value capture is paired with a significant, and ongoing, increase in capital intensity. In at least some cases, that will mean possible nationalizing of networks. In other cases, competitors might be forced to consider sharing network facilities, to stave off such intervention.

That would overturn nearly a half century of moves to privatize assets and introduce competition.

“Emerging market telcos have no choice but to fundamentally change the structure of industry assets through the unification of networks via nationalization, centralization under a regulated return utility, or more aggressive commercial network sharing,” argues James Sullivan, J.P. Morgan Chase Head of Asia Equity Research.

Simply, between now and 2024, telco capital investment and operating expense will climb, while revenue growth lags. But there also is regulatory risk. India, the Philippines, Thailand, Malaysia, Indonesia and Turkey are countries where capex pressures are the big problem.

Regulatory risks exist in Indonesia, Brazil, South Africa and Malaysia, he argues.

Markets with the potential for the most extreme margin compression include the Philippines, India and South Africa; while more defensive margin markets are Nigeria and China, says Sullivan.

Bringing stakeholders together to understand changing supply and demand issues, and the business model for Internet access, is a key focus of the Spectrum Futures conference. Here’s a  fact sheet and Spectrum Futures schedule. Sullivan will be sharing his perspective at the event.

Comcast Isn't the Biggest "Cable" Company Anymore

The largest U.S. “cable” company is not cable company, anymore. Counting firms the old way, Comcast is the biggest “cable TV” company.

1. Comcast: 22,400,000
2. Charter Communications: 18,421,145
3. Cox Communications: 4,540,280
4. Altice: 3,948,000
5. Mediacom: 862,000

But it long has been the case that there are several ways to deliver linear entertainment video: cable TV networks, satellite and now telco networks. Using a “linear video entertainment” definition, rather than counting by access network technology, the market shares are different.

1. AT&T U-Verse and DirecTV: 26,000,000
2. Comcast: 22,400,000
3. Charter Communications: 18,421,145
4. Dish Network: 13,909,000
5. Verizon FiOS: 4,700,000

Cloud Services Leaders Growing at 53% to 162% Annual Rates

Amazon Web Services (AWS), Microsoft, IBM and Google combined control well over half of the worldwide cloud infrastructure service market, and are growing briskly. Google’s cloud infrastructure grew at a 162 percent annual rate, while Microsoft’s cloud services business grew at a 100-percent rate, from the second quarter of 2015 to the second quarter of 2016, according to Synergy Research Group.

Amazon, the market leader, grew at a 53 percent rate, while IBM’s business grew at a 57 percent rate.

In aggregate the big four grew their cloud infrastructure service revenues 68 percent in the second quarter of 2016, while the next 20 largest cloud providers grew by 41 percent and all other smaller providers grew by 27 percent.

The market as a whole grew by 51 percent. Amazon’s business is almost three times the size of its nearest competitor and has a clear lead in all major regions and most segments of the market.

source: Synergy Research Group

For Verizon, Being #3 to Google and Facebook Might Not be a Bad Outcome

Few--if any--observers seem to think Verizon will gain market share against Google and Facebook with its combined AOL-Yahoo business.

“Verizon Communications Inc.'s planned purchase of Yahoo! Inc. for $4.83 billion in cash would make the telecom giant the clear number-three player in the U.S. digital ad market,” says Seth Shafer, SNL Kagan research analyst. “However, gaining ground on Google Inc. and Facebook Inc. could prove difficult.”

Some might argue that being third, and even lagging Google and Facebook by quite some measure is not such a bad outcome for Verizon.

Challenging either Google or Facebook in a serious way might seem fanciful to most observers. But, at a minimum, the new Verizon unit is likely a $4 billion or bigger annual revenues business, which is sort of a minimum for Verizon to bother with.

Also, a market positioning of “we offer an alternative to Google and Facebook” would seem to be sustainable.

source: SNL Kagan

Automation, Machine Learning are Issues in Developing and Developed Nations

Automation, including use of artificial intelligence and machine learning, might boost productivity, but also cause job loss, in developing as well as developed countries. That is not a new problem.

The advent of the industrial revolution caused similar stresses two hundred years ago. But as a growing number of us might say, high productivity gains that lead to job stresses have to be coupled with high efforts to ameliorate those stresses.

An increase in inequality seems to be a byproduct of technology revolution, and therefore creates a challenging new problem to solve.

source: World Bank

source: World Bank

India Mobile Subscriptions Drop in May 2016

One does not often see mobile subscriptions decline in any developing country market, in any particular quarter or month, but that seems to be what happened in India in May 2016, when total mobile subscriptions dropped about one percent.

In urban areas, mobile subscriptions dropped two percent. In rural areas, subscriptions grew.

But most of the dip comes from share losses by several mobile providers, not necessarily an  across the board decline in net subscribers for most mobile service providers.

Fixed voice subscriptions also dropped, but that is not entirely unexpected in many markets.

source: TRAI

Consumer Revenues Propel Service Provider Revenue Growth

Among the big changes in the telecom business over the last three to four decades, one of the most prominent is the growing revenue contribution (and profit contribution, in many cases) made by consumer accounts, as opposed to business customers.

That was not always the case. In the past, super-high profits from business customers funded operation of networks also serving consumers.

These days, it is consumer revenues that lead growth for most service providers.

The other big shift is the relative contributions made by mobile, as opposed to fixed, revenues.

As I mentioned during a recent keynote address for Telegration business partners (a U.S.-based sales organization focusing on enterprise and mid-market customers), when conducting analyses of Internet adoption on a global basis, one can essentially ignore all fixed network access, look only at mobile Internet access, and still get the trend right, and the magnitudes of usage about right.

That is quite a change from historical patterns, where business user revenues accounted for about a half of total revenues in developed countries. But the direction of change is mostly towards greater consumer revenues (What EY calls “smart operator”).

A few service providers might opt for becoming “mostly” wholesale providers for retail partners, in which case the revenue contributors shift dramatically to “business” revenue (wholesale). So far, we see little evidence that service providers are willing to retrench as wholesale capacity suppliers, however.

And that means the percentage of revenue earned from consumers is going to dominate, in the future.

source: EY

Except for about 10.5 percent of Internet users in Asia, for example, who do get access using a fixed network, substantially all the rest of the Internet users do so using mobile networks.

In some Asian countries--such as the Philippines and Thailand--perhaps 20 percent to 25 percent of people use a fixed Internet connection (primarily urban residents). But in many other countries, usage of fixed Internet access can be in single digits, or less than one percent.

As we start to connect the half of people who do not presently use the Internet, the percentage of mobile or wireless users will climb, while the percentage of fixed network users shrinks. And we are talking about roughly two billion new Internet users, in Asia alone.

That is not to say other access methods will not emerge, but it is hard to ignore the fact that nearly 90 percent of Internet access in Asia now is provided by mobile networks.

Rural coverage, language relevance, device prices and recurring access costs all are issues. Still, mobile has to be reckoned the primary delivery vehicle.

As the 80/20 rule suggests, "20 percent of activities produce 80 percent of the results." For Internet access, the practical application is that only mobile really matters, where it comes to consumer Internet access.

Will that change in the future? It is possible. If Internet of Things develops as a key revenue driver, then enterprise or business revenue contributors will grow, again.

source: ITU

Bringing stakeholders together to understand changing supply and demand issues, and the business model for Internet access, is a key focus of the Spectrum Futures conference. Here’s a  fact sheet and Spectrum Futures schedule.

90% of Asia Internet access is Provided by Mobiles

Across Asia, about 58 percent of people still do not use the Internet, according to the International Telecommunications Union.

If you exclude China, Japan and Korea, plus the city-states of Singapore and island of Taiwan, the percentage of Internet non-users can, in some cases, range upwards of 76 percent.

Except for about 10.5 percent of Internet users in Asia, who do get access using a fixed network, substantially all the rest of the Internet users do so using mobile networks. That is not to say other access methods will not emerge, but it is hard to ignore the fact that nearly 90 percent of Internet access in Asia now is provided by mobile networks.

Rural coverage, language relevance, device prices and recurring access costs all are issues. Still, mobile has to be reckoned the primary delivery vehicle.

As I mentioned during a keynote address for Telegration business partners (a U.S.-based sales organization focusing on enterprise and mid-market customers), when conducting analyses of Internet adoption, one can essentially ignore all fixed network access, look only at mobile Internet access, and still get the trend right, and the magnitudes of usage about right.

As the 80/20 rule suggests, "20 percent of activities produce 80 percent of the results." For Internet access, the practical application is that only mobile really matters, where it comes to consumer Internet access.

Bringing stakeholders together to do something about that is the mission of the Spectrum Futures conference. Here’s a  fact sheet and Spectrum Futures schedule.

source: ITU

Why Bundling?

U.K. service provider bundled subscriptions will have grown by 20 percent from 2015 to 2020, with quadruple play revenues growing 300 percent, while total multiplay market revenues grow by 34 percent, Strategy Analytics predicts.

Dual-play subscriptions will peak in 2016 and begin to decline as customers move to triple and quad-play bundles, the firm estimates. By 2020, quadruple-play subscriptions will represent more than 21 percent of bundled subscriptions in the U.K. market.  

The bundled services trend--in any market--tells you quite a lot about the state of the telecommunications market. There are some advantages in terms of customer acquisition and retention.

Simply, most bundles are a form of discounting: “buy more, save money.” Once consumers have made those decisions, churn tends to drop, because dropping any one service means losing the cost savings of the whole bundle.

But bundling also tells you how hard it is (perhaps how “nearly impossible” it is) to build a modern fixed network and build a business model on any single service.

Competition, more than anything, is the cause. Assume a single-purpose network (voice, and entertainment video being the historical examples). Under monopoly conditions, the addressable market is nearly 100 percent of locations. The actual take rates were above 80 percent for video and above 90 percent for voice.

But add competition and the business model is unsustainable. Even if only two competitors split a market, neither is likely to gain more than half the available market. That means actual subscribers are gotten from about 40 percent to 45 percent of locations.

In other words, subscription gross revenue--all other things being equal--is cut in half. Add more competitors and the numbers get worse. And things never are equal. New competition also leads to price cuts. So gross revenue falls more than half.

Bundling allows any single provider to create a new revenue model based on smaller shares of bigger markets. Even if no single provider has more than half the market share in any product segment, any successful provider can maintain gross revenue by selling multiple products to the same customer.

A simple illustration: a supplier that formerly had 100 percent of the market for one product might instead have 33 percent of the market for each of three products. In such a case, the total number of sold units remains the same as under monopoly conditions.

Gross revenue per unit,and profit margins, are issues, but you see the logic.

The most important observation might be that bundling now is necessary because no fixed network can survive on revenues from a single product.

More change is coming, since virtually all existing legacy products in the bundle face product substitution or outright abandonment.

AT&T, Comcast Boost Usage Caps for Internet Access Customers

As with virtually everything else connected with Internet access, speeds keep getting faster, and usage (where there are any limits at all) allocations keep getting bigger, for U.S. customers of leading fixed network Internet service providers.

AT&T has used a variety of usage caps for customers of its U-verse Internet access service, with caps based on user access speed. Users at 768 Kbps to 6 Mbps had a 300 GB cap.

U-verse users on connections of 12 Mbps to 75 Mbps had 600 GB caps. Customers on anything between 100 Mbps and 1 Gbps had a terabyte (1,000 Gb) monthly usage cap.

But AT&T has bumped usage caps usage caps higher. But AT&T now appears to have bumped usage allowances up to 1,000 Gb for all customers on access connections of 768 kbps or faster.

As a practical matter, that means all U-verse customers, and customers of AT&T’s gigabit service have no caps at all.

source: AT&T

Charter Communications, as a condition of the approval of its purchase of Time Warner Cable, will not be allowed to impose any usage caps for seven years.

In markets where it is testing usage caps for its Internet access customers, Comcast has raised the monthly usage from 300 Gb to 1,000 Gb (a terabyte).

Comcast says its typical customer uses about 60 gigabytes of data in a month.

To be sure, some advocate no caps or any kind, ever, arguing that there actually are not cost implications for unlimited use. ISPs would disagree, of course. All networks are dimensioned for some amount of expected usage, and further designed to support peak loads.

Capital has to be invested when peak loads grow substantially. So the amount of usage by typical and the heaviest users does matter.

The terabyte cap allows for viewing of 700 hours of high-definition video, or about 23 hours at is each day. No single person watches that much video, and even if shared between four people, that amounts to nearly six hours per day, per person.

Power users (less than one percent of Comcast’s high speed access customer base) who want more than a terabyte can sign up for an unlimited plan for an additional $50 a month, or they have the option to purchase additional buckets of 50 gigabytes of data for $10 each.

As a practical matter, nearly all of you who actually have kept track of data usage under both “unlimited” and data plans with large buckets of usage, would likely agree that, as a practical, a reasonably-sized usage bucket is virtually indistinguishable from an “unlimited” plan, as a practical matter.

Usage limits vary on U.S. mobile services as well. Sprint and T-Mobile US have been bigger proponents of unlimited usage than have Verizon and AT&T.

But AT&T recently has been offering unlimited mobile data usage for mobile customers who bundle their service with AT&T’s DirecTV service.