Moral Outrage Over "Loosely Coupled Networks" is Misplaced

Lots of start-ups find they actually change revenue models models on the way to building a sustainable business, compared to what they originally were funded to undertake. But there are lots of more-subtle ways businesses wind up providing value in the Internet ecosystem, for end users and other business partners, even when nothing “Internet” is actually directly related to their actual revenue streams.

These days, it is common for Internet service providers, for example, to lament the way value, revenue and equity valuations are created by successful application providers who do not have a formal and direct relationship with access providers.

That leads to ISP efforts to create such business and therefore revenue relationships between some popular application providers. One might say there is almost a sense of moral outrage that in a loosely-coupled ecosystem, companies are able to build big, valuable Internet-based businesses without necessarily having a direct business relationship with any access provider.

One might well argue that a healthy ISP business, broadly defined, is in the consumer interest, the public interest or  the national interest. One might well debate various ways to ensure that this outcome is achieved.

But some of us might argue that the sense of moral outrage is misplaced. One might argue that telecom service providers would not have preferred the loosely-coupled “Internet” as a major communications architecture able to rival their own “closed” and tightly-coupled networks.

One might argue about the degree to which telecom organizations and interests, as opposed to “Internet” organizations and interests, “created” the Internet. But it is hard to argue that telecom interests did not help create the protocols and networks, or that global service providers have not selected Internet Protocol as the foundation of their next generation networks.

That is not to say that all IP networks are “the public Internet,” or that all business models using IP are equivalent. They are not.

But the moral outrage about loosely-coupled networks is more than a bit wrong. Everyone now agrees that this is the way software gets written and that this is the way modern networks operate. The actual ownership of applications and services will vary (some tightly coupled, but most only loosely coupled).

But loose coupling (“over the top” apps) is the way we all have decided modern networks will work. That is not to say, as some once did around the turn of the century, that “bandwidth wants to be free.” That is not to argue the importance of maintaining viable access networks, or the legitimate challenges that have to be faced as massive changes occur in ISP revenue sources.

But the moral outrage really is misplaced. The layered model, by definition, presupposes separation of the application and other layers, including physical and transport layers. It therefore is not surprising at all that new revenue models and business categories now exist, and that such businesses exist without the “permission” of participants working at other layers.

The emergence of loosely coupled networks is profoundly disturbing for legacy access providers, to be sure, even if all service providers now accept such models as the foundation of their own next generation networks.

But any sense of moral outrage or entitlement is wrong and misplaced. Without question, we will need viable and profitable ISPs to support the Internet ecosystem. And video applications do pose issues for ISPs that are very complicated and challenging in a loosely coupled framework.

But that’s the nature of the networks we all have chosen to build and use. In that sense, and without diminishing the legitimate need for strong, financially viable ISP businesses, moral outrage probably is not helpful.

This is the communications world we all have chosen to live in, and some would argue it is a good model. The magnitude of transition issues for access providers should not be underestimated. But in a loosely coupled world, application providers create value, they do not steal it.

Is 3rd Place the Best any Smart Phone Provider Now Can Hope For?

Whether aiming for third place in any competition is a good thing, or a bad thing, depends on perspective. For the acknowledged "best" competitor in any endeavor, third place is a defeat. For an up and coming new competitor that never has won at that level, third is a big win.

In the smart phone business, it appears that "third" place in sales volume or market share is about the best any contestant other than Apple or Samsung now can aspire to. That is not to say that state of affairs is permanent. 

But it might now be fair to say many observers seriously doubt any of the "other" contenders have a realistic shot at anything other than third place. That might not be such a "bad" thing for lots of entrepreneurs, though. 

With a different cost structure, niche markets, specialized products or method of delivery, plus retail price, lots of competitors "too small to matter" can make a living in many businesses. That  has been true in the communications business for a few decades, at least. 

In other words, if you want to be a whale, only a few can succeed. If being something else works, lots of space exists in most communication markets. The scale will be different. So will the gross revenue and profit margin. But those niches always exist.

Specialists serving small business segments, premises-based products such as business phone systems, repair, refurbishing, language populations, migrants, prepaid and other niches provide examples. 

That is not to say the niches are permanently defensible. If the biggest providers decide they need to be in the businesses specialists occupy, and if the "whales" can figure out a way to sell at a profit (one reason whales do not pursue some lines of business or customer segments is that they cannot do so profitably), then other contestants can find themselves squeezed out of the business. 

Each business is different, but the "rule of three" process is likely at work in most parts of the communications business. By that rule of thumb, one should expect to see only three leading contestants in any market. 

Some might also suggest that in most markets, market share is unevenly shared by those three competitors. It would not be unusual to expect the share of the lead contestant to be twice that of the number two provider, and for the share held by the number two provider to be twice that of the number three provider, as a general rule. 

For contestants in the tier one part of the access provider business, and the smart phone business, the rule of three will be an uncomfortable reality for many. but that isn't the game most entities in the communications business are playing, in any case. 

Pandora on Mobile Illustrates Service Provider Bandwidth Paradox

Pandora says it is introducing a 40-hour-per-month limit on free mobile listening for its users, something that Pandora says will "affect less than four percent" of its total monthly active listeners. The average listener spends approximately 20 hours listening to Pandora across all devices in any given month.

The direct issue for Pandora is licensing fees. The key issue for access providers is more complicated. By definition, "interesting and valuable apps" are the reason people want to use the Internet. So apps create the demand for Internet access services, especially broadband access.

But with the advent of video as the dominant media type affecting global Internet transmission requirements, access revenue and profit margin are an obvious key issue for access providers, who argue they are not being compensated properly for the value of their access and transmission facilities. 

It might be correct to say that access providers worry they will not be fairly compensated in the future. By most estimates, tier one access providers are making healthy profit margins on access services. Smaller providers have a much-bigger problem. 

The bigger issue really seems to be application revenue, particularly the issue of whether application providers, especially those providing lots of video, might in the future also pay some sort of "access fee." 

That would be a major switch from the traditional one-sided business model where retail end users paid for the full price for use of the network. In the proposed two-sided model, access providers would be paid  both by end users and third party business partners, as is the case in much of the media and content business (cable TV subscriptions, for example, where revenue comes largely from end user "access" with significant "program network carriage fees" and some "advertising" revenue as well). 

The paradox for an access provider is that Internet apps both create the business opportunity and represent a major cost driver. That obvious tension might not, by itself, be too big a challenge. The bigger problem really is that legacy revenues that underpin the business are going away. 

In that sense, it is not so much that Pandora or YouTube are breaking the business model, but rather than the shrinking voice and messaging businesses are forcing service providers to recover most of their costs from Internet access services. 

In that sense, it is not the "Internet apps" that are the problem. It is the declining revenue from other major sources. 

To be sure, some might say the additional problem is that the past value chain is being disrupted. In the past, the "access" was embedded in the "application" provided by a service provider. In other words, voice was the app the customer wanted, and the cost of network access was embedded in the retail cost of using voice.

These days, "voice" increasingly is a separate app from "access." But that might just be another way of saying the real problem for access providers is the legacy revenue disappearing, not the mismatch of value and revenue earned by participants in the Internet value chain. 

Entertainment video is helpful, but actual profit margins in video entertainment have fallen dramatically, by perhaps 50 percent over the last decade or so. 

Pandora says per-track royalty rates have increased more than 25 percent over the last three years, including nine percent in 2013 alone and are scheduled to increase an additional 16 percent over the next two years. So Pandora has to cover the costs. 

Pandora notes that users can listen for free for as many hours as desired on desktop and laptop computers; pay $0.99 for unlimited listening on mobiles for any month when usage exceeds the limit; , or subscribe to "Pandora One" for unlimited listening and no advertising.

No Consumer Demand for 1-Gbps Internet Access?

Time Warner Cable's Chief Financial Officer Irene Esteves continues to get coverage every time she speaks in an investment or other conference about the state of consumer demand for 1-Gbps access, seemingly prompted in virtually every case by Google Fiber's Kansas City 1-Gbps network and service. Esteves repeatedly has said that Time Warner Cable does not see the demand for such speeds. 

Some will be tempted to argue this is a typical effort by a quasi-monopolist to dismiss a competitor's better and disruptive offering, and rather incorrect, since people in Kansas City do seem to be buying Google Fiber. 

A few might say the statement is an effort to stave off, as long as possible, or perhaps indefinitely, the need to invest major new sums in access technology. 

Others will note that since Time Warner Cable faces Google directly in Kansas City, the question is rather an obvious question for investors to ask. There is some truth to all such interpretations. 

On the other hand, to understand the comment that "we just don't see the need of delivering that to consumers." one has to unpack the statement and put it into context. 

Esteves is not necessarily dismissive of Google Fiber.  "We're in the business of delivering what consumers want, and to stay a little ahead of what we think they will want," she said, and seems always to say when asked about the state of demand for 1-Gbps access. 

A fair way to rephrase might be "At the moment, at the prices we would have to charge, we believe few customers would want to buy 1-Gbps access." The statement is highly conditional. 

At very low prices, many consumers would buy 1-Gbps. At significant prices, far fewer will do so. At high prices, a small percentage will purchase. 

A corollary might be that "right now, at significant prices compared to our other offerings, few consumers we sell to would willingly pay the incremental prices to get the fastest speeds."

The "consumers don't want it" is a highly conditional statement. Time Warner Cable and all other executives know full well there is some set of circumstances that could drive very high take rates. 

Time Warner Cable's business objective is to supply any future forecast level of demand, under competitive market conditions, at a profit, without investing prematurely or wildly in ways that harm its financial prospects. 

Some might say Time Warner Cable is craven, stupid or just wrong about end user demand for 1 Gbps access. That's unfair and incorrect. Time Warner is making a highly conditional statement about demand under a finite set of circumstances. The answer will be different under a different set of specific circumstances.

More than Half of Global Internet Users Rely on Mobile

Application providers, device manufacturers, policy makers, service providers and policy advocates all have clear interests where it comes to the actual state of broadband access “supply.”

And mobile access now complicates all evaluations of “access supply.” For example, a new study sponsored by inMobi found that 50 percent of the average global mobile web users now use mobile as either their primary or exclusive means of using the Internet.

That can have important implications for any evaluations of “unserved” or “under-served”  populations, for example. If significant percentages of people prefer to use mobile access compared to fixed network access, that could skew analyses of “problems” in the broadband access arena.

In other words, it is not necessarily a “problem” if many users prefer to buy one product rather than another. At the same time, the state of any nation’s broadband access infrastructure properly would have to entail both mobile and fixed network access in any meaningful evaluation of latency, speeds or cost.

The InMobi study included more than 15,000 mobile users in 14 markets across all continents.

Will Mobile Disrupt "Search?"

Search's future depends, many would say, on how consumers behave on mobile, especially on whether "search" changes as users adapt to smart phones. One key issue, for search applications, is "what" people search for. 

Already, behavior on mobiles has shifted in the direction of "information I need right now" related to commerce (shopping, restaurants, directions to a place) and present location. For Google and other search engines, that is a potential problem. Yelp or Amazon.com become new ways to search. 

In other ways, social networks becomes new ways to "search," as well. 

How Big a Problem are Roaming Charges?

High gross revenue and high profit margin services are both good and bad for service providers. Though it is obvious why service providers might like service that offer high gross revenue and high margin, those conditions are all the incentive competitors need to enter a market. 

And, in many cases, such circumstances drive regulators to take the margin out of a business by regulating it away.

Roaming services are such an issue, even if currently representing only about six percent of total service provider revenues. Also, roaming is a bigger problem in some regions than others. The European Union and roaming between Australia and New Zealand are some examples of regions that are taking steps to curb bill shock and roaming charges.

Juniper Research predicts mobile roaming will represent more than $80 billion worth of revenue by 2017, compared to over $46 billion in 2012, despite a general trend to lower roaming charges. In 2017, roaming will represent about eight percent of operator billed revenues.

In some cases, roaming revenue might represent as much as 10 percent of revenues, though.

Informa predicts roaming revenue will grow 86 percent globally over the next five years, with revenues of $67 billion by 2015. That means roaming will account for 6.3 percent of total mobile service revenues worldwide by then. 

If operators are to capitalize on what can remain a very high-margin business, they have to stop the current trend of roamers shutting off their data roaming capabilities when traveling. B

usiness travelers often stop using their mobiles when traveling internationally, representing lost revenue, of course.

Bill shock, the unexpected surge in a mobile phone bill that often happens when a user is roaming, especially internationally, is a growing problem for service providers, for several reasons. The first reason is simply that after the first “shock,” users learn either to limit their use of mobile devices when in a roaming situation, switch providers, adopt prepaid or take other actions to prevent future shocks.

Excessive roaming charges also increase customer irritation and draw regulator ire. As in the European Union, high roaming charges not addressed by service providers themselves can lead to regulator action to force lower rates.

Sandvine says that audio and video streaming account for 28 percent of roaming data as observed on a tier-one European mobile network. That tends to suggest high roaming charges are an issue for traveling consumers as well as business travelers.

For example, a three-minute YouTube video uses approximately 10 MBytes of data which would cost a European mobile subscriber about 7€ while roaming in Europe.