5G Capex Will Grow, but Less Than Many Expect

There now is growing evidence that 5G capex actually will not lead to much of an increase in service provider investments, at least for many tier-one service providers. Three UK, Swisscom and Verizon are among the service providers who maintain that 5G investments will be made within the typical capex budget ranges each of the firms historically has maintained, at least compared to other periods when the next-generation mobile network has to be built.

source: Macquarie Capital

Others, including AT&T, Sprint and T-Mobile US, might well find capex increasing, partly as a cyclical matter. Capex requirements always build at the start of a new mobile generation, then taper off as the new network reaches completion.

source: Edge of the Cloud

“I think you'll see capex stay within reasonable bounds of what you've seen from us in the past,” says Matthew Ellis, Verizon CFO. By that he means capital expenditures for the full year 2018 will be between $16.6 billion and $17.0 billion. That noted, Verizon and AT&T have spent in 2018 more than originally forecast by company executives.  

5G IoT Connection Density Will Drive Edge Computing

Aside from support for ultra-low latency applications, another driver of edge computing in the 5G era is the potential increase in the number of devices seeking to communicate over a mobile network.

Around 10 percent of enterprise-generated data today is created and processed outside a traditional centralized data center or cloud. By 2022, Gartner predicts this figure will reach 75 percent, says Santhosh Rao, principal research analyst at Gartner.

That suggests the magnitude of the demand for edge computing and fog computing facilities. When 75 percent of enterprise data has be processed outside a cloud or enteprise main computing center, an awful lot of edge computing will have to happen.

Gartner defines edge computing as data processing at or near the source of data generation. Edge computing serves as the decentralized extension of the campus networks, cellular networks, data center networks or the cloud.

Connection density is part of the International Telecommunications Union specification for IMT-2020 (5G) because so many more devices are expected to be connected to networks when internet of things appliances and sensors proliferate. The 5G standard stipulates that 5G networks must support up to one million connected devices per square kilometer (0.38 square miles).

That increase in IoT devices to be supported also is a driver for thinking about edge and fog computing. With so many additional devices communicating with servers, it will be quite helpful to minimize the amount of traffic that has to cross network backbones to reach cloud computing centers.

Will Verizon 5G Fixed Wireless Upset ISP Economics?

Competition really changes the economics of consumer access networks, limiting revenue and constraining profit margins while increasing the amount of stranded capital investment. In the fixed networks business, capital is stranded when deployed, but earning no revenue.

Consider any one-product service provided by a single provider with 80 percent installed base or take rate. Add one competent new supplier, and theoretical “maximum” share for each provider drops to 40 percent. Add a third competent supplier and theoretical maximum share drops to 33 percent.

As a practical matter, dropping from 80 percent household take rates to 40 percent automatically doubles the cost per customer. When there are three competent suppliers in any single market, theoretical market share can triple per-customer costs of any network.

That change alone would destroy the original economics and business model. Triple play now is foundational because it directly addresses competitive market dynamics: three products can be sold to a smaller number of potential customers.

If each unit sold (for example) is about $40, then per-account revenue could be $120, not just $40 if a single product were the only supplier option. Such bundling practices help service providers compensate for lower take rates.

And that is the principal threat Verizon’s fixed 5G initiative poses for AT&T, Comcast and Charter. If Verizon is successful at taking accounts, the theoretical maximum market share for any of the providers drops from 50 percent (two equally skilled providers split the market) to 33 percent (three equally skilled providers each take a third of the market).

As always, opinions vary about the potential amount of substitution, in urban or rural markets, with the greatest skepticism about impact on rural markets. In many cases, that is because of signal propagation characteristics of millimeter wave spectrum in such bands as 28 GHz and 39 GHz.

It is reasonable to suggest that many believe lower-frequency signals in the 3.5-GHz band might be more important in rural areas, as 3-GHz and 5-GHz frequencies already are used by fixed wireless providers in rural areas.

Analysts at CoBank are in the camp that believes millimeter wave spectrum will not have very much impact on rural internet access, and also believes fixed wireless using millimeter spectrum will be less successful than Verizon estimates.

Still,  others might note that, in urban and suburban areas, even using skeptical estimates, Verizon alone might be able to take significant market share from other internet service providers using cabled approaches.

In the first five years or so, such gains might only represent share gains between 11 percent to 18 percent, in the areas where Verizon builds fixed wireless networks. In substantial part, such skepticism is founded on transmission distances that might range from 500 foot  to 1,000 foot cell radii.

To be sure, skepticism on the part of many observers, such as rural telcos who rely on cabled approaches, is to be expected. After all, fixed wireless is yet another alternative to existing telco or cable TV networks.

CoBank analysts believe the general absence of cable TV operators from the ranks of bidders for new millimeter spectrum indicate those firms are not worried about millimeter wave spectrum approaches. If they were threatened, CoBank analysts argue, cable operators would be bidding for spectrum, either to deny the use of such spectrum to would-be competitors, or to use the platform themselves.

Some of us would disagree with that logic. For institutional reasons, cable executives have disliked running services over any platforms they do not own, with a strong preference for HFC whenever possible. Already looking at use of leased facilities to support their early mobile efforts, cable executives might resist adding more one platform, especially if they conclude the revenue upside is limited.

Cable operators also seem to favor lower-frequency bands, such as 3.5 GHz, in part for reasons of cost, in part for reasons of signal propagation.

Nor would cable operators be able to simply “warehouse” spectrum very effectively, on a long term basis. “Use it or lose it” is the standard policy for new spectrum awards. And even if cable operators wanted to warehouse 28 GHz or 39 GHz spectrum, lots of other spectrum in the millimeter bands will be released for commercial use, and much of that spectrum will be available on an unlicensed basis (or licensed on a no-fee basis).

source: CoBank

Will 5G Eliminate Mobile Price Gap with Fixed Access?

Assuming that 5G delivers on its promise of faster speeds, around a third of customers indicated that they’d be willing to pay $5.06 more per month on average, versus $4.40 more per month for mobile customers, PwC says.

More would pay a premium for 5G if it provided “better quality video” on their mobile device. Some 29 percent of respondents suggested they would pay more for better video, compared to 25 percent of respondents similarly willing to pay more for better video on their fixed service.

The key caveat here is that most people do not know what the term 5G means. They have never had a concrete experience with 5G. They may not know they have to buy new devices. They have not been presented with a concrete offer. Nor have they been able to evalue better 4G services against new 5G offers.

Moreover, it is possible that the real value of 5G for new use cases will emerge most clearly on the enterprise side of the market, to support internet of things use cases, not consumer broadband. We soon will begin to test actual demand in the real world, and surprises could emerge.

People simply cannot make fully rational choices about products they have no experience with, with no actual pricing information, terms and conditions of use. Steve Jobs’ advice likely is warranted in that regard. People cannot tell you whether they want a product they have never seen.

A survey by PwC found that many consumers believe they already pay too much for home internet access. Some 51 percent of respondents agreed that their fixed network service was too costly.

Some 36 percent of respondents likewise agreed that their mobile internet access was too costly.

That finding is somewhat paradoxical, as PwC notes the average cost of fixed network internet access is $0.34 per gigabyte, compared to $20.02 per GB for some mobile services. Others would contest that characterization of mobile internet costs.

source: Mobile Experts

Most would likely agree that mobile data tends to cost an order of magnitude more than fixed access. But many believe 5G millimeter wave networks will produce mobile costs in line with costs of fixed networks. That already can be seen in some cases for 4G networks.

Spectrum abundance is going to replace capacity scarcity in the mobile business in the 5G era. The key point is that the historic price relationships between fixed and mobile internet access services are changing, with price parity between mobile and fixed network prices becoming possible for the first time.

Why Spectrum Sharing is Such a Big Deal

If you have ever wondered why spectrum sharing is such a big innovation, take a look at how U.S. spectrum for communications is presently allocated, as shown in the National Telecommunications and Information Administration frequency allocation chart.

As you can see in this portion of the chart, there is almost no spectrum that is presently not allocated for some uses, and some users. So when new demands arise, such as a need for additional mobile spectrum, something has to give.

The solution in the past has been to reallocate spectrum, moving present users elsewhere in the band. But that takes time and money. Spectrum sharing is a way of opening up existing spectrum for new uses, without the cost and time of moving existing users.

Who Wins 5G, and Why?

A dominant or at least quite prevalent line of thinking these days (not the least because some business interests gain from such thinking) is that whoever gets to 5G first, in terms of coverage, “wins” the race to wring advantage from 5G.

While 5G coverage is necessary, it is not sufficient to wring huge financial or economic benefits from the new network. The reason for that thought is what happened in the 4G era, when one might similarly have argued that early deployment would lead to economic benefits associated with leadership in a “growing or leading” part of technology and economic life.

So here’s the argument, in a nutshell: the real upside in the 4G ear came from application providers, not access providers. Sure, the networks had to be there, but the incremental value and growth came from ability to create compelling, widely-used applications.

And that was not a direct result of having deployed 4G networks. Instead, the advantages flowed to places where app creation skills and capabilities were more advanced. Europe did not “fall behind” because it did not deploy 4G; it “fell behind” because its app-creation industries were not as well placed.

And that is why China and the United States generally are considered the key areas where new innovations might come in the 5G era. But it has nothing to so with deploying the access networks, in a direct sense. It has to do with the ability to conceive and commercialize new apps and use cases that take advantage of 5G network attributes (along with the other changes in networks that come with 5G).

Use of Millimeter Spectrum Will Be Essential by 2022

Though much depends on what happens with the proposed T-Mobile US merger with Sprint, it is likely that the leading U.S. mobile service providers will require millimeter wave spectrum by about 2022, as they will no longer be able to supply capacity demand using only assets below 6 GHz.

In other words, the traditional low band and new mid-band spectrum (3.5 GHz, 5 GHz) will not be sufficient (presumably including offload to Wi-Fi, and assuming spectrum aggregation is used) to support operations.


source: Mobile Experts

A shift of video consumption to mobile, and away from fixed networks, is a prime driver of demand. "From our perspective, mobile is winning, and cable is losing," said Joe Madden, Mobile Experts principal analyst. In a broad sense, all linear services are losing share.

And the losses are both to streaming alternatives and “mobile streaming” alternatives.

"In the US market, the transition has already begun. T-Mobile offers unlimited Netflix over their network. AT&T is launching a direct-to-consumer streaming service for HBO and other Time Warner content.

Cable companies are losing TV customers quickly, but will soon begin to lose internet customers as well; the market is being forced to adapt or fall apart."

But there also are two elements to this potential shift: fixed network providers losing some market share in internet access to 5G networks (fixed and mobile modes) and linear service providers losing some account share to mobile versions.

To be sure, spectrum supply is growing, along with demand. South Korea recently auctioned 280 MHz of the 3.5 GHz band and about 10 times that amount in the 28 GHz band.
‘
In the United States, the Federal Communications Commission will auction spectrum in the 24-GHz to 47-GHz bands, in addition to 3.5 GHz CBRS band and the 3.7-GHz to 4.2-GHz band.

Many countries are auctioning 3.5-GHz spectrum. Such moves create more spectrum abundance in some countries, especially when spectrum aggregation and use of unlicensed spectrum or shared spectrum are considered.

Traditionally, a shift to smaller cell sizes (cell division) has been one of the two primary tools for growing capacity, in addition to getting more spectrum. Through 1995, smaller cells were the more important means of creating usable capacity.

Use of small cells will increase dramatically in the 5G and millimeter wave eras, beyond greater use of small cells to support 4G networks.


source: Frank Rayal

After 1995, additional new spectrum has become more important. Still, smaller cells have yielded the great bulk of the usable capacity increases.


source: Science Direct

5G Will Drive Entertainment Industry Revenues

New 5G mobile networks will have a big enabling role on the media business, researchers at Ovum predict. Advertising and new forms of enhanced video will likely drive much of the revenue growth, Ovum predicts.

The average monthly traffic per 5G subscriber will grow from 11.7GB in 2019 to 84.4GB per month in 2028, at which point video will account for 90 percent of all 5G traffic, according to a study by Ovum, sponsored by Intel.

The average monthly traffic per 5G subscriber will grow from 11.7GB in 2019 to 84.4GB per month in 2028, at which point video will account for 90 percent of all 5G traffic.

source: Intel

Between 2019 and 2028 media and entertainment companies will be competing to win a share of a nearly $3 trillion cumulative mobile-enabled revenue opportunity. As always, access providers will have to work to create roles for themselves in the 5G media ecosystem.

Experiences enabled by 5G networks will account for nearly half of this revenue opportunity (close to $1.3 trillion), Ovum suggests.

As early as 2025, 57 percent of global wireless media revenues will be generated by using 5G networks and devices.

Revenue earned on 5G networks are expected to grow steadily.

By 2022, nearly 20 percent of total revenues--$47 billion of $253 billion--will be earned on 5G networks. By 2025 more than 55 percent of total revenues--$183 billion of $321 billion--will be earned on 5G networks. In 2028 nearly 80 percent of total revenues--$335 billion of $420 billion--will be earned on 5G networks.

Cisco long has forecast that video will dominate network traffic. “Globally, IP video traffic will be 82 percent of all IP traffic (both business and consumer) by 2021, up from 73 percent in 2016, Cisco has predicted.

Global IP video traffic will grow threefold from 2016 to 2021, a compound annual growth rate of 26 percent. Internet video traffic will grow fourfold from 2016 to 2021, a CAGR of 31 percent.

source: Cisco

Higher-definition video is one driver of capacity demand. Where standard definition video works at 2 Mbps, high definition video requires 7.2 Mbps. Television at 4K requires 18 Mbps.

source: Cisco

Openwave Mobility expects that up to 90 percent of 5G traffic could be mobile video. AR video can use 33 times more data than 480p video, notes John Giere, president and CEO of Openwave Mobility.

In the 4G era, especially from 2010 to 2015, increased minutes of use have been driving higher consumption of video on mobile networks.

Since 2015, growth in mobile video has come significantly as a result of a move to higher bandwidth HD content, rather than greater watch time only.

Security is an Issue for 62% of U.S. Adults Own at Least 1 Connected Device

Security is an issue for all connected devices, but especially for consumer internet of things or connected devices such as televisions, which are designed for convenience first, security second. 

Most Futuristic 4G Apps Did Not Flourish

It is very hard to predict what the 5G killer app, or set of apps, might be. We have been wrong about killer apps for 3G and 4G.

Many innovations hoped for in the 3G era arguably flourished only in the 4G era. And though some predictions about 4G proved correct, many did not. New 4G networks were expected to support user live streaming of video (no), ad hoc mobile gaming (no), cloud-based apps (yes), navigation with augmented reality (no) and telemedicine (no).

So it is conceivable that although 5G will see the early years for many new use cases, those applications will thrive only in the 6G era.  

Early on in the 3G era, video calling was a hoped-for new “killer app.” That did not happen. But it has become commonplace on 4G networks. In a similar way, content services were expected to flourish in the 3G era. That did not happen until 4G.

Augmented reality apps were supposed to develop on 3G networks. That still has not happened.

In fact, many would find it hard to point to a killer app  for 3G. Eventually, new apps do emerge. And some might say the early value of 4G was just speed.  

You might argue text messaging was the new killer use case for 2G. You might suggest mobile email and Internet access was the legacy of 3G. Video entertainment is developing as the singular new app that defines 4G.

Internet of things is expected to drive the futuristic new use cases for 5G. Many believe 5G fixed wireless, though, will represent the early new use cases, beyond bolstering consumer internet access speeds and augmenting 4G capacity in geographic areas where congestion is an issue on busy cell sites.  

Still, it is fair to say we are likely to be surprised by what 5G brings, or fails to bring, as has been the case for the other next G mobile networks.

Among the possible surprises could well be that hoped-for new use cases do not flourish in the 5G era, but take longer to reach near-ubiquity. That is why some believe 5G will wind up being more like 3G--a platform whose innovation lagged expectations--than 4G, a network platform that some would argue largely developed on the promise of new apps enabled by greater capacity.

As was the case for 4G, which provided significantly faster speeds, making video consumption feasible, 5G will boost access speeds, by an even greater margin. But the bigger long-term change is the shift to ultra-low latency.

It would be a reasonable supposition that use cases for higher speed will develop sooner, with greater scale, than use cases for ultra-low latency. And that is why some believe 5G might be more like 3G than 4G.

The wild card is the role of enterprise apps and use cases. Some of us argue that what makes 5G clearly different from all prior mobile generations is that brand-new use cases are more likely to emerge in the enterprise segment of the market.

In the case of prior generations, growth and use cases were lead by consumer applications. That is unlikely to be the case for 5G, if internet of things actually emerges as the killer set of apps.

3.5-GHz Auction Prices Too High?

Over the past few years, some have worried about the cost of 5G spectrum, although spectrum prices are dropping, generally speaking, in part because there is a huge increase in supply, and because mobile operators must now more carefully weigh the cost of new spectrum against expected financial return.  

Also, firm strategies now vary. Some firms believe use of unlicensed spectrum will be more important. Others substitute small cells for additional spectrum. Some need additional spectrum more urgently than others, based on present holdings.

Recent auctions of 3.5-GHz spectrum have no clear pattern, especially since the various auctions featured different amounts of total spectrum and different license allotments (bigger or smaller amounts of spectrum per license), and there always are local market drivers (some contestants have greater needs for spectrum).

The point is that supply and demand issues affect price, as always. Finland offered the most 3.5-GHz spectrum, at 390 MHz. Spain and Italy each sold 200 MHz. U.K. regulators auctioned 150 MHz.

Finland’s prices wound up at 0.04 euros per megaHertz pop (a MHz POP represents one megahertz of bandwidth passing one person in the coverage area).  Spectrum sold for 9.07 euros per MHz POP in Spain, but a whopping 0.51 euros per MHz POP in Italy. U.K. spectrum sold for 0.17 euros per MHz POP.

On the other hand, at least one Australian official worries that a recent big merger between TPG and Vodafone will reduce demand and lead to lower prices.

source: GSMA

One might simply argue that supply and demand will  work. Whatever the limits on new spectrum at 3.5 GHz, regulators simply must make more spectrum available in other bands. More supply takes care of pricing pressures. Releasing more unlicensed spectrum, spectrum sharing, spectrum aggregation and additional spectrum in the millimeter wave bands all will help ensure there is plenty of 5G spectrum and that prices will not be onerous.

Supply in the 3.5-GHz auctions will be something of an issue, in most countries, as there is not lots of spectrum available there.

source: GSMA

French regulator Arcep’s chief Sebastien Soriano announced it will be challenging to keep prices low for a 2019 spectrum auction, especially when supply cannot keep up with demand, but said it wants to find a way to do so.

Traditional government views of spectrum auctions as easy ways to raise government revenues also are issues. Policymakers have to balance the need to make lots of spectrum available with the desire to raise revenues.

As you would expect, firms that have paid high prices justify their actions by arguing the new spectrum will help reduce costs per gigabyte, as well as supporting all the new end user demands for capacity.

High spectrum prices, though, have been big problems for mobile operators. Recall the high prices paid for 3G spectrum in many countries, which nearly bankrupted several firms in Europe, the high prices paid in India and in some other countries.

More supply will help keep prices within reasonable ranges.

Wi-Fi and 5G as Peer Networks

With the growing reliance on network virtualization, new questions can be raised about the control plane for access that combines 5G and W9-Fi, and perhaps especially creating new ways of integrating mobile and Wi-Fi networks as peers.

The Wireless Broadband Alliance argues three broad approaches are possible, including access centric, core-centric and above-the-core.

The access centric approach is represented by 3GPP Release 13 for LTE access, with LTE WLAN Aggregation (LWA) for integrating trusted Wi-Fi with LTE and LWIP for integrating untrusted Wi-Fi.

Core-centric approaches were created by 3GPP Release 8 for the integration of untrusted Wi-Fi and in Release 11 for integrating trusted Wi-Fi into LTE’s Evolved Packet Core. Core-centric integration within a 5G deployment can either be achieved using Non-Stand Alone (NSA) or Stand Alone (SA) configurations.

source: WBA

Above-the-Core integration uses Multi-Path TCP and MultiPath Quick UDP Internet Connection (QUIC) protocols, and have been proposed for integrating Wi-Fi and mobile networks.

Above-the-core solutions are designed for mobile devices that have several or many wireless interfaces simultaneously active.

When the different access networks have different delay characteristics, above-the-core solutions can selectively decide to use only the lower latency path to deliver improved user interaction.

Above-the-core solutions can use the multi-path capability to handle the transition between the access network coverage areas.

When the different access networks have different probability of coverage, above-the-core solutions can use the multi-path capability to opportunistically leverage the local area coverage whilst having an “anchor path” over the wide area network.

When different access networks are deployed by different organizations, (a carrier Wi-Fi provider and a mobile carrier), above-the-core solutions can aggregate the connections with different IP points of attachment.