Gigabit or 100 Mbps? Largest U.S. ISPs Make Different Choices

Why is Comcast, the largest U.S. internet service provider, pushing to make gigabit service available to all its customers, while Charter Communications, the second-largest ISP, only will sell 100 Mbps?

In each case, the decisions reflect strategic choices about what is required in their markets. Charter, the second-largest ISP, takes a minimalist approach at least in part because many of its systems are in rural areas where 100 Mbps is a top-of-the line service. Comcast, on the other hand, operates mostly in big-city markets where competition from other gigabit providers can be expected.  

That does not mean those suppliers actually expect most consumers to buy their fastest service. In fact, they probably expect that most consumers will not do so. But those expectations also drive thinking about how much investment, and what type of investment, to make it their current generation access platforms.

That is true in other markets, as well.

In the United Kingdom, direct fiber connections on at least one access link remain a small fraction of total internet access connections is about two percent, according to U.K. government estimates. All-copper digital subscriber lines account for 44 percent or so of total connections; fiber-to-curb about 34 percent; and cable TV connections about 21 percent of total.

Speeds of Broadband Technologies as seen in December 2016 (ordered by median speed)
Provider	Ratio of tests	Down Speed of bottom 10% Mbps	Median Mbps	Mean Mbps	Median Upload Mbps	Mean Upload Mbps	Down Speed of top 10% Mbps
FTTH/FTTP	1%	16.7	49.1	72.6	16.3	32	160.1
Cable	20.6%	8.9	43.4	52.4	5.4	6.4	104.9
VDSL2/FTTC	33.9%	13.3	30.1	31.8	7.1	7.6	51.9
Mobile	N/A (*)	2.3	14.5	21.3	3.1	4.4	46
Fixed Wireless	0.14%	3.1	14.3	27.8 (**)	2.1	14.2 (**)	55.4
Satellite	0.04%	1.6	12.2	12.9	0.1	0.3	25.3
ADSL/ADSL2+	44.3%	1.1	5.5	6.8	0.6	0.6	14.8
(*) Mobile was not included in the proportion of tests as the assumption is the majority of the tests are people who have a fixed line option too. (**) We advise caution on these two wireless mean speeds, as they appear skewed to due a small number of tests that are symmetric and in the 200 to 300 Mbps region, this may be people testing connected via Ethernet to a mast or as one or two fixed wireless providers suggest they provide Fibre to the Premises in a small part of their footprint.

source: Thinkbroadband

In the United States, cable TV companies lead, in terms of speeds, for most consumers, even if independent providers and telcos support fiber-to-home networks. In the third quarter of 2016, cable operators had 63 percent of the installed base of fixed internet connections, and have added virtually all the net new accounts (more than 100 share of new accounts), as they  have been for a few years.

That means that cable access platforms will use hybrid fiber coax (fiber plus copper), even if others (AT&T, Google Fiber, independent ISPs) deploy fiber to home networks. Cable operators use HFC because they can: gigabit speeds are possible on the standard hybrid networks. Telcos and independent ISPs must opt for FTTH to supply gigabit speeds.

Over the next couple of years, all of Comcast’s customers (Comcast is the largest U.S. ISP) will be able to buy gigabit service, and many of the other leading cable operators will be making that same move. For the moment, Charter Communications, the second biggest ISP, will hold top speeds to about 100 Mbps, across all its territories, though some customers (former Time Warner Cable customers) might be able to buy 300 Mbps.

In rural areas, fiber to home might not always be possible, but cable hybrid fiber coax networks might. That is important because the industry-standard DOCSIS 3.1 platform supports gigabit service on standard HFC cable networks.

The point is that decisions about “speed” vary, even between the largest and number-two U.S. ISPs. So do decisions about “minimum viable” access platforms.

Why 5G Will Have to Be Different

There are 5G skeptics who argue ISPs cannot afford it, or that other elements of the business model are questionable, mostly built around economies of scope and scale that might not develop, and certainly not by the 2020 time frame touted by backers in Korea, Japan and the United States. And some simply question whether most consumers actually will want to pay for it, much less need it.

Reliance on new millimeter wave spectrum also causes some concern, as signal propagation will be an issue. Use of small cells will require significantly-new backhaul infrastructure as well, while frequency-agile handsets might be pricey.

But there is one area where supporters and skeptics do agree, and that is that if and when 5G emerges as a sustainable platform, it will be because big new revenue streams and applications have emerged, as well. Virtually all agree that will, or could happen, because of the internet of things and machine-to-machine applications.

In other words, should 5G prove sustainable, it will be because new and vital enterprise apps have developed, although new fixed wireless services in the consumer segment might prove to be quite helpful, for the business case.

There are clear signs that such new thinking already is driving capital investment decisions. Consider the way Verizon is deploying new optical fiber in Boston. In the past, FiOS has either an “either, or” proposition: either full residential deployment or not.

Now, Verizon architects seem to have shifted to a network that uses dense fiber trunks suitable for supporting macro and small cells, and then extends the distribution network from there. In some ways, that One Fiber network is a mirror image of the network cable operators want to deploy to support their eventual W--Fi-first mobile networks.

Some leading cable operators, with a heritage serving the consumer segment, are building dense public Wi-Fi networks on the back of their consumer internet access services, essentially trying to create an enterprise service on the backs of a consumer service.

Verizon is taking the other tack, essentially building an enterprise network (macro and small cell backhaul) that helps create a new IoT capability, and that in turn lowers the cost of creating a  consumer capability (fiber to the home). It already is clear that tier-one service provider fixed network revenues are driven by the enterprise segment, not consumers.

Many would argue that the XO Communications acquisition is part of that overall direction, as XO contributes a fiber-rich enterprise customer network. And, as always is the case, having direct fiber access to any anchor tenant lowers the cost of reaching the next set of customers.

Somewhat ironically, “lots more fiber” to support enterprise apps and IoT makes high-bandwidth services for consumers more feasible. But the new framework is “bandwidth to the customer,” not always “fiber to the customer.” One Fiber definitely is “fiber to the tower” or “fiber to the cell site.” I

In many cases One Fiber might also mean “fiber to the business.” Once that is done, the economics of supplying gigabit to the small business or residential customer are a lot easier. In some cases that might mean full fiber-to-home deployment. In other cases, fixed wireless might be the solution.

The point is that the legitimate concerns about the 5G business model are related to enterprise services, enterprise networks and fiber deployment strategy in a way that has not been true for earlier mobile network generations. In that sense, and others, 5G is different.

Why Both 100-Mbps and Gigabit "Top Speeds" Make Sense

Decisions about internet access speeds always are a mix of supply and demand drivers, as suppliers invest in capabilities they believe potential customers will buy, at specific price points, at levels that are sustainable long term.

That is why the second-largest U.S. internet service provider--Charter Communications--sells 100 Mbps connections as its top offer, while Comcast and other cable companies are rolling out gigabit connections.

Charter could upgrade to a gigabit, but clearly believes the market will pay for 100 Mbps. If so, then investing in more-expensive gigabit connections does not make business sense.

Comcast and others (AT&T, many independent ISPs and other cable companies) think upgrades to a gigabit are required for competitive reasons (headline speeds and marketing), even if they believe most consumers will not choose to buy such services.

The key observation is that nobody actually has found that most consumers are willing to buy gigabit connections when they also have a choice of 100-Mbps up to 300-Mbps choices that cost less.

In other words, investing in gigabit platforms almost always--so far--involves a determination that most consumers will not buy that product, and instead will opt for a lower-speed--though still fast--connection.

So why supply gigabit services? The answer is because “our competitors do so.”

Comcast primarily operates in major urban markets, where competition is more robust, and where it faces Google Fiber offering gigabit services. That is true for AT&T and CenturyLink as well. Even if only 10 percent of customers actually choose to pay for gigabit services, that still sets market expectations, and no leader wants to face the marketing claim that “it is not the leader” in speeds.

Customer demographics also can play a role. Charter historically has operated in smaller markets, where competition is less robust, though the new Time Warner Cable assets primarily are in larger urban areas.

Comcast, on the other hand, mostly operates big-city networks, where it faces competition from other ISPs presently, or soon, to offer gigabit speeds.

So gigabit headline speeds matter, in big markets, even if suppliers realize most consumers will not buy them, yet. Google Fiber, and some other gigabit providers, also have found demand for gigabit connections less robust than they had hoped. EPB reports that about eight percent of its internet access customers buy the gigabit service, for example. Google Fiber might have wound up getting 10 percent or less take rates for its gigabit service, priced at $70 a month.

Under such conditions, a range of decisions, ranging from “top speeds of 100 Mbps” to “top speeds of a gigabit” or even 2 Gbps, make business sense. Demand matters when supply is considered.

Cable Operators Largely Drive U.S. Fixed Network Speed Increases

Competition in U.S. internet access markets, despite some concerns, seems to have spurred some of the most-aggressive speed upgrades ever. According to Ookla, U.S. fixed network access speeds have averaged more than 50 Mbps for the first time, ever.

Between 2015 and 2016, fixed network internet access speeds grew more than 40 percent.

One might be tempted to argue that it is gigabit access offers from Google Fiber, AT&T or Comcast and other cable operators that are driving the speed upgrades. That is unlikely. There simply are not enough actual gigabit customers to affect national trends. Where gigabit and other speeds are sold, it seems unlikely that more than single digits worth of customers actually buy.

Instead, the key shift is upgrades by cable operators, particularly upgrades into the hundred-megabit range.

source: Ookla

Mobile speeds also have grown by more than 30 percent since 2015 with an average download speed of 19.27 Mbps in the first six months of 2016.

Overbuilder Model Remains Challenging

Though many have high hopes for the role of municipal or other types of public-private internet access ventures, competing as the third provider in an access market where cable companies and telcos already operate has always been a tough proposition.

Google Fiber is the latest, but hardly the only example. Overbuilders--as fixed network access providers competing against the incumbents are known--always have had a hard time sustaining their businesses because market share is so hard to wrest from the incumbents, despite episodic waves of enthusiasm for overbuilder business models of various types.

Though many think that is because of nefarious behavior on the part of incumbents, the big problem is simply the dynamics of market share in a three-provider fixed network market. Even assuming equal levels of skill and commitment, it has been very difficult for “number three” providers to wrest more than 20 percent share in any market.

That, in turn, “strands” a great deal (80 percent, potentially) of the network assets, and dramatically raises the “cost per customer.” For some key services, such as video entertainment, matters are even worse, as the competition also includes Dish Network and DirecTV (owned by AT&T), which further reduces the likelihood that any overbuilder offering video is going to get significant share.

Some believe Google Fiber, for example, got 10 percent or less of its Google Fiber internet access customers to buy its video service. If Google Fiber in some instances got 10 percent internet access market share, that implies video share of about one percent. And that is not sustainable.

And though some believe Google Fiber might be sold, that does not exactly square with the investment Google Capital recently made to acquire a stake in RCN and Grande Communications, two overbuilders that are being acquired by TPG, a private-equity firm. It would seem incongruous--if Alphabet really wanted to rid itself of Google Fiber--to simultaneously invest in other overbuild assets.

A decade and a half ago, a wave of overbuilders sprung up, with a fairly-standard argument about sustainability. By offering superior triple-play packages, overbuilders would be able to get 15 percent to 20 percent market share, eventually. At the top end of that range, and three anchor products to sell, the logic was sound.

Few have actually managed to hit that 20-percent market share level, though. Arguably, the few that have done so operate in markets where there are not, in fact, two strong incumbents (telco and cable). EPB in Chattanooga, Tenn., for example, is the poster child for overbuilder hopes, having gotten as much as 45 percent of consumer market share in its service area (with share defined as revenue-generating units, not “accounts” or “homes”).

EPB’s internet access share might be about 27 percent, its video share lower than that. An interesting statistic, in that regard, is that about eight percent of EPB’s internet access customers buy the gigabit service.

EPB’s market share is highly unusual in most overbuilder markets, as EPB arguably competes with Comcast, not AT&T, which has negligible video share in Chattanooga.

In a triple-play market, that is important. Comcast might have 61 percent video share, while EPB might have 36 percent share, leaving only three percent video share held by AT&T. Essentially, EPB has become the number-two provider, relegating AT&T to third place, something that is not the case in most other U.S. markets where three mass market fixed network suppliers compete.

The point is that the economics of overbuilding have been difficult for decades, which is why so few are attempted and why so few succeed.

As Google Fiber’s experience indicates, the overbuilder business model remains challenging, unless one of the incumbents essentially chooses not to fight.

Why Mobile Video Might be So Important

Most U.K. consumers now buy at least parts of a bundle of services (internet access, voice, video, mobile), EY study found. Some 93 percent of U.K. broadband households now have some form of bundle, EY found. That same study also found that TV and mobile bundles score best in terms of satisfaction and loyalty.

That is an indicator of why Verizon, focusing more on becoming a mobile advertising platform for video services and app, and AT&T, which is more intent on becoming a force in mobile content delivery, are working on mobile content delivery systems. If the U.K. preferences wind up being seen in the U.S. market, then the “best possible” bundle will be video entertainment plus mobile service.

At the moment, the most-popular U.S. bundle likely is “internet access plus TV.” That is a preference parallel to “mobile plus TV,” with one twist. Where the most-popular fixed network bundle arguably is TV-and-internet, the most-popular mobile package could naturally become “mobile voice, mobile internet, mobile video,” for the simple reason that buying mobile internet access always comes with voice and messaging included.

In that sense, the “natural” mobile bundle is going to be “bigger” than the natural fixed line bundle, simply because purchasing mobile video will assume the presence of mobile internet access, which in turn presumes the customer also gets voice and messaging (the services are stacked upon each other).

That natural bundle also shows why mobility platforms might become even more powerful: they are the ultimate platform for bundling all ubiquitous consumer communications services and most ubiquitous apps.

The other study finding that reinforces all thinking about bundles is that 55 percent of consumers would buy a bundle “only” if it also represents a price discount. That is logical. Consumers and suppliers are used to the notion of volume discounts.

source: EY

Business Model, Not Technology, is Key to Mobile Substitution for Fixed Net Internet Access

As much as platform capabilities underpin mobile substitution for internet access, fundamental changes of business model are more important. In the U.S. market, for example, the packaging is quite different.

Mobile broadband is priced according to usage, generally in the form of buckets of use. Fixed internet access is priced based on speed (faster speeds cost more), but usage allowances are big enough that usage effectively is “unlimited.”

In other words, mobile retail packaging is based on usage, while fixed access is packaged on “speed tier.”

That poses a key problem for mobile service providers who want to encourage users to substitute mobile access for fixed access: packaging has to replicate what consumers presently expect. That means a shift away from “usage-based pricing” and towards “speed-based pricing” to a large extent.

Some glimmers of that already can be seen. As the tier-one mobile operators move to encourage consumption of video services on mobile devices, they increasingly are exempting data usage. That effectively “levels the playing field” where it comes to video entertainment service data charges.

When consumers buy a fixed network video service, they do not expect to be charged separately, and additionally, for bandwidth. They buy content, and use of the network is allowed. That is the new model mobile operators are moving towards. There are two different angles of that expectation.

When users consumer Netflix, Amazon Prime, Spotify or Pandora content over a fixed network, they do not worry about the impact on their data usage, as the fixed network plans are abundant enough that this is not a concern. So usage allowances that eliminate the need to worry about data usage are foundational.

Mobile operators provide that same sort of assurance by simply allowing consumption of some video services without imposing data usage charges.

Where mobile data might cost about $9 to $10 per gigabyte (GB), fixed access might cost as little as 15 cents per GB.

source: Mobile Experts

The point is that, as important as platform innovations are, the retail packaging is even more important.

What "EBITDA" Instead of "GAAP" Profit Tells You

Without wanting to be unduly bearish, the global “telecom” industry is less healthy than it appears. Consider only the shift that has to be made in describing “profit.” According to Ericsson, global revenue will climb about 2.4 percent each year to 2018, with growth of earnings “before interest, taxes, depreciation and amortization” of one percent to 2018.

That shift from “generally accepted accounting principles” to EBITDA tells the story: the global telecom industry no longer is “profitable” in the GAAP sense.

Companies that operate in capital intensive, such as telecom, “do not give investors accurate depictions of performance through the EBITDA margin,” says Investopedia. In other words, in capital-intensive telecom, EBITDA is inaccurate as a measure of operator performance.

That is why “generally accepted accounting principles, or GAAP, do not include EBITDA as a profitability measure, and EBITDA loses explanatory value by omitting important expenses,” the site says.

Revenue is important, but the more-telling shift is the use of “EBITDA” rather than “profit.” EBITDA is a measure of cash flow, or operating efficiency. That is important, but it is not the same thing as “profit,” commonly understood.

To be sure, there are reasons for using EBITDA: it isolates operating performance, without the potential distortions of financing activities. At the same time, positive cash flow does not necessarily mean a firm is “profitable.”

That is not always a long-term problem. The entire U.S. cable TV industry, in its major urban growth phase, always had high cash flow (EBITDA), but zero profits, as all available cash was plowed back into growth. Eventually, when the construction phase ended, cable operators shifted to actual GAAP “profits.” So cash flow matters. A lot.

The problem for the global telecom industry is that while there is growth, that growth is heavily to be found in Asia and Africa. In most other regions, the business is quite mature or getting that way. Where there is not high growth, use of EBITDA arguably masks some structural issues.

Mergers, acquisitions, new products and operating cost reductions are reflections of the underlying trends. In one clear sense, telecom clearly is a declining industry: all its legacy products are in a mature mode, while some have been declining for more than 15 years (international long distance, voice).

To survive, much less prosper, completely new products, at massive scale, will be needed. That is why “internet of things” and “connected cars” are so important: they hold the promise of supplying those big new revenue sources.

But we have to recognize that the shift to “EBITDA” instead of GAAP “profit” tells you something very fundamental about the business.

source: Ericsson

Mobile is Least-Favored Retail Payment Method, Survey Finds

Retail mobile payments have developed rather more slowly than many have predicted would be the case. The basic objection has been that switching from existing retail payment methods to mobile payment adds too little value to drive rapid switching. That seemingly remains the case.

Security and privacy remain the top stated consumer concerns, though an argument might be made that the real reason is that the innovation simply does not yet add enough value to be worth the “bother.”

When asked about which forms of payment they find most secure, cash topped the list of responses to a survey conducted by Walker Sands. Some 46 percent indicated that was a concern.  Credit cards were cited by 27 percent as “most secure,” while debit cards were seen as most secure by 22 percent of respondents.

Mobile payments ranked last each of the past two years, at one percent, in terms of “preferred method of payment.”

The majority of consumers cite security (61 percent) and privacy (58 percent) as the two primary factors that make them hesitant to use mobile payment applications.

About14 percent of consumers say they have no hesitation to use mobile payment services. There are other issues, such the disparity of retailer terminals able to accept one or more mobile payments systems.

For some of us, those are real, but secondary issues. The key problem remains that value is not high enough to prompt massive behavior change.

source: Walker Sands

OTT ARPU Might Matter More than Number of Accounts

source: Chetan Sharma

The decline of the U.S. linear subscription TV revenue might happen slower than you think, for a number of reasons. Between 2016 and 2019, for example, though the number of over-the-top subscriptions will grow--boosting the number of active accounts by perhaps 12.7 million units, according to eMarketer--revenue will not change as quickly.

To be sure, OTT account volume already has surpassed linear account volume. U.S. linear subscriptions in 2016 represent about 93.8 million subscriptions. Assuming linear accounts do not grow, or decline too fast, that might mean OTT, already the share leader in terms of total accounts (93.8 million 2016 linear accounts and 187 million OTT accounts), will represent as much as 68 percent of all video entertainment subscriptions.

But average revenue per account is highly disparate. Where a U.S. linear around drives between $80 and $120 a month in revenue, an OTT subscription might drive $7 to $11 per account, per month. In other words, linear accounts can represent as much as an order of magnitude more revenue (10 times) as an OTT subscription.

In 2016, U.S. OTT video revenues might have represented $7 billion. Linear video produced $102 billion.

Were OTT subscription revenues (advertising or transaction revenues also will be generated) to grow 10 times, that would still directly displace a bit less than the current value of linear subscription revenue.  

source: Strategy Analytics

One way of looking at the market is that a customer contemplating purchase of an entertainment video service can buy a single linear subscription, or perhaps 10 OTT accounts, and spend the same amount of money per month.

If one assumes that over-the-air broadcast TV can be gotten using an off-air antenna, for no monthly subscription price, and pre-recorded material from OTT, then the third bucket of content is “live sports” and, for a small number of customers, 24-hour news.

So the issue, long term, is which bundles of purchases allow a potential consumer to replicate all--or enough--of the wanted content using OTT sources only (and assuming over-the-air TV can be gotten at no incremental cost), and what that implies for total monthly spending.

Linear services are adding more features (including OTT or on-demand access), so the issue is not a simple “linear versus on-demand” dynamic.” Also, more linear content is being packaged in OTT form. And few customers want “all” the programming currently offered in big linear bundles. Relatively few want any of the highly-specialized channels, many do not want sports, most do not want news and off-air is an alternative for traditional broadcast network fare.

Logic suggests consumers will evaluate value against price, suggesting a workable set of OTT replacements, for most consumers, will offer “most” of what a linear subscription features, at lower price.

That implies lower revenue per account, and likely lower total monthly spending per user or per household. The reason is what happens in a typical product lifecycle. At some point, every popular product saturates, and revenue per account, as well as industry revenue, falls. Some might argue that OTT is a “new” industry with a “new” growth curve. Others might see OTT as the replacement product for linear delivery.

Both points of view can be correct, simultaneously. One other observation likely is germane. The internet tends to create product replacements that produce less revenue than the legacy business model. The phrase “trading digital dimes for analog dollars” illustrates the notion.

Since most of the primary underlying cost of content is driven directly by content rights access (perhaps 40 percent of total cost in the linear model), and since owners will act to protect those revenue streams, we might surmise that content costs are unlikely to fall much, in the transition from linear to OTT delivery.

source: eMarketer