No Need for 5G?

At least some observers might have argued that “we do not need 3G.” Some might have argued there “is no need for 4G.” So it is not surprising that some argue there is no need for 5G, either.

One argument might be that data growth actually is not as robust as many believe it is. Video is the reason most-often given for why capacity is needed. But some might point out that video already represents more than 50 percent of total mobile network. Some might argue that future growth, though substantial, is a glass half filled. In other words, there will be significant growth, but not orders of magnitude growth.

Many of the new use cases likewise will be satisfied by existing fixed network assets.

Others might argue that, although 5G will enable many internet of things apps, 5G will not be needed for many such apps and use cases.

The point, some might argue, is that the commercial revenue drivers for 5G are not entirely clear. So argues William Webb, Ofcom senior technologist, for example.

“Users will not value the higher data rates that are promised and will not need the higher capacity forecast,” Webb  argues.

The countervailing argument is that demand will exist, or can be created. That essentially rests on the view that services and revenues based on 4G have reached--or will reach--saturation soon.

Since 5G offers new capabilities, it also can support new use cases and revenue models, either speeds much faster than 4G or with a different set of features (supporting low battery consumption use cases).

Others might argue that 5G will enable new business models precisely because it eliminates the price and performance advantage fixed networks traditionally have held over mobile networks, making mobile a full network substitute for the first time, and essentially allowing mobile to cannibalize fixed network revenues.

Networks that can deliver the highest speeds and greatest reliability will command the highest average revenue per connection, Juniper Research argues. That means 5G actually will be strategic for mobile operators.

That noted, demand will have to be created. In terms of commercial IOT revenues, Juniper Research believes average revenue per connection from IoT will be “disappointing, including smart cities and digital health.”

For starters, such devices will not require much bandwidth, implying relatively low revenue per connection.

That might not be so much the issue as the relative value of the transmitted data, in context. Almost by definition, the value of data collected from any single sensor hinges on its predictive role for the application it supports. A single sensor that can predict the failure of an expensive piece of machinery might have very high value.

A single sensor measuring pressure, temperature, vibration or performance might therefore be highly valuable. But many sensors, though valuable, will not have large immediate value.

Customers are not likely going to be willing to pay a high percentage of total potential gain or loss to be alerted to changes that suggest such potential gain or loss. In other words, customers might well be willing to pay a few percent of the value of any product’s value, but not 100 percent percent of that value, especially when a simple “replace it when it fails” approach is possible.

How much recurring cost will a consumer be willing to spend to know a light bulb is going to fail, compared to the alternative of simply replacing that bulb when it fails?

In other words, the market value of a recurring payment to learn about future device or component failures varies. Such value might be very high for many industrial and commercial purposes, but generally very low for most consumer applications.

That is one reason some believe 5G fixed wireless will be so significant. A single consumer 5G fixed wireless connection, in the U.S. market, for example, should generate monthly revenue between $50 to $100, depending on bandwidth, about the range of similar fixed network connections.

That is a big deal, when many internet of things connections might generate only a few dollars a month of revenue.

Also, consumer demand for internet access--including higher-bandwidth versions--is well known, and large. That means it has potential to directly and powerfully affect total service provider revenues. Many other new IoT apps will generate comparatively small revenues, at first, and so not move the revenue needle appreciably.

Fixed wireless could have a significant impact.

That impact might not be seen in all markets, or even many markets, at first, Juniper Research, for example, argues that most 5G connections, in 2025, will be active in just three countries:

China, the United States and Japan.

Together those countries will have 55 percent of all 5G connections by 2025, Juniper Research estimates. The United States will account for over 30 percent of global 5G IoT connections by 2025, with the highest number of 5G connections for fixed wireless broadband and automotive services.

Where Cable ISPs Compete, They Win?

The U.S. internet access market is one of the few in the world where cable TV operators have been driving consumer internet access speed advances over the last decade. The United Kingdom is another such market.

In 2016, for example, on U.S. fixed networks, “average” speed increased 40 percent in a single year, and most of that was driven by Comcast and other cable TV operators.

In the United Kingdom, in 2016, Virgin Media, the U.K. cable operator, was far and away the fastest ISP, offering speeds more than twice as fast as BT or BT’s wholesale customers and about four times faster than ISPs using unbundled local loop access.

source: Ookla

source: Ookla

But speed is not the only significant business model impact in the U.S. and U.K. markets. At least as important--and arguably more important-- is the dramatic change in potential market shares obtainable by any former telco in such a market.

Facing accomplished competitors with scale, skill and other business resources, including their own facilities, a former incumbent telco might reasonably expect that its addressable market shrinks as much as half. In the U.K., Virgin Media’s network reaches perhaps a quarter of locations; in the United States cable operators have close to 100 percent coverage of telcos.

Verizon’s experience with its FiOS service suggests a telco facing a cable opeator might get 40 percent to 45 percent of the internet access market, even when fiber to the home is the access platform.

"At the end of the second quarter of 2017, cable had a 64 percent market share versus 36 percent for telcos,” said Bruce Leichtman, Leichtman Research Group president and principal analyst.

Unless something breaks the current trend, telos could collectively become something of an afterthought in the access business, with market share as low as 28 percent by 2020, according to New Street Research.

source: New Street Research

Stranding 60 percent of the deployed capital in FTTH access networks is one very good reason for some service providers to look at 5G fixed wireless. If the maximum share is range bound around 40 percent to 50 percent, then any solution that minimized stranded assets will improve the business case.

Absent a Disruption, U.S. Telcos Will See Internet Access Share Between 28% and 45%

How well can any telco do, in terms of internet access market share, when facing accomplished competitors with scale, skill and other business resources, including their own facilities?

Verizon’s experience with its FiOS service suggests the answer is “40 percent to 45 percent of the market,” even when fiber to the home is the access platform.  

"At the end of the second quarter of 2017, cable had a 64 percent market share versus 36 percent for telcos,” said Bruce Leichtman, Leichtman Research Group president and principal analyst.

Unless something breaks the current trend, telos could collectively become something of an afterthought in the access business, with market share as low as 28 percent by 2020, according to New Street Research.

source: New Street Research

Stranding 60 percent of the deployed capital in FTTH access networks is one very good reason for some service providers to look at 5G fixed wireless. If the maximum share is range bound around 40 percent to 50 percent, then any solution that minimized stranded assets will improve the business case.

Also, if one believes internet access is the anchor service for consumer customers, continued share loss is dangerous. In the second quarter of 2017, most of the telco share losses came from the three former rural telcos--CenturyLink, Frontier Communications and Windstream.

source: marketrealist.com

source: IHS

ISPs	Subscribers at End of 2Q 2017	Net Adds in 2Q 2017	% Change
Cable Companies
Comcast	25,306,000	175,000	0.69%
Charter	23,318,000	267,000	1.15%
Altice	4,004,000	2,000	0.05%
Mediacom	1,185,000	6,000	0.51%
WOW (WideOpenWest)	727,600	-1,400	-0.19%
Cable ONE	521,724	-1,603	-0.31%
Other Major Private Company	4,845,000	15,000	0.31%
Total Top Cable	59,907,324	461,997	0.77%
Phone Companies
AT&T	15,686,000	-9,000	-0.06%
Verizon	6,988,000	-23,000	-0.33%
CenturyLink	5,868,000	-77,000	-1.31%
Frontier	4,063,000	-101,000	-2.49%
Windstream	1,025,800	-21,800	-2.13%
Cincinnati Bell	307,100	-300	-0.10%
FairPoint	304,193	-1,160	-0.38%
Total Top Telco	34,242,093	-233,260	-0.68%
Total Top ISPs	94,149,417	228,737	0.24%

source: Leichtman Research

What Needs Explaining is Telecom Price Increases

Even if one assumes there is relatively-constant pressure on retail communications service products, those price trends for fixed and mobile network services need deciphering.

Global prices, measured as a percentage of gross national per-capita income, have fallen at least since 2008, according to the International Telecommunications Union. But what requires explanation is higher prices, where they happen.

In the U.S. market, since at least 2009, prices have generally fallen for some products such as mobile service, mobile voice and texting.

Also, internet access prices have fallen about five percent since 2009. But prices for fixed network voice and content subscriptions have risen.

source: Federal Reserve Bank of St. Louis

Even prices for internet access services, generally stronger in some quarters because consumers now are buying faster services that cost more than slower services, have dipped since 2009, with most of the drop happening in 2017.

source: Bureau of Labor Statistics

The exception to the trend of falling prices are subscription TV services, which have seen growing prices since 2009. So what makes content services different from internet access and mobility? Internet access is the classic “dumb pipe” service, hard to differentiate and subject to Moore’s Law fundamental trends (constantly increasing quantity, constantly dropped price per unit).

Subscription TV is a content service, more analogous to websites, music and fashion than a “telecom” service. Also, the value of mobility these days is arguably more weighted to internet-accessed apps and content than to carrier voice and messaging functions. That is to say, more of the value of a mobile service now is the dumb pipe access to content and apps, and less the carrier voice and messaging services.

source: BLS

Ironically, one might also note that prices for fixed network voice service, a product far fewer consumers now buy, have grown since 1996, when local telecommunications was deregulated in the U.S. market. Since about 1996, prices have climbed about 35 percent. There are a couple reasons, amongst them the ability of suppliers to raise prices more easily, despite the countervailing trends of declining demand and greater competition.

The other issues are likely that the mix of business and consumer lines has been changing, with a greater percentage of business lines, compared to consumer lines, than has been the case in the past. Also, consumers who do not value fixed network voice services already have deserted for mobile services. The consumers who remain likely place a higher value on fixed network voice.  

source: ITU

On the whole, basic economic principles seem to be at work. Generally speaking, demand for any product will grow with lower prices and fall with higher prices. Higher prices for fixed network voice have definitely been accompanied by lower take rates.

Globally, fixed telephone accounts seem to have peaked about 2006.

source: ITU

Yes, 5G is a Gamble. But it is, in Some Markets, a Very Necessary Gamble

The commercial revenue drivers for 5G are not entirely clear, argues William Webb, Ofcom senior technologist. The “vision is flawed,” he argues.

On the other hand, in many markets, mobile operators will require speeds that “can compete with fiber services,” says Sam Barker, Juniper Research analyst. That means 5G is necessary, in the same way that optical fiber has been necessary to boost fixed network bandwidth (no matter how deep into the distribution network a service provider deploys it).

In that sense, it is not so useful to know that perhaps 1.4 billion 5G connections will be in service by 2025, up from one million in 2019, the anticipated first year of commercial launch, as Juniper Research now forecasts will be the case.

Many, perhaps most, of those connections likely will be accounts that already were buying 4G services. That is a familiar situation for many fixed network service providers moving from copper access to optical fiber: for nearly every account gain for “fiber-based” internet access, those providers “lose” a customer formerly buying copper-based internet access.

So 5G--though a big gamble--is likely a necessary gamble, illustrating the point that it is the business model for 5G which poses the single greatest challenge, not the technology, not the need to support small cells or better radios, not even the capital investment and spectrum, even if all are crucial elements.

“Users will not value the higher data rates that are promised and will not need the higher capacity forecast,” Webb  argues. One does not have to agree with that sentiment to note that there are huge business model challenges.

Perhaps more debatable are some of the reasons Webb believes exist, such as

“technological advances” being “insufficient” to support the platform. Many would disagree, at least to the extent that such advances will not be available in the time frame, and with the cost parameters, required to support the business case.

Webb also argues that mobile operators “are insufficiently profitable to afford it.” Some might argue that is largely true in some markets, but not all; or true for some providers in markets, but not all; and also strategically irrelevant. If survival requires what 5G can provide, and 4G cannot, then the investments must be made.

That also is not a new situation. One might argue that, at least for a couple of decades, that same situation has been true for most deployments of fiber to the home. Even when they never say so in public, executives of firms making the investments understood, and understand, that the decision to upgrade to some next generation platform is not driven by classic investment criteria--because the profit from doing so will be X--but because failure to make the investment cedes the market (and perhaps the whole business) to new competitors.

That is important. The next generation network upgrade is not based on classic investment criteria, but for strategic reasons. “You get to stay in business” is the driver, not “our revenues will grow by X.”

Webb essentially argues that the 5G investment should not be made. There merit to that argument, perhaps strategically, often tactically, and often for some suppliers, compared to others. But, in some markets, as the colloquial expression suggests, “good luck with that.”

There are markets--developed markets where 4G adoption is saturated, for example--where there basically is no option but to make the leap. The simple answer is that suppliers have run out of things to sell customers on 4G networks, especially new things that generate incremental revenue.

So it is noteworthy that Juniper Research now argues the U.S. market will have the US alone “highest number of 5G connections for fixed wireless broadband and automotive services.”

In some markets, where strategic investment in next generation networks has been most important, 5G offers a way to dramatically reduce the costs of next generation network infrastructure; immediately offering a way to produce incremental revenue gains and boosting competitive positioning.

In other words, 5G supports what 4G cannot: a way to address a huge new revenue segment: replacing fixed network internet access, in territory and outside the current footprint. Doing so out of region is more challenging, but at least in principle, would offer a chance for the largest regionally-based fixed network service providers (AT&T, Verizon, CenturyLink) as well as new entrants, to offer internet access services with a sustainable business case.

In some instances, that is not just because 5G network platforms are available, but also because other innovations, such as spectrum sharing, bandwidth aggregation across licensed and unlicensed bands and new access to unlicensed assets and new spectrum, are available.

The point is simply that even if the 5G business case is uncertain, as it is, there are markets where the gamble to deploy must be taken.