Mobile Towers are Created Highly Unequal

source: techneconomy blog

Whether you look at revenue, profit or traffic, mobile network cells are not “equal.” Basically, a Pareto distribution typically holds: perhaps 80 percent of traffic is generated by 20 percent of total tower sites.

Looking at usage on a given day, by any single user, perhaps 80 percent of traffic is carried by just three towers. About half of traffic is carried by one tower. The remaining 20 percent of traffic is carried by 28 additional cell sites.

The implications are that some towers are highly profitable, others are self-supporting and some towers probably lose money.

source: FCC

The traditional rule of thumb for revenue and profit for fixed networks in the U.S. market is that service providers make money  in urban areas, break even in suburban areas and lose money in rural areas.

source: techeconomy blog

One way of illustrating the pattern is to note that population density and network cost are inversely related. An analysis by the Federal Communications Commission shows that the cost of networks, per location, grow dramatically as density falls.

Without subsidies of some types (governmental or internal subsidies by the service provider), rural area services likely are not possible.

Rural market capital investment always is higher than investment in urban or suburban areas.

Other studies of 3G network traffic in Western Europe suggest that 20 percent of the towers carry  60 percent of the 3G data traffic.

In U.S. and U.K. Markets, Cable Operators Supply Disproportionate Share of "Really-Fast" Internet Access Connections

source: Huawei

In 2016, Virgin Media has about 19 percent market share of U.K. Internet access subscribers. Yet, in 2014, Virgin Media had 56 percent of “superfast” connections.

In 2016, Comcast had about 24 million U.S. Internet access customers, compared to 16.6 million for AT&T and seven million served by Verizon.

Comcast has in operation a program to upgrade 100 percent of those locations to gigabit Internet access.

Verizon does not yet have a gigabit fixed network upgrade program in place. AT&T is building actively, but it is unclear how many customer locations actually are passed. CenturyLink also is building aggressively.

Google Fiber has not released customer account numbers, but is believed to serve accounts in the low six figures.

That means Comcast alone will have more gigabit-capable locations than all the other ISPs put together.

In both U.K. and U.S. markets, then, cable operators are supplying a disproportionate share of all “superfast” (24 Mbps to 100 Mbps), 100-Mbps and higher, or gigabit capabilities.

Over the medium term, virtually all access platforms--and some new platforms not yet tracked--will be supplying gigabit connections.

Mobile Data Pricing Now is Unstable

source: Jackdaw Research

With the recent addition of “does not count against data cap” DirecTV viewing on AT&T iPhones policies, and with Sprint and T-Mobile US exempting streaming video from data usage caps, it is clear we are in an unstable period for mobile data pricing.

The issue, of course, is that mobile service providers are exempting the most data-intensive apps from usage calculations, and usage drives requirements for network investments.

The pricing anomalies are easy to illustrate. On a revenue-per-bit basis, narrowband apps such as messaging and voice produce the highest returns, video the lowest returns (even if entertainment video represents much more gross revenue).

Assume a fixed network ISP sells a triple-play package for a $100 a month retail price, where each component--voice, Internet access and entertainment video--is priced equally (an implied price of $33 for each component).

How much bandwidth is required to earn those $33 revenue components? Almost too little to measure in the case of voice; gigabytes for Internet content consumption and possibly scores of gigabytes for video.

So, by some estimates, where voice might earn 35 cents per megabyte, revenue per Internet app might generate a few cents per megabyte.

Video might generate fractions of a cent per minute of use (access fee compared to usage).

In addition to the fact that revenue per megabyte tends to drop over time, the bigger issue is that profit per megabyte, and revenue per megabyte, is inversely related to consumption of bandwidth, from an access provider standpoint.

Customer Resistance an Issue for Some Smart Parking Deployments

source: Redtone IoT

How particular smart cities services--including smart parking-- will sustain themselves is a big issue. Irrespective of “soft” value (less air pollution, less traffic), how revenue can be generated to pay for the smart parking infrastructure and operations remains an issue.

User opposition is among the potential roadblocks or issues. One suburban shopping area--which has featured free parking--now is converted to a an app-based paid parking system.

The Reston Town Center in Virginia is activating the new system Sept. 12, 2016, and local residents are--as you would expect--unhappy about the switch to paid parking. Some are unhappy about potentially needing to use the app system, as well.

ParkRTC customers can use an app, website or on-site pay stations that accept credit or debit cards or exact change only.

Park Assist is the system supplier for the area of 50 shops, 30 restaurants, an 11-screen cinema, and other amenities, as well as 9,000 parking spaces in seven parking garages.

Passport is the mobile payment parking provider.

Garages rates will range from $2 for the first hour to $24 for 12 to 24 hours. The street rate will be $3 for the first hour and $6 for between one and two hours.

The system relies on license plate recognition, LED-based space indicators, electronic display boards to indicate quantity of available spaces, and online space counts in real time.

Upon entering one of the center’s seven garages, drivers will locate an available parking space using green and red light indicators. Once parked, the system will read the vehicle’s license plate, and the driver must initiate a parking session, through either the ParkRTC app or ParkRTC.com using a pre-set four-digit pin, or at a pay station.

Using the app or website, driver credit cards will be charged automatically based on their pre-selected length of stay. If using the app or website, drivers will have the option of extending stays remotely.

To be sure, the smart parking features are less the issue than the conversion to paid parking. But the issue remains: would the smart parking have been instituted if the parking revenue were not available to support it?

The business case will be different in urban areas where paid parking already is the norm, to be sure. Still, potential customer confusion or resistance is among the obstacles. No value chain is complete without the customer who supplies the revenue.

source: Gartner

Performance Gaps Rarely Persist--Across Countries, Regions or Service Providers

It always is dangerous to make longer-term predictions based on where technologies or service providers are at the moment. The reason is simply that capabilities can change rapidly, even unexpectedly.

Over the last couple of decades, it has been argued that the United States was “way behind” Europe in use of mobile phones, way behind Japan in access speed, or more recently that Europe as “way behind” the United States in 4G network availability and adoption.

Others have argued that U.S. Internet access prices were high, compared to other countries perceived as leaders. But price is relative. One has to adjust for general price levels across countries, and then to adjust for retail plan differences, to derive price per megabit per second, for example.

Even in 2007, when the price differentials were said to be quite disparate, on a cost per Mbps, U.S., French, German and Japanese prices were comparable.

The point is that such gaps always have closed.

source: OECD  

Many have argued that average or peak U.S. Internet access speeds lagged either Europe or world levels. Those gaps also will close. Since 2011 alone, U.S. Internet access speeds have tripled.

From 2015 to 2016 alone, U.S. Internet access speeds  got 40 percent faster. Much of the credit for those advances goes to U.S. cable TV companies, who are rapidly increasing speeds.

Mobile M2M Will Generate $67 Billion in 2021 Revenue, Ovum Predicts

Mobile machine-to-machine (M2M) connections (not including NB-IoT) will reach 733 million globally by 2021, researchers at Ovum predict. That will drive total mobile M2M service revenues to a global annual total of $67 billion in 2021.

Global mobile M2M connections will reach 733 million in 2021, about  8.1 percent of all mobile connections, up from 4.2 percent in 2015.

Between 2017 and 2021,  total M2M service revenues will grow at a compound annual growth rate of 13.3 percent.

The biggest revenue contributions will come from Asia/Oceania, North America and Western Europe.

The Asia/Oceania market will generate US$ 22 billion, North America will represent US$ 16 billion and Western Europe will create US$14 billion in revenue.

Of all current generations, LTE will be dominant in the long term, accounting for 212 million connections in 2021, Ovum predicts.

source: Ovum

Special Access Prices Are Not Evidence of Market Power, Phoenix Center Argues

Prices in the U.S. special access market (business data services) actually do not actually indicate there is market power, argues George Ford, Phoenix Center for Advanced Legal and Economic Public Policy Studies chief economist.

Why is that important? The U.S. Federal Communications Commission argues there is market power exercised, which it believes explains prices in the special access market. Ford argues that is an unsupported assumption.

If market power is a “bad thing” because it leads to higher prices than would occur in a competitive market, one has to ask what the competitive price might be, as part of the determination of whether market power exists, says Ford.

The answer cannot be “marginal cost of providing the next unit of output,” as that ignores all the sunk costs in the full network.

Telecom markets tend to be oligopolistic, so the assumption of “a perfectly-competitive market” tends to fail, as a useful analytical assumption.

Instead, Ford argues, the relevant “competitive price” in real-world markets is the price that arises from the maximum level of competition supported by the demand- and supply-side conditions of the actual market.

In other words, if market conditions are such that only two firms can profitably offer service, then the noncollusive duopoly price is the “competitive price.”

Firms enter when it is profitable to do so, and they do not enter when it is not profitable to do so. If a market has only two dominant providers, and entry by other firms is lawful, there probably is a reason only two firms operate.

There are real policy implications. “Telecommunications markets are often served by relatively few firms not because of some random process or poor public policy, but because the size of the market is small relative to the fixed cost of providing service (or, equivalently, the fixed costs are high relative to the size of market demand),” says Ford. “If only two firms can profitably serve a market, it is of no value to lament the fact there are not ten firms doing so.”

“Nor is sensible to use the equilibrium price for five firms as a regulatory benchmark in a market that can be served by only two firms,” Ford argues. “If the five-firm price was meaningful, then there would be five firms in the market.”

In other words, the determinants of price are likewise the determinants of the number of firms.

“The lack of entry is not an indicator of market power, it is an indicator that entrants do not believe there is sufficient excess profit in the market to justify the capital costs to serve it,” Ford notes.