Gigabit Network Investments Might Not Always Return Cost of Capital

Sometimes, service providers make strategic investments not strictly related to "return on invested capital." That might seem irrational. It is not.

Large or small, fixed network telcos continue to face very-tough decisions about high speed access. For several decades, telco planners have modeled financial returns from fiber-to-home projects and generally have faced a hard reality.

In most cases, though such investment often has to be justified for strategic reasons, the financial return often is difficult and tenuous. Verizon Communications, for example, has in the past argued that a significant portion of the return comes not from the ability to offer new services, but from reduced operating expenses.

One might conclude Verizon no longer believes the expected maintenance savings are as great as were originally expected. In 2010, Verizon’s suspended its FiOS program in major metro areas where it had not already begun construction. In 2014, Verizon executives said they would consider expanding FiOS deployments when doing so would recover the cost of capital committed to the effort.

In other words, FiOS probably does not return its cost of capital in many markets.

To be sure, there are other considerations. Verizon and AT&T now drive revenue growth from the mobile business, so returns on invested capital are much higher when available capital is invested in the mobile business.

In the first quarter of 2014, Verizon operating income for the mobile segment was an order of magnitude higher than for the fixed line business.

In the third quarter of 2014, mobile segment operating income margin was 31.9 percent and segment earnings (EBITDA) margin on service revenues was 49.5 percent.

Compare that performance with results from the fixed network segment.

Fixed network  operating income margin was 2.3 percent in the third quarter of 2014, up from 1.5 percent in third-quarter 2013. So mobile operating income margin was an order of magnitude higher than fixed network operating income margin.

Smaller telcos without mobile assets will not have the option of directly available capital to the mobile network.

So the decision about investing in fiber-to-home facilities remains a challenge. Cincinnati Bell, for example, sold its mobile business and half of its data center business to raise capital to build its “Fioptics”  fiber-to-home network.

Analyst Craig Moffett of Bernstein and Company estimated in 2009 that Verizon’s cost per subscriber was about $4,000, while estimating the expected revenue from a FiOS-connected household was just $3,200.

Though network element costs arguably have declined since then, the business case remains challenging. So why do telcos move ahead? There are strategic reasons. Unless a fixed network telco upgrades, it might be unable to compete with cable TV operators.

Simply, the investment in fiber-to-home has to be made so “you get to keep your business,” as one executive said. Strict return on capital considerations are secondary.

Sometimes a business decision has to be made for reasons other than strict return on invested capital grounds. Investment in gigabit networks would appear to be such a decision.

Source: Craig Moffett, Bernstein & Co.

Phone Installment Plans Turn Out to be Revenue Neutral for Mobile Operators

When unsubsidized mobile phone plans first were introduced, there was some concern about the impact on consumer adoption of advanced smartphones, which could have slowed mobile data revenue growth.

In practice,  installment plans seem to operate pretty much as before, the only difference being that mobile service provider cash flow arguably is better when most customers are on installment plans.

Importantly, there has been no significant negative impact on adoption of smartphones and sales of data plans.

That is the Verizon Wireless experience, at any rate.

“From a pure financial perspective, the profitability of a customer is exactly the same under the subsidy model in some models over a two-year period of time,” said Fran Shammo, Verizon CFO. “So there’s no different in the profitability of what that customer generates.”

There is a difference in cash flow and a shift of revenue between accounts. Equipment sales revenue is higher, and recurring service revenue is lower. A device sold on an installment plan might mean the total cost of the device is booked as revenue, even if the actual payments and receipt of cash will e booked over a year, two years to 30 months.

In most cases, though, total revenue from a customer on a traditional contract featuring a subsidized device, and total revenue from a customer on an installment plan, will be about the same, Shammo said.

On an “average revenue per account” basis, including both recurring service and installment plan revenue, Verizon reported five percent revenue growth. Looking only at mobile service revenues, growth was 3.5 percent in the third quarter of 2014.

Cloud Data Centers: How WAN Providers Earn LAN Revenue

The cloud computing business in some cases is remaking the old distinction between “local area network” and “wide area network.”

Traditionally, the business models for LAN and WAN were as different as the physical scope of the networks.

Wide area network services were provided by entities that operate “outside the building.” Local area networks worked “inside the building.”

Businesses and consumers WAN “rent” services but “own” LAN infrastructure (Wi-Fi routers and in-building wiring and servers).

So the ownership modes always have been different. People and organizations own their own LAN infrastructure, and do not pay recurring service fees. Customers pay for WAN services, which are owned by others.

The capital-intensive WAN business requires large organizations with lots of capital to invest in building an operating huge networks. Its revenue model is “recurring access and transport services.”

The LAN ecosystem includes suppliers of consumer electronics appliances, system integrators in the small and mid-sized business segment and consultants and integrators in the enterprise segment.

The shift to cloud computing is in some cases causing WAN providers to blur the lines with the LAN space, at least in the data center “customer” segment. In other words, where LAN operations have not traditionally created revenue opportunities for WAN providers, ownership of data centers now drives traffic to the WAN, and also generates direct real estate revenue.

To be sure, it can be argued that the direct revenue is generated by “hosting” servers in a data center (a real estate transaction), not “moving bits” between customers in a data center. Some might argue ownership of a data center now creates a recurring revenue stream for moving bits within the building (something that formerly would have been a LAN function).

In that sense, at least where it concerns data centers, WAN providers who own data centers might be said to be earning revenue for data communications that might formerly been non-revenue LAN communications.

Global Cloud Xchange has put data centers at the heart of its strategy, even going so far as to change the name of the company from the former Reliance Globalcom.

Tata Communications has done something very similar, with a key nuance.

Like other service providers, such as Verizon, Tata has become an owner and operator of data centers and cloud infrastructure  In that instance, “transport” revenue is earned in a different way, in the form of adding transport to the “real estate” services and cloud infrastructure capabilities.

Global Cloud Xchange does not seem to be buying and operating data centers and colocation facilities that support “meet me” rooms.

Instead, Global Cloud Xchange seems to be architecting its transport network to provide transparency for server-to-server communications, so the long-distance connection acts like a cable connection between servers in the same building, on the same floor.

Tata, for its part, has created IZO Public, a cloud enablement service. Tata recently inked an interconnection agreement with Google, providing business customers a way to connect and build their public cloud services with consistently good user experience.

The IZO platform provides predictable routing and connections to data centers over the Tata Communications global network. In 2014, 24 percent of the world’s Internet routes travel over the Tata Communications network.

Cisco’s latest Global Cloud Index estimates that global data center traffic will grow nearly 300 percent between 2013 and 2018.

Although the amount of global traffic crossing the Internet and IP WAN networks is projected to reach 1.6 ZB per year by 2018,  the amount of annual global data center traffic in 2013 is already estimated to be 3.1 ZB, according to Cisco.

By 2018, “data center to end user traffic” will constitute 17 percent of total. About nine percent of traffic will move from data center to data center. About 75 percent of global data center traffic will stay within the building, moving from server to server.

That explains the high interest by capacity providers in data centers. In the past, most of the revenue made by wide area network providers was supplying capacity across the wide area network.

In the future, it is likely much revenue will be made supporting data communications between servers and entities within data centers, and some of that will fall to WAN providers that own data centers.

ISPs Represent 30% of Total U.S. Domestic Capex by the Top 25 Firms

Leading telcos, cable companies and application providers were clearly among the U.S. firms making the biggest 2013 domestic capital investments, according to the Progressive Policy Institute.

A study of the top 25 firms with the highest domestic capital investment shows 2013 capital investments of $152.5 billion. Of the top 25 firms, number one was AT&T, which invested $21 billion. Number two was Verizon, which invested $15.4 billion.

Intel ranked seventh, committing $8.4 billion to domestic capex.

Comcast was seventh, investing $6.6 billion. Google was 12th, spending $4.7 billion. Apple ranked 15th, investing $3.8 billion.

Time Warner Cable was ranked 21, spending $3.2 billion in 2013. Microsoft, ranked 23rd out of 25, made an outlay of $3 billion. Amazon, at 25th position, invested $2.6 billion.

So when five of the top 25 say a new proposed change in regulation will cause a slowdown in domestic capital investment, that is a non-trivial matter, as would be the case if Google, Apple Intel and Microsoft were to argue that a proposed regulatory change would choke off their investment.

In fact, AT&T, Verizon, Comcast and Time Warner represent 30 percent of 2013 domestic capital investment made by the top 25 firms, investing $46.2 billion.

The software firms--Google, Amazon, Microsoft--invested $10.3 billion, or nearly seven percent of domestic capex by the top 25 U.S. firms.

Apple and Intel invested $12.2 billion, representing eight percent of domestic capex by the top 25 firms.

One might simply note it is a big deal if new regulations slow down investment by the ISPs that represent 30 percent of total capex by the top 25 U.S. firms.

Over a three-year period, AT&T ranked first of 10, investing $60.5 billion. Verizon was second at $46.6 billion. Intel ranked sixth at $24.6 billion. Comcast was ninth at $17.6 billion. No app or device suppliers made the list of the top 10.

"Pay for Priority" Already is a Widespread Reality of the Consumer Website Experience

Fully 81 percent of consumers surveyed by the University of Delaware Center for Political Communications say they oppose “allowing Internet service providers to charge some websites or streaming video services extra for faster speeds.”

In one sense, the responses are not surprising. At some level, a rational person would likely conclude that if any ISP can charge a website for expedited delivery, that has to be reflected in higher costs to use that website, over time.

Also, past surveys universally have found that consumers dislike advertising embedded into their TV experiences. But when asked whether they would rather pay more, to avoid the commercials, people routinely decline, indicating an acceptance of the irritation of ads if it also means "free" content access.

Something like that undoubtedly is at work here.

Incidentally, one can generate equally unhelpful results asking people whether the Internet should be regulated like television and radio. People say they do not like that either, and that is how one survey probed for attitudes about network neutrality.

One suspects fairly comparable results might therefore have been gotten if the question something like “do you think a network should be able to charge a website or streaming video service extra for faster speeds.” It is logically the same scenario: the website spends money to a transport services provider to gain faster delivery of its packets.

Consumers might rightly guess that would somehow be reflected in higher costs to use the websites that do so.

Consumers cannot be expected to know that just one content delivery network of many (Akamai) provides exactly those services to 33 percent of “Global 500” companies as defined by Fortune magazine, including the largest 30 media and entertainment companies.

Akamai also is paid money for expedited packet delivery by “all 20 top global e-commerce sites,” according to Akamai.

Fully 97 of the top 100 online U.S. retailers also use Akamai to speed up performance.

In addition, Akamai is use dbgy more than 150 of the world's leading news portals, nine of the top 10 U.S banks, seven of the top 10 world banks, eight of the top 10 U.S online brokers, nine of the top 10 largest newspapers, eight of the top 10 online publishers, and nine out of 10 top social media sites, to name just some of the industry segments and leading firms that pay money to Akamai to expedite delivery of their packets.

If such payments are reflected someplace in each of the paying company business models (it has to be), then “higher prices” already are reflected in the websites and services. The upside, of course, is better user experience when interacting with the sites.

The point is that most respondents, if rightly guessing “my costs” have to be higher, also do not understand that such costs already are widely embedded in leading website cost structures, and that better user experience is the benefit such costs enable.

One guesses most respondents also would object to some new policy that slows down packet delivery or speeds. But that would also mean they object to content delivery networks such as Akamai, even if nobody knows payments for expedited delivery are any everyday occurrence for the most-popular websites, and that, yes, those costs are reflected in the cost of the delivered products.

The point is that popular understanding of a complicated subject such as “network neutrality” is further complicated because people do not understand the ways content delivery networks, though representing some incremental cost, also provide experience benefits that app providers widely believe are worth the price.

Consumers who use those sites are benefitting, and also are paying for the expedited delivery. It isn’t much of a problem, if it is a problem at all.

Why Cloud Data Centers Drive WAN Provider Revenue

source: Cisco

Traffic patterns are rapidly shifting from the traditional north-south client server model to an east-west server to server, said Jayshree Ullal, Arista president and CEO.

What that implies is a shift of data traffic. “We found that most of the server generated traffic in

the cloud data centers stays within a rack, while the opposite is truefor campus data centers,” say Theophilus Benson and Aditya Akella of the University of Wisconsin–Madison and David A. Maltz of Microsoft Research–Redmond.

And make no mistake, traffic generated by cloud computing increasingly dominates the pattern of global wide area network, metro and data center traffic. Data center traffic, as measured by Cisco,  includes data center-to-user traffic along with data center-to-data center traffic and traffic that remains within data centers.

Cisco’s latest Global Cloud Index estimates that global data center traffic will grow nearly 300 percent between 2013 and 2018.

By 2018, 76 percent of all data center traffic will come from the cloud, while 75 percent of data center workloads will be processed in the cloud.

But that might not even be the most significant prediction. Quantitatively, the impact of cloud computing on data center traffic is clear, Cisco argues.

Most Internet traffic has originated or terminated in a data center since 2008.

Where in the past most traffic (voice) functionally originated and terminate at a central office, though that traffic was passively transmitted to an end user telephone, now most global traffic originates and terminates at a data center.

And Cisco means that literally. Most traffic that ultimately traverses the WAN will follow communication within a data center between co-located servers.

The whole rest of the communications network essentially becomes a passive connection between an edge device requesting some operation, and the server that processes and serves up the result.

So in a meaningful sense, data centers are the central offices of the new network. As “switching” in the telco or mobile central offices might have been likened to the “brain” of voice operations, with the rest of the network the nervous system, so now data centers are the brains of the global network.

That will have any number of ramifications. The architecture of a voice network was build on the switching offices. Even the location of specific central offices was built on the amount of signal loss through a pair of copper wires connecting central offices with end user locations.

The architecture of new global networks is built on co-located servers inside data centers, and then connections between data centers over the wide area network, often with more traffic crossing the metro area than transiting the WAN, and then only traffic moving between a metro area data center and an end user location.

Although the amount of global traffic crossing the Internet and IP WAN networks is projected to reach 1.6 ZB per year by 2018,  the amount of annual global data center traffic in 2013 is already estimated to be 3.1 ZB.

The implications could not be clearer. By 2018, data center to end user traffic will constitute 17 percent of total. About nine percent of traffic will move from data center to data center. About 75 percent of global data center traffic will stay within the building, moving from server to server.

That explains the high interest by capacity providers in data centers. In the past, most of the revenue made by wide area network providers was supplying capacity across the wide area network.

In the future, it is likely much wide area network provider revenue will be made supporting data communications between servers and entities within data centers.

In other words, cloud data centers now are a major driver of WAN provider revenues. As potential owners of cloud data centers, WAN providers earn revenue by enabling data communications within the data centers, while also positioning to capture data center to data center revenues.

Network Neutrality Should Not be a "Clever Hack"

Sometimes it takes a brave observer to say the “emperor has no clothes.” You might say some witty British technology observers have done so.

“Almost 20 years ago Congress decided that its regulators couldn't apply old Bell-style phone rules to Internet services,” says Andrew Orlowski, a writer for The Register. “The judicial system is getting fed up with being asked to throw out this distinction: it's not the courts' job to make new laws, or repeal old ones.”

While a couple of times the lawfulness of common carrier regulation of Internet access has been challenged--successfully--in court, some continue to advocate that the Federal Communications Commission do something many argue it has no authority from Congress to do.

“The FCC is the creature of Congress, not its master,” notes Orlowski. There are some other avenues network neutrality supporters could support, that would come under jurisdiction courts have suggested already exist (promoting broadband adoption).

But that is not the remedy some now propose.

Orlowski points out what might be obvious. Zeal for some ends now seems to justify nearly any ends.

Disparaging what he characterizes as an effort to pull off a “clever hack,” Orlowski points out the larger problem of which efforts by some are only an example.

“Why not bypass it (democracy) wherever and whenever you encounter something you don't like?” he rhetorically asks.

The matter is not good intentions or ideas about good policy. Most network neutrality supporters honestly and sincerely believe Internet freedom is served by those policies. 

Perhaps all who oppose net neutrality object only to a very few specific notions about network neutrality. 

Even oppoinents of net neutrality agree that end users have the right to use any lawful application, They agree that no Internet service provider should be able to block or degrade the performance of any lawful application.

The single salient objection is that voluntary or optional use of content delivery networks by end user customers, and the ability of an ISP to offer such services, should be lawful, not unlawful as network neutrality supporters propose.