Build it and They Will Come

Like it or not, our history with 4G networks, which was in many respects a “build it and they will come” exercise in hope, will characterize 5G as well. Few claim to be sure precisely where new revenues will be created, though as a generic, most believe internet of things and other ultra-low latency use cases will develop, at scale.

In the near term, the more-prosaic use cases will be capacity to support consumer mobile broadband. The near-term new use case is 5G fixed wireless. Ever since 3G, observers have expected creation of many new use cases killer apps and revenue streams, but such use cases have had to be discovered and created.

It would not be unfair to characterize the actual use cases as somewhat unexpected. 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. But the actual use cases often are unexpected.

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. So “build it and they will come,” as discredited as that might be, seems to be what happens when next-generation mobile networks are deployed.

“Build it and they will come” became a discredited phrase in the collapse of the telecom and internet bubble that destroyed trillions in equity value. Over a two-year period between 2000 and 2002, 500,000 jobs were lost in the U.S. market alone.

source: Morgan Stanley

The Dow Jones communication technology index dropped 86 percent; the wireless communications index, 89 percent.

Some $2 trillion worth of equity value in telecom companies vanished. Scores of firms went bankrupt in the 2000 to 2002 period alone.

One clear problem was investment far in advance of actual commercial market demand, despite the correctness of the long-term vision of capacity growth fueled by use of the internet. Simply, between 1999 and 2002 too much new capacity was added to the market.

Many who lived through that period developed a healthy respect for skepticism about any “build it and they will come” investment plans. Clearly, that approach failed during the height of the telecom bubble.

source: Science Direct

And still, “build it and they will come” was largely a reality for 3G and 4G and remains so for 5G. The problem always is that we cannot predict which big new use cases and revenue will develop.

No prudent executive will claim to be staking the future on investments whose revenue models are uncertain. Yet that is precisely what has happened, to a large extent, since 3G. Sure, “faster speed” is an easy justification. Beyond that, nobody has been terribly good at predicting how that will enable new apps, use cases and revenue.

Video--With All its Issues--is the Biggest Consumer Market Opportunity for Fixed Network Telcos

Fixed network telcos face extreme threats to their core business models as sales of legacy products diminish and as average revenue per unit sold drops, even as usage skyrockets. But there is some disagreement about the wisdom of telcos getting into the video entertainment business, and in what roles.  

Consider only the role of video distributor. One conclusion many observers reach about linear video services is that the product is declining enough that it is not a fruitful area for telcos to pursue.

Linear video in the U.S. market is past its peak, to be sure, even as streaming alternatives proliferate. Most believe the revenue implications are clear enough: linear services with higher average revenue per unit (ARPU) will be displaced by streaming services with lower ARPU. But many, perhaps most streaming customers will buy more than one service. So, ultimate revenue impact is not as clear as one might suppose, if there were a one-for-one substitution of streaming for linear accounts.

source: Strategy Analytics

Others might point to Comcast’s strategy, which has been to diversify away from linear video distribution since 2008, as providing further evidence of limited potential. But that is a matter of incumbent firm strategy.

In the competitive era, telcos have taken video share as cable companies have taken internet access and voice share. The point is that what is rational for Comcast is not rational for AT&T or Verizon.

source: nScreenMedia

But there are several reasons why it makes total sense for fixed network telecom firms to  invest in linear video, even as it declines. For starters, there are very few mass market services in high demand by consumers and requiring network services. There is voice, there is internet access and there is video. All other sources remain niches, and mostly small niches.

All existing “at scale” telco products (voice, messaging, internet access, most business services) are flat to declining. Also, mobility is the biggest revenue driver for many service providers. The corollary is that the entire fixed networks business is smaller, and, in many cases, shrinking rather than growing.

So much sales volume now is in the mobile domain that the total fixed network business often is quite small, compared to mobility. That is true for AT&T, SK Telecom, Elisa and Swisscom, for example.

AT&T now earns close to 70 percent of total units sold from mobility (business and consumer), about 18 percent from video about five percent from fixed network voice and less than 10 percent from voice services.

The gross revenue and profit percentages do not match completely, but the big point is that for AT&T, fixed network voice and internet access are too small to make a difference in overall revenue and especially growth.

The other important point is that the video distribution business, on AT&T’s fixed network, already is bigger than voice and internet access put together.

source: Huawei, IHS Markit

Recent moves by AT&T might suggest what is possible. By units sold, AT&T product sales in the consumer market are more than 50 percent of total. Internet access represents less than 25 percent of units sold, while voice produces more than 25 percent of products sold.

Other major telcos have different sales profiles, but many leading telcos earn significant revenue from video distribution services, rivaling in many cases revenue earned from providing internet access services.

source: Huawei, IHS Markit

As a practical matter, there are few brand new revenue sources big enough to make a difference on fixed network service provider financial statements, where it comes to consumer services.  

How Much Share of the Enterprise Edge IT Market Can Service Providers Take?

Market share in the software-defined wide area networking market is difficult to pinpoint, as it combines sales both of networking hardware and carrier services.

SD-WAN hardware includes appliances and routers, SD-WAN software that includes orchestrators, gateways, cloud routers and firewalls, dashboards, management systems among others, and SD-WAN services that includes service provider managed SD-WAN services.

Beyond that, early estimates vary by nearly an order of magnitude. Observers might debate the importance of SD-WAN as a service provider product, but the strategic upside comes in two ways.

First, it is possible that SD-WAN, as a way of connecting branch offices, becomes a growth market in its own right, displacing other solutions enterprises have used. So it is possible that carriers can take much market share from appliance and software suppliers.

source: Gartner

The longer-term possibility is that SD-WAN becomes the enabler of other value and revenue service providers can supply. 

So far, as most of the SD-WAN activity has been appliance-based and installed by enterprises or other end user organizations, most of the sales activity has been garnered by edge equipment suppliers, not service providers.

source: Gartner

The expectation is that, over time, more share will be taken by service provider offers (SD-WAN as a service, if you like that formulation). That belief is bolstered by the likely use of SD-WAN as a replacement for other carrier services.

On the other hand, some might argue that since SD-WAN increasingly bundles the functions of encryption, path control, overlay networks and subscription-based pricing together, the edge equipment approach to SD-WAN in many instances will be a substitute for other purchases enterprises have made in the past (routers, security appliances and software, special access services).

source: Juniper Networks  

In a broad sense, the service provider opportunity likely will involve building on SD-WAN connectivity to supply additional features and services.

source: IDC

The issue, perhaps, is how much of the value of edge equipment market overall service providers can replicate “as a service.”

source: Open Networking Summit

FTTH or Time Warner? It Is Not a Close Call

Would AT&T have generated more incremental revenue if it had not bought Time Warner, and instead had plowed that capital into a massive fiber to home upgrade?

The numbers suggest AT&T made a better choice buying Time Warner.

AT&T spent $85 billion to acquire Time Warner, with an immediate quarterly revenue boost of $8.2 billion. Were AT&T able to invest in fiber to home and then take an incremental five percent share of market everywhere it operates, is perhaps $2.2 billion in annual revenues, assuming $50 a month in gross revenue, or about $180 million a month in incremental revenue.

It is not clear how much upside exists for AT&T, in terms of fixed network internet access revenue, even if it were to dramatically extend its FTTH footprint, but you might argue that the best case for AT&T, for a massive upgrade of its consumer access network, is about 10 percent upside in terms of consumer market share, facing cable operators already leading the market in accounts and speed, with a clear road map for additional speed increases that easily match anything AT&T might propose, and arguably at less cost.

So here’s one take on the alternatives of buying Time Warner or using that capital instead to expand the AT&T FTTH profile. Consider the incremental revenue generated from each alternative.

Assume first that U.S. telcos could take 10 percent more market share from cable TV suppliers. Incremental revenue might then be less than $4.4 billion annually. Consider that AT&T has footprint covering perhaps 69 percent of U.S. homes. So make the incremental revenue for AT&T $3 billion, or $250 million per month.

Also, it would take some years before that degree of new FTTH assets could be put into place. Over any three-month period, AT&T might expect incremental revenue ranging from $540 million to $750 million per quarter, the former figure representing five percent share gain, the latter representing 10 percent share gain.

Neither comes close to the $8.2 billion per quarter AT&T picked up from the Time Warner acquisition.

Verizon has different strategic issues, compared to its main fixed network competitors.

Significantly, Verizon has a small geographic footprint, compared to any of its main fixed network competitors. Verizon homes passed might number 27 million. Comcast has (can actually sell service to ) about 57 million homes passed. Charter Communications has some 50 million homes passed.

AT&T’s fixed network represents perhaps 62 million U.S. homes passed.

Assume there are 138.6 million U.S. housing units, of which perhap 92 percent are occupied (including roughly seven to eight percent of rental units and two percent of homes). That suggests a potential base of 128 million housing units, including rooms rented in homes or apartments, that could buy services from a fixed network supplier.

That implies Verizon has the ability to sell to about 21 percent of homes; Comcast can sell to 45 percent; AT&T can market to 48 percent of occupied homes; while Charter can sell to 39 percent of U.S. occupied homes.

The point is that Verizon has more to gain than AT&T, Comcast or Charter from investing in internet access outside its traditional geography.

In principle, Verizon faces the same issue as does AT&T when weighing alternative uses of scarce capital.

As it deploys 5G fixed wireless, there are two key issues: how much market share and revenue can Verizon gain, and what else might Verizon have done with its investment capital? It all depends on one’s assumptions.

Some argue that, over seven years, Verizon might gain only 11 percent to 18 percent share in markets where it can sell 5G fixed wireless. Verizon believes it will do better, and some believe a 20-percent share is feasible. Verizon itself predicts it can get about 23 percent share, as a minimum, over seven years, representing about 6.3 million accounts.

Assume Verizon fixed wireless gross revenue is about $60 per account (a blend of the $50 from Verizon mobile customers and $70 from non-customers). Assume annual revenue of perhaps $720.

Assume Verizon spends about $800 per location on 5G fixed wireless infrastructure (radios, backhaul, spectrum costs), even if those same assets can be used to support other users and applications.

At 20 percent take rates, that implies a per-subscriber network cost of perhaps $4000.

Assume a cost of perhaps $300, over time,  to turn up service to accounts. That implies a rough break even in months. Assume total capex investment of perhaps $4300 per account. At $720 annual revenue, that implies breakeven on invested capital in six years.

But assume half the cost of the capital investment also supports revenue generation from other users and use cases (mobility, business users, internet of things). In that case the fixed wireless capex is perhaps $$2150 per customer, and breakeven on capex is a bit more than three years, assuming the only revenue upside is internet access revenue.

Logically, one would have to add churn reduction in some cases, and so the lifetime value of a customer; incremental advertising opportunities; some possible upside from voice services or wholesale revenue. None of that is easy to quantify with precision.

The point is that potential return might fall well within a framework of payback in three years.

Whether that is a “good” investment or not depends on what else might have been generated from other capital deployments.

Over a seven-year period, Verizon might have committed $13 billion in capex to generate revenue from six million fixed wireless accounts (about $1.85 billion per year). It is hard to image any alternative use of capital at that level that would result in annual revenues of $4.3 billion in internet access revenues alone.

It is in fact quite hard to create a brand new business generating as much as $1 billion a year in incremental revenues, under the best of conditions.

source: ITU

So, back to the importance of video revenues, as difficult as the Time Warner debt burden might be, the renamed Warner Media already generates $32 billion in annual incremental new revenue for AT&T. Virtually nobody other than its competitors is likely happy about the new $55 billion worth of new debt AT&T has acquired.

Still, the issue is what else AT&T could have done with $55 billion that would immediately create $32 billion in new revenues. Personally, I cannot think of another transaction that would have produced that much new revenue, immediately.

AT&T could have spent that money on fiber to home upgrades, to perhaps gain five percent to 10 percent additional market share in the consumer internet access market, in region, over perhaps five to seven years. The upside, even at 10 percent share gain, does not approach the Warner Media contribution.

Global Telecom Revenue Flat (Not Adjusting for Inflation) Through 2022

On a non-inflation-adjusted basis, global telecom services revenue will grow at a compound annual growth rate (CAGR) of 0.8 percent, IDC researchers now predict. In any market with inflation rates of at least 0.8 percent, actual revenue will decline.

Product segments within the industry can have faster or slower growth rates. In fact, revenue earned in the fixed network segment generated by data services will grow 22 percent in 2018 and at a four-percent CAGR through 2022, IDC predicts, driven by uptake of internet access services.

Mobile revenue will grow at 1.2 percent through 2022, but fixed network voice revenues will decline at five percent annually to 2022.

"Developed and mature markets will only show marginal gains now, driven by technology migration and bandwidth needs," said Eric Owen, IDC group vice president, EMEA Telecommunications & Networking. South Asia and Africa are the two regions that will see the fastest revenue growth, IDC predicts.

source: IDC

Watch What Telecom Execs Do, Not What They Say

How will U.S. service providers grow revenues over the next three years? A clue: watch what company executives do, and what they have done, not what they say.

Mergers and acquisitions, partnerships and organic growth is what U.S. executives tell KPMG will drive revenue growth.  KPMG’s 2018 U.S. Telecom CEO Outlook provides insights from 82 telecom industry CEOs in the United States on their expectations for revenue growth.

But with organic growth so slow in telecom, neither organic growth nor growth generated by channel partners is going to move the revenue needle too much.

source: KPMG

source: KPMG

Consider only the matter of price increases between 2011 and 2017. The consumer price index has grown slowly over that period, up about nine percent. Over the same period, telecom business service provider prices have not grown at all, declining a bit less than one percent.

source: USTelecom

One reason for that flatness in business services is product substitution. Though the products might not always be full substitutes, best effort cable modem service is much more affordable than other business data access alternatives. Where a T1, supporting 1.544 Mbps, might cost $145 a month, a 100 Mbps to 200 Mbps cable service might cost only $70 a month.

Where a 45-Mbps special access connection might cost  $720 a month, a 300 Mbps to 500 Mbps cable access service might cost $100 a month.

source: USTelecom

So, in practice, mergers and acquisitions are likely to lead any measurable amount of revenue growth for most companies,  no matter what respondents say, simply because it is the fastest way to make a meaningful pivot in business strategy or fill a strategic gap.

Organic growth simply is too low to make a meaningful difference in growth rates. In North America, according to recent work by Strategy &, between 2010 and 2016 the bulk of revenue growth recorded by service providers globally came from horizontal acquisitions.

In other words, firms acquired similar assets in the same lines of business from other telecom firms. That noted, there has been somewhat more acquisition of vertical assets since about 2012.

source: Strategy &

source: Strategy &

Someday, "Connectivity" Will Not be Your Value Proposition

With the caveat that any attempt at predicting the future is inherently hazardous, Nokia Bell Laboratories is creating a Future X Lab to illustrate its thinking about how networks will supply value in coming decades.  

That effort supports a vision of the future centered on the FutureX network that dramatically reshapes understanding of the purpose and function of a next generation network that will have moved beyond connectivity as its source of value.

“Creating time” is one way of illustrating the difference between “we connect you” and “we create time” as the core value proposition. By “creating time,” Bell Labs means greater productivity, in both personal and work spheres.

source: Nokia Bell Labs  

In this illustration, note the base of the “analog needs” pyramid: “free Wi-Fi.” That illustrates the connectivity conundrum, which is that revenue per bit keeps falling, while value shifts to other areas higher on the stack.

In the coming era, the value proposition will have to be recreated, in ways that transcend “connectivity.” And it has to be said, such proposals will seem ethereal and “academic” for people who work in commercial enterprises.

There are other obvious issues. Few people, few executives and few firms will be able to invest sufficiently in efforts to create solutions for “new needs.” It is commonplace in the tier-one portion of the business to note that new revenue streams smaller than $1 billion cannot get investment because the financial return is too small to “move the needle” on overall revenues.

Smaller firms, in niche areas of the market, are even more constrained, as there generally is little free capital or cash flow to invest in “broader” initiatives not central to the core business model.

For reasons of scale, the “normal” or “typical” way tier-one firms grow is by acquisition. That is unlikely to change, no matter how great the imperatives for industry transition.

On the other hand, observers often decry the “cultural” issues that “prevent” firms from changing. That, too, is likely correct, but largely irrelevant. If the strategic direction is to reinvent the whole business by participation at higher levels within the value ecosystem, then it is almost pointless to worry about changing culture in the legacy areas.

The people who work in the new areas will run those parts of the business. In other words, efforts to “change culture” at the lower levels of the business stack probably are largely wasted effort.

If and when service providers become relevant at supplying solutions at higher levels of the value stack, different people--with different skills required for those parts of the business--will be supplying those solutions.

The point is that connectivity providers eventually will find new roles (larger or smaller) within the value ecosystem. Many firms will lack scale to do much other than supply local connectivity, and will be acquired or sold. Firms with scale will gain scale, but vertically within the stack, rather than simply horizontally.

If Bell Labs is correct, the key value proposition for the “communications” business will shift from “we connect you” to something else, related to solving big human or business problems. It might seem quite ethereal. Someday it will be seen as eminently practical.