How Much Mobile Substitution for Fixed Network Home Broadband, and How Fast?

Observers continue to debate the importance of 5G fixed wireless as a driver of increased home broadband market share for Verizon and T-Mobile, as well as a cause of slower cable TV home broadband net additions.

Cable operators predictably deny that fixed wireless will prove a threat, arguing that hybrid fiber coax speeds are fast enough to stay ahead. But financial results posted by Verizon and T-Mobile suggest that fixed wireless already has proven to be a key driver of home broadband market share gains.

The issue now is the fixed wireless rate of growth, compared to new fiber-to-home additions. Despite cable executive denials, more observers believe FTTH and fixed wireless are going to eat into cable's market share over time. How fast that happens is an issue.

But there have been signs of mobile ability to substitute for fixed network service over the past five years or so.

Some studies show that users in 2018 often found 4G mobile internet access was faster than their home broadband using Wi-Fi. 

A key caveat here is that Wi-Fi device speed inside the home is not the same thing as the speed delivered to the home by the internet service provider.

source: Deloitte 

Another possible caveat is that it is not clear how many of the speed comparisons used home Wi-Fi compared to public hotspot Wi-Fi. It is almost certainly the case that home Wi-Fi runs faster than public Wi-Fi. 

Wi-Fi is slower than line speed for a number of reasons, so to note that 4G data access is faster than home Wi-Fi is not to say that 4G mobile networks perform better than ISP fixed networks. 

It is to say that a smartphone or other device able to use a 4G network might often experience faster speeds than delivered to that same device by the in-home Wi-Fi. 

This might be even more relevant for comparisons of home Wi-Fi experience compared to 5G mobile network experience, as 5G is going to be faster than 4G, once the networks are substantially or fully populated. Even as early as 2020, early 5G was faster than Wi-Fi, according to Opensignal.  

That also goes for 5G fixed wireless, as it is rolled out on a variety of frequencies globally. Generally speaking, if 4G was faster than Wi-Fi, there is a very good chance that 5G will be even faster, in comparison to Wi-Fi. 

source: Opensignal 

In countries that launched 5G early, 5G has proven to be faster than 4G and Wi-Fi, according to Opensignal. The sole exception was the U.S. market, where early 5G used low-band spectrum that does not support mid-band or millimeter wave speeds. 

source: Opensignal 

All of this is important for the development of 5G fixed wireless as an alternative to fixed line networks. In some markets, 5G fixed wireless is expected to be a key challenger to fixed network service. 

Fixed wireless has become a major driver of Verizon home broadband net new account additions. That also is true of T-Mobile home broadband net account additions in the U.S. market. 

In its first quarter of 2022, for example, fixed wireless supplied 85 percent of Verizon net home broadband account additions. 

For its part, T-Mobile got fully a third of total net home broadband additions in the U.S. market in 2021, and all of those accounts gained used fixed wireless. 

So 5G mobile and fixed wireless speeds will be a huge material factor in driving Verizon and T-Mobile home broadband market share gains, as well as a limitation on cable TV operator net additions as well.

How Fast is "Fast Enough?"

Determining what an acceptable internet access speed “ought to be” requires knowing how many users are sharing a connection, what they do when connected, how often they are connected simultaneously, the days of week and times of day when that happens. 

And it is hard to avoid the conclusion that most of the discussion we hear about “how fast is fast enough, ” when it comes to internet access and home broadband,  is a political or lifestyle or values statement, not a technology statement. 

The minimum amount of bandwidth to support one user is not the same as that to support four users, each connected simultaneously to a few devices. Even assuming every user requires streaming video support, most of the time, minimum bandwidth requirements are not that stringent. 

source: Minim

In fact, the single best rule might be the number of concurrent users and concurrent video streams that must be supported. Not many households will find they really require speeds above 500 Mbps, assuming latency performance and upstream bandwidth requirements are tolerable. 

All that said, would I willingly downgrade from my gigabit connection, which I know I really do not need, even back to 500 Mbps or 600 Mbps? No. But that is not based on any real performance advantage I can perceive. All of my requirements are, in principle, met by a connection operating at 25 Mbps. 

I haven’t had a connection running that slow in a few decades. My consumption keeps growing, but as a technology matter I cannot really argue that I generally “need” speeds as high as I pay for. 

 

source: Minim

In a sense, the behavior is many decades old, sort of done for the same reasons as I always have purchased mobile data usage buckets that exceeded my expected usage.

The logic is similar to the reasons people often buy “unlimited usage” consumption plans that typically exceed their anticipated usage. 

The value is predictability of payment amounts, not a fine-tuned analysis of price versus usage relationships. In fact, people tend to pay more than they need to, to assure predictability of the recurring payments. 

In a similar manner, having a speed “up to Xbps” means that when the network gets congested, I still can expect to have .6Xbps as a realistic experience. Obviously, what matters is the speed one actually experiences at the most-congested part of the day, when running the most demanding applications, with the most concurrent devices or users.

EU Looks at Allowing ISPs to Treat"Some" App Providers Unequally

Network neutrality has always been a slippery, impossible to define concept, notable more for its help or hindrance to business models of various participants in the internet ecosystem. Under the rubric of “treating all bits the same,” policymakers and advocates have prevented quality of service mechanisms for consumer internet access; blocked the access equivalent of “toll free” calls and generally imposed effective price regulation on internet access providers. 

All that despite the fact that application providers routinely pay money to ensure that their own bits are “not treated the same,” using content delivery networks to circumvent public internet routing delays and uncertainty. 

Now, in an ironic twist, European Union regulators are looking at imposing just such “unequal treatment of bits” on a handful of large application providers. 

Allowing ISPs to extract fees from some app providers for the privilege of allowing bits to be delivered over ISP access networks. 

One if almost forced to conclude that the network neutrality debate was never about equal treatment; equal access or anything else related to the delivery of bits over ISP access networks. 

It seemingly always was about the perceived revenue and cost advantages and disadvantages faced by various ecosystem participants. It is hard to reach any other conclusion given the extreme range of regulator opinions.

First, “equal treatment” to benefit app providers. Now, “unequal treatment” to benefit ISPs. In addition to all that, there are other political concerns, principally the impact of policies on domestic suppliers of apps, content or access. 

If we are honest we will stop pretending “network neutrality” had much to do with “protecting bits from discrimination,” and recognize it was a political move designed to help or hinder some parts of the internet ecosystem, just as it now--in reverse--is similarly designed to help or hinder ecosystem participants. 

In the latest incarnation, it is ISPs who need “revenue help.” The business simply is not growing in Europe, and ISPs seemingly have won the argument that it is they who need help, not app providers. 

source: ETNO

As often is noted, app providers have enjoyed revenue growth, while ISPs have seen their revenue shrink since the early 2000s. 

 

source: ETNO

The point is that network neutrality is shown to be a sham. The new proposals will impose unequal treatment of bits. It is the exact opposite of “net neutrality,” whatever that was supposed to be, and to some of us the concept never had integrity. 

The same people who argued for “equal treatment of all bits” also agreed that sometimes ISPs would have to treat bits unequally to preserve network performance. 

Maybe the better advice would be to stop picking winners and losers under the charade of some sort of “fairness” or “equal treatment.” It appears to be nothing of the kind. Instead, we have governments picking winners and losers for political reasons.

If You Hate Meeings, Do Not be a CEO

Frustrating though it might be, CEOs of larger organizations spend very little time with customers: about three percent, according to a survey conducted by Harvard Business School professors Michael Porter and Nitin Nohria in 2006. 

About 72 percent of CEO time was spent in meetings.

Broadly speaking, no more than 21 percent of CEO time was spent on anything connected with business strategy. About a quarter of time was spent on function or business unit reviews and another 25 percent on “people and relationships.”

About 16 percent of time dealt with “organization and culture issues.”

source: Harvard Business Review

Does Crypto Intrinsic Value Matter?

Some believe intrinsic value does matter, and crypto currencies do not possess such value. Others make the argument there is intrinsic value. 

It matters as crypto's role and value in coming Web 3.0 and metaverse use cases might hinge, to some extent, on user belief in such intrinsic value. 

The Digital Divide Will Not Always be a Problem

Scarcity--both real and imagined--drives the prices and perceived value of nearly all products and services. “Lack of” also drives the political agendas of virtually all organizations and entities who promote an agenda. 

Those organizations require resources to operate, and resources mean jobs, prestige and power. So what happens when a “problem” is essentially solved? Do organizations disband, or do they find some other “new problem” to work on, thus inviting continued support of the entity?

Almost always, the latter is chosen over the former. So we can virtually predict that, eventually, policy proponents are going to stop talking about the “digital divide” and move on to some other problem related somehow to “inability to buy broadband internet access.”

Already, many point to “digital literacy,” which is a demand issue, not a supply issue, as a substantial remaining problem. In other words, it is not the quality of the available broadband access that limits use, it is the skills of potential users. Faster broadband does not fix that impediment. 

But to the extent that generational differences exist, that problem eventually fixes itself. Younger generations are more comfortable with all new technologies than older generations, and as each generation passes, the “lag” evaporates. 

There will likely always be “differences” in available speed, latency, reliability or price between remote areas and urban areas, to be sure. Summer fruits and vegetables cost more, and are less fresh, in the winter. 

Still, at some point, internet access is going to be good enough that bottlenecks to experience and value will shift elsewhere in the ecosystem and value chain. 

Where servers are located; what customer premises gear is needed; how pricing and packaging models are crafted; which indoor transmission platforms are operating and processing speed and power could well determine whether internet apps, services and devices work at all or work properly. 

Most are now too young to have encountered it, but back in the 1980s global communications policymakers actually were concerned about how to create “voice access” platforms for most people, as “half the people have never made a phone call.” That might have been true in the 1980s or even 1990s. It no longer is true. 

We have “solved” the problem of humans having access to voice communications. We likewise will solve the “digital divide” in a meaningful sense: not defined as absolute parity of speeds, latency or cost per bit, but in the sense of “access” no longer being a barrier to usage. 

And that will lead a whole bunch of people and organizations to find some other new problem to solve.

Will Significant 5G Revenues Come from B2B? Maybe Not

The conventional industry wisdom is that incremental new 5G revenues will come from business customers, not consumers. The bad news is that, in some regions, those new business-related 5G revenues might be quite small, by 2025.

You would be hard pressed to find any observers who do not believe edge computing, private networks and network slicing will lift revenue for mobile operators over the next decade In the Asia-Pacific or any other region.

The only question is the magnitude of those increases. And that is where matters get tricky. Some forecasts suggest sharp drop offs in Asia-Pacific mobile revenue through 2025, compared to trends up to 2019. 

But most forecasts call for revenue in the range of $230 billion to $390 billion by about 2025, with total revenue--fixed and mobile--closer to $500 billion in the region. 

If 5G revenue earned by mobile operators in the Asia-Pacific region by about 2025 reach $24 billion, then 5G would represent between six percent and 10 percent of mobile operator revenues.

If one assumes that consumer mobile connections represent 90 percent of 5G revenue in 2025, and using the higher figures of $24 billion in 5G revenue, then edge computing, network slicing and private networks together would only represent perhaps $2.4 billion in revenue.

That is a small amount contributed by three new revenue sources. 

But some believe 5G might contribute less, perhaps contributing $14 billion in mobile revenues  by about 2025. In that case, 5G would represent between four percent and six percent of mobile operator revenues in 2025. 

In that case network slicing, private networks and edge computing would be negligible revenue contributors, generating perhaps 1.4 percent of mobile operator revenues. 

At such levels, the impact of changes in subscription volume, average revenue per account, increases in internet access revenues and market share changes will have far more impact on mobile operator revenues than network slicing, edge computing and private networks.

How High is Home Broadband Churn?

If we can assume a monthly churn rate for home broadband of about two percent a month, annual churn could reach nearly 25 percent of the installed base. As often is the case for consumer surveys, behavior might not match stated intentions. 

source: TiVo 

Those stated intentions seem out of line with actual monthly churn rates in developed markets, which seem to hover between 0.75 percent and 1.25 percent per month. That suggests annual churn in the range of 12 percent of the installed base. \

source: Analysys Mason 

Metaverse is a Decade Away

Some technology transformations are so prodigious that it takes decades for mass adoption to happen. We might point to artificial intelligence or virtual reality as prime examples. Now we probably can add Web 3.0 and metaverse to that list. 

At a practical level, we might also point to the delay of “new use cases” developing during the 3G and 4G eras. That is likely to happen with 5G as well. Some futuristic apps predicted for 3G did not happen until 4G. Some will not happen until 5G. Likely, many will not mature until 6G. 

The simple fact is that the digital infrastructure will not support metaverse immersive apps, as envisioned, for some time. Latency performance is not there; compute density is not there; bandwidth is not there. 

In fact, it is possible to argue that metaverse is itself digital infrastructure, as much as it might also be viewed as an application supported by a range of other elements and capabilities, including web 3.0, blockchain and decentralized autonomous organizations, artificial intelligence, edge computing, fast access networks and high-performance computing. 

source: Constellation Research 

Scaling persistent, immersive, real-time computing globally to support the metaverse will require computational efficiency 1,000 times greater than today’s state of the art can offer, Intel has argued. 

To reduce latency, computing will have to move to the edge and access networks will have to be upgraded. 

All of that takes time, lots of capital investment and an evolution of business models and company cultures. Metaverse is coming, but it is not here today, and will take a decade or more to fully demonstrate its value. Major technology transformations are like that.

Web 3.0 Will Not Prevent the Rise of Powerful New Platforms

Many would argue that since Web 3.0 is the future of the internet, and since blockchain is among the key enablers of Web 3.0, that blockchain is therefore the future of the internet. We might at least agree that blockchain is part of the foundation of the future internet, as we might argue for artificial intelligence, edge computing or the metaverse. 

One of the principles of Web 3.0 is that it is more distributed, in terms of ownership of data. That is inherently part of the design of blockchain, so there is a clear logic there. Some proponents of Web 3.0 also tout some other possible advantages, including user ownership of their own data. 

Many argue that decentralization will prevent the rise of new gatekeepers that have been a criticized feature of Web 2.0. And this is the tricky part. It remains unclear whether technology decentralization necessarily leads to dispersed power within the ecosystem, or not. 

Keep in mind that the internet is, by design, similarly disaggregated. Owners of apps and services do not have to own networks to reach their users or customers. Functions within the ecosystem similarly are disaggregated. The use of layers allows a modular approach to supplying and upgrading functions. 

At least in principle, any end user can reach any other end user, so long as that is lawful. But it does not seem likely that new platforms will be prevented from arising. Though any entity can use blockchain, that does not prevent the rise of new platforms, any more than leaders can be prevented from emerging in any industry.

The existence of a public road, rail, airline or other infrastructure does not prevent the emergence of auto, airline or electrical and energy leaders. Blockchain might, in some cases, eliminate “middle man” functions for commerce, content or application supply. 

But that disintermediation does not prevent new platforms from emerging. Suppliers will still exist. And some suppliers will gain leadership of markets. Efficiency is the benefit of blockchain: it allows disintermediation.  

source: WallStreetMojo 

But disintermediation in no way prevents the rise of powerful platforms. It simply allows greater supplier efficiency. So though some believe Web 3.0 necessarily prevents the rise of centralized power on the internet, some will disagree. In any market, for any product or service, leaders emerge. The databases, currencies and technologies we use do not seem to affect such processes.

Is Growth an Unsolvable Problem for Service Providers?

Virtually all observers praise AT&T's "return to connectivity" as the fundamental business strategy. Some hail a new era for the company. Others might point to aggressive marketing tactics that could be hard to sustain longer term, even if they work in the short term.  

But monopoly market dynamics are fundamentally different from those with competition. Slow growth is not a problem for a regulated monopoly that earns a guaranteed--if low--return from investments made with almost zero risk. 

But that same business is fraught with danger in a competitive situation, where profit margins are squeezed; bad investment choices have real consequences and new competitors reduce the effective size of the market any single firm can grab. 

The simplest analogy: in a monopoly market the theoretical share is nearly 100 percent. In a competitive market with two competent suppliers the theoretical market share is 50 percent, In a market with three competent suppliers theoretical market share is reduced to 33 percent of total. 

In practice, a stable competitive market often will have a 4:2:1 pattern of market share among the top-three firms.  

In a mature competitive market it is conceivable that one supplier gets 50 percent share; a second 25 percent; a third 12.5 percent and the rest is divided amongst scores to hundreds of suppliers. But the biggest three suppliers can have close to 90 percent share. 

Few--if any--national communications markets have reached that shape, which suggests the markets remain unstable. 

The access business (voice, internet access, messaging, mobility) has other problems, though. Competition has meant declining profit margins; a lower return on invested capital and, often, lower average revenue per account over time. Revenue growth also is a persistent issue.

And that is the fundamental conundrum big access companies (telcos, cable TV, other ISPs) face. Competitive access markets feature low rates of growth; ARPU pressures; profit pressures and low rates of financial return on invested capital. 

“Sticking to the basics” (connectivity services) was always a low-growth business in the monopoly era. In the competitive era it often is a “close-to-zero growth” or even “negative growth” sort of business. 

That remains a key issue for connectivity providers that “sticking to the core business” does not necessarily solve. Market share gains and losses will remain a key variable under conditions where big gains in ARPU are close to impossible.

Wireless Power Delivery: Kilowatts at a Kilometer

Yes, substantial amounts of electiical energy can be converted into microwave radio frequencies and delivered without wires. 

Monetization of Higher Data Consumption Remains a Key Issue for ISPs

Virtually all internet service providers worry to some degree about monetizing growing data consumption on the part of their customers. But monetization is a bigger problem in some markets than in others. 

If consumption and revenue were strictly linear, as once was the case for long-distance telephone communications, the highest usages would correlate with the highest revenue, all other things being equal. Looking at mobile ARPU, some markets including India and South Africa show the monetization issue.

soruce: Cisco, Kagan Research 

In most markets, though, monetization is sticky on the revenue side. Average revenue per user might not increase as usage grows. To remain viable under such circumstances, an ISP must reduce costs per delivered bit or find additional revenue sources with higher profit margins and stronger revenue growth profiles. 

Who is the "Speaker?" The Platform or the User? Does it Matter?

Free speech always has been a difficult and complicated subject in the United States. Time, place and manner restrictions have been upheld as lawful. But the First Amendment to the U.S. Constitution only binds the federal government. 

The Amendment says “Congress shall make no law. respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.”

Over time, many have emphasized a few key words. It is “Congress” that can “make no law” regarding the “establishment” of any religion or abridging “freedom of speech” or “press.” 

The Constitution therefore restricts the federal government, not other entities, jurists have concluded. But the meaning of “make no law” has been debated. Though intended to protect political speech, the courts have, over time, concluded that other forms of expression with political implications also are protected.

All those issues now are complicated, many would argue, by suppression of political speech by social media platforms. To be sure, such entities are not bound by the First Amendment to the U.S. Constitution. Neither are newspapers, radio stations or other publishers of content. 

But such issues have been raised before. Consider the issue of “who is the speaker whose rights are protected? In the 18th century the right was said to be held by the owners of printing presses. In the 21st century it is social media platforms. 

But where jurists might agree that a newspaper is a “speaker” for reasons of protection, who is the “speaker” on a social media platform? Is it the platform (which insists it is not responsible for the views expressed on its sites) or the users of the platform? 

And, to be sure, in either case, no matter which definition is used, the constitutional protection of speech might not apply. The platform, speaking for itself as a legal entity, has the right to express its own views. What is unclear is whether, for all other purposes, the views expressed on the platform are distinct. 

Though courts have refused to consider private property venues areas of protected speech, that arguably remains an issue. In other words, is a major social media platform the equivalent of the village commons. So far, courts have not agreed. 

Still, naked suppression of political speech arguably rankles most people. And at least so far, none of the historical precedents seem to provide much room for adapting First Amendment law to 21st century political speech.