Tuesday, November 30, 2021

Technology Never Happens in a Vacuum


It builds on earlier innovations and requires people to make it happen.

Monopoly, Duopoly, Oligopoly: Can Competition Happen Under All Scenarios?

Most communications regulators believe duopolies are injurious to competition, innovation and investment. And yet the degree to which that is true is a bit unclear. 


Uganda has a mobile duopoly and yet subscriptions keep climbing. In the fixed networks market, while some would argue that the cable operator-telco duopoly is not competitive, others would point to declining prices, heavy investments in bandwidth supply and available speeds as evidence that competition is producing results. 


In the telco world, faster home broadband is nearly synonymous with fiber to the home upgrades. In the U.S. and many other markets, the issue is available bandwidth, not physical media. 


More than 80 percent of U.S. homes can buy gigabit per second internet access if they choose, from the local cable operator. And though U.S. telcos are stepping up their optical fiber access investments, fewer homes are reached by FTTH. 


Still, according to the Federal Communications Commission, 88 percent of U.S. homes can buy internet access at gigabit speeds. The NCTA says home broadband speeds have increased 1880 percent over the last decade alone.  


source: NCTA 


It is possible to debate whether infrastructure or retail competition produces better outcomes, especially since, in many markets, only retail competition (using wholesale access from one facilities supplier) is feasible. 


Some claim U.S. home broadband prices are too high, the typical argument being that U.S. a la carte prices (the retail tariff for internet access, not purchased in a bundle) are higher than prices in other countries.  


Adjusting for currency and living cost differentials, however, broadband access prices globally are remarkably uniform. 


The 2019 average price of a broadband internet access connection--globally--was $72..92, down $0.12 from 2017 levels, according to comparison site Cable. Other comparisons say the average global price for a fixed connection is $67 a month. 


Looking at internet access prices using the purchasing power parity  method, developed nation prices are around $35 to $40 a month. In absolute terms, developed nation prices are less than $30 a month. 


According to a new analysis by NetCredit, which shows U.S. consumers spending about 0.16 percent of income on internet access, “making it the most affordable broadband in North America,” says NetCredit.  


In Europe, a majority of consumers pay less than one percent of their average wages to get broadband access, NetCredit says. In Singapore, Hong Kong, New Zealand and Japan,  10 Mbps service costs between 0.15 percent and 0.28 percent of income. 


The point is that home broadband prices fall everywhere, over time.

source: Strategy Analytics 


Looking at 95 countries globally with internet access speeds of at least 60 Mbps, U.S. prices were $62.74 a month, with the highest price being $100.42 in the United Arab Emirates and the lowest price being $4.88 in the Ukraine. 


According to comparethemarket.com, the United States is not the most affordable of 50 countries analyzed. On the other hand, the United States ranks fifth among 50 for downstream speeds. 


The point is that duopoly, oligopoly or even monopoly can produce retail competition. There is room to argue about how much competition, investment or innovation is possible. But connectivity no longer is a “natural monopoly” in terms of retail competition. 


There is a stronger argument for infrastructure monopoly in many markets. And duopoly might be the only realistic outcome in some mobile markets, even if most regulators believe three is the minimum number of mobile firms necessary to promote robust competition. 


Is U.S. Home Broadband on Cusp of Major Change, or Not?

The U.S. home broadband market has been dominated by cable operators for two decades. Depending on the source, cable operators have held close to 70 percent of the installed base of accounts and have had up to 100 percent of net new additions (market share) in many years. 


Many believe that now will change, though there is disagreement about how much change is possible. 


U.S. telco fiber-to-the-home (FTTH) lines will pass 82 million American households by 2027, nearly double the 44 million households passed today, Cowen predicts. 


Led by AT&T, the four biggest telcos (AT&T, Verizon, Frontier and Lumen) will account for the lion's share of those deployments, together passing more than 71 million homes with fiber.

source: S&P Global Market Intelligence


U.S. cable operators will pass another five million homes with fiber lines over the next six years as well, largely deployed by Altice USA. Cable operators have deployed five million FTTH lines already, says Cowen. 


By 2027, telcos will get adoption by about 43 percent of homes passed by the FTTH networks, Cowen predicts. The issue is the percentage of those new FTTH lines that are market share neutral upgrades by existing telco digital subscriber line customers, and what percentage will come from market share taken from other providers. 


Cowen believes relatively small amounts of market share will actually be taken from cable operators. Cable operator market share is expected to decline from 61 percent today to 58 percent in 2027.


Telco share, meanwhile, will climb from 25 percent to 27 percent in 2027. The magnitude of the shift is where there is disagreement. - 


Some might call the Cowen share forecasts too conservative. Moffettnathanson, for example, believes a 50-50 split of the installed base between telcos and cable operators is possible. 


Telco executives, for example, believe they will do better than Cowen suggests. Frontier, for example, expects to reach about 45 percent of the installed base when it completes its FTTH upgrades.


Some telco executives also believe  fixed wireless will play a role in their share gains. Altogether, observers predict some shift of installed base and market share as telcos go up tempo on FTTH upgrades and 5G fixed wireless gets marketed. How much change could happen is the issue.


Before Metaverse There was Second Life

Before the metaverse there was Second Life. It has been a decade and a half since Second Life was heralded as the next big thing: virtual worlds. Second Life still is around, but did not really become the next big thing. In fact, significant new technologies often take decades to become commercially relevant or ubiquitous.

  

Advanced technology often does not get adopted as rapidly as the hype would have you believe. In fact, most useful advanced technologies tend not to go mainstream until adoption reaches about 10 percent. That is where the inflection point tends to occur. That essentially represents adoption by innovators and early adopters.


source: LikeFolio


Consider mobile phone use, among the most-ubiquitous products used globally. On a global basis, it took more than 20 years for usage to reach close to 10 percent of people. The point is that even a truly useful or transformative new product or technology can take a decade or more to reach the early adopter stage, which is when 10 percent of people or households use an innovation. 


source: Quora


That is why Sigmoid curves are the rule for product or technology diffusion. The S curve has proven to be among the most-significant analytical concepts I have encountered over the years. 


It describes product life cycles, suggests how business strategy changes depending on where on any single S curve a product happens to be, and has implications for innovation and start-up strategy as well. 


source: Semantic Scholar 


Some say S curves explain overall market development, customer adoption, product usage by individual customers, sales productivity, developer productivity and sometimes investor interest. It often is used to describe adoption rates of new services and technologies, including the notion of non-linear change rates and inflection points in the adoption of consumer products and technologies.


In mathematics, the S curve is a sigmoid function. It is the basis for the Gompertz function which can be used to predict new technology adoption and is related to the Bass Model.


 I’ve seen Gompertz used to describe the adoption of internet access, fiber to the home or mobile phone usage. It is often used in economic modeling and management consulting as well.


The next big thing will have first been talked about roughly 30 years ago, says technologist Greg Satell. IBM coined the term machine learning in 1959, for example.


The S curve describes the way new technologies are adopted. It is related to the product life cycle. Many times, reaping the full benefits of a major new technology can take 20 to 30 years. Alexander Fleming discovered penicillin in 1928, it didn’t arrive on the market until 1945, nearly 20 years later.


Electricity did not have a measurable impact on the economy until the early 1920s, 40 years after Edison’s plant, it can be argued.


It wasn’t until the late 1990’s, or about 30 years after 1968, that computers had a measurable effect on the US economy, many would note.



source: Wikipedia


The point is that the next big thing will turn out to be an idea first broached decades ago, even if it has not been possible to commercialize that idea. 


Metaverse seems to follow the pattern.


Sunday, November 28, 2021

6G Poses Challenges for Fixed Networks

Since most observers expect 6G to feature access speeds at least 10 times greater than 5G, that implies top speeds in the range of 100 Gbps, with a lower range in the 10 Gbps region. 


 

source: Navixky


In fact, some believe top speeds could reach a terabit per second. 


 

source: Navixky


All that has to be a driver of thinking about upgrades of cabled internet access networks. If, within the next decade, mobile networks can reasonably deliver gigabit to 10 Gbps speeds, that makes them substitutes for the fixed network. And if, in a decade, 10 Gbps to terabit-per-second speeds are feasible, then cabled networks might in some cases be the second-fastest networks available to consumers. 


All of that could have big implications for internet service provider market share. The issue in some markets will not be “cable or telco” but rather “mobile versus fixed.”


Monopoly Versus Wholesale: When Each Happens

Wholesale is a useful option in the communications business. Mobile virtual network operators build their whole businesses on it. Global communications and roaming depend on it. Capacity requirements often use wholesale mechanisms.


But wholesale is not the same thing as "monopoly."


Sanctioned monopolies--public or private--typically exist because there is some market barrier to sustaining multiple suppliers. Roads, airports, seaports, electricity and natural gas supply, water and sewage systems provide obvious examples. In the decades prior to 1980, telecommunications was widely believed to be such a case. 


Virtually all telecom operations globally were official government monopolies, and often also operated directly by the government. All that began to change in the 1980s, gathered force globally in the 1990s and now competition is the common pattern in communications. 


Still, some believe growing capital investment, low growth and challenged profit margins will lead to a lessening of competition, at best, and re-monopolization at worst. And that is the issue in Malaysia, as the Digital Nasional Berhad, a government entity, has been given a monopoly on 5G spectrum, something virtually unprecedented, and also is supposed to become the sole national provider of 5G infrastructure. 


That presumably also means future generations of mobile network, such as 6G, also will supplied as a monopoly by the DNB. 


There is an argument for such intervention when the markets are not working. That was believed to be the case for most of the history of global telecommunications. 


And, to be certain, there are concerns in some quarters about the long-term viability of multiple facilities-based mobile service providers. This is more than the consolidation we already commonly see in the competitive telecom business. 


In some regions, some suggest profitability is so difficult that future markets might only sustain one supplier. So, the industry would return to monopoly supply. Sustainable profits are the issue, in some cases, some argue.  


It does not seem that facilities-based competition in Malaysia is so ruinous that private entities are unable to sustain private 5G, though. 


A study produced by DT Economics and sponsored by the GSMA argues a 5G monopoly owned by the government points out that the mobile market in Malaysia is not broken.


 “Malaysia has already achieved a competitive mobile market,” the report argues. “ It has four major Mobile Network Operators (MNOs), each having a market share of between 18 and 28 percent, with one additional smaller player (with a market share of less than 5 percent).”


“Malaysia also has a small, but a well-established, Mobile Virtual Network Operator (MVNO) market hosted by the largest four MNOs,” the authors say. 


The market arguably has gotten more competitive with each mobile next generation network, to be sure. The issue is the degree to which the private entities are unable to sustain themselves based on competitive access to spectrum and facilities-based networks. 


source: DT Economics


If if one accepts the thesis of higher capital investment in 5G or other future networks, it is not necessary to re-monopolize either spectrum or infrastructure. Mobile firms are capable of creating infrastructure sharing mechanisms that reduce costs, but on a voluntary basis. 


One also can argue that mobile firm mergers accomplish many of the same objectives as infrastructure sharing or monopoly facilities: a reduction of cost, while arguably maintaining a higher rate of innovation. 


Monopoly and wholesale are not synonymous, of course. 


Wholesale is the norm for international communications, as in the case of data access, messaging or voice services. No single company has a network that reaches “everywhere,” so that is obvious. Domestic operations are far more complex.


In most markets, fixed network capacity has been a near-monopoly and wholesale underpins competition. In mobile markets, facilities-based competition is typical, though wholesale access remains important for mobile virtual network operators where MVNOs are lawful. 


As a rule, movements toward wholesale have been gaining ground in the local access markets. Mobile operators divesting their cell tower assets and leasing access provide the best examples, but more recent actions have mobile operators outsourcing their information technology platforms for 5G to hyperscale cloud computing suppliers. 

source: McKinsey


The broader strategic issues concern value: where can service providers create new value most effectively, and how does that drive capital investment decisions? 


Some might argue DNB confuses wholesale with monopoly. The former almost always happens because it makes good business model sense. That latter almost always happens because there is a failure of business models. 


Some would argue there is no evidence of market failure requiring re-monopolization.


Saturday, November 27, 2021

Would Malaysia DNB Ever Become a Mobile Retailer?

Malaysia is taking an unusual strategy for licensing 5G spectrum: it has built a government-owned 5G wholesale network and restricts the ability of mobile operators to refarm their licensed 4G spectrum to support 5G. The Digital Nasional Berhad wholesale network will begin commercial operations soon, but the leading Malaysian mobile operators are opposed to the move. 


DNB says renting will cost half or perhaps 40 percent of what the mobile operators would have to spend to build their own networks.  


In Malaysia, mobile device data consumption grew 35 percent in 2021. Nationally, fixed network broadband penetration is about eight percent, where mobile adoption is in excess of total population. In other words, most people rely on mobile devices for internet access in Malaysia. 

source: Khazanah Research Institute 


That is an issue in Malaysia now that the government is building a national 5G wholesale network and placing obstacles in the way of mobile operators who might otherwise wish to refarm 4G spectrum to support 5G. 


The government has said  5G wholesale rates will be about 40 percent less than the annual cost of operating an owned 4G network.  

source: Khazanah Research Institute, Malaysian Communications and Multimedia Commission 


But the leading mobile operators are not convinced. They say a lack of transparency about future plans is an issue. There is no clear guidance on whether the government might itself decide to become a retail competitor, for example. 


As always, that would not be advantageous to the other wholesale customers of the network.


Thursday, November 25, 2021

KKR Bids for Telecom Italia

It is official: private equity firm KKR is making a bid to take Telecom Italia private. Said to be the biggest-ever private equity bid for a public telecom service operator in Europe, the $12 billion deal seems to be opposed by Vivendi, which owns 24 percent of Telecom Italia. 


The KKR playbook would normally involve an effort to streamline, rationalize and reposition the assets. KKR is believed to be interested in separating Telecom Italia’s network assets from the retail operations, turning part of TI into a wholesaler of capacity, likely with a heightened optical fiber position, while retail operations are conducted separately, using the wholesale network. 


At first glance, the proposed deal looks like a standard private equity deal: buy an underperforming asset, make changes and then sell. But the deal might also reflect another private equity focus: buying infrastructure assets to hold longer term, as an alternative asset. 


Perhaps a likely scenario is that KKR hopes to dramatically improve financial performance before selling the asset to an investor that wants the long-term cash flow. 


Telecom Italia, for its part, also fits the “go private” scenario: it has high debt and shrinking recurring revenues and profits, arguably impairing its ability to invest in digital infrastructure including fiber to home facilities. 


Among the key drivers for telecom privatizations is the perception by asset owners that public markets will not positively reward the firms, in terms of equity valuations, commensurate with their revenues, cash flow or potential growth prospects.


Another key driver is private equity firms with lots of private capital to invest, and assets that offer long-term and predictable cash flows to institutional investors such as pension funds and other entities with long time horizons that view infrastructure assets as equivalents to other long-duration fixed-income assets such as bonds.

 

Also, asset diversification is another motivation for investors.

  

There is a good reason why any number of public telecom firms have been taken private, and why others are considering similar moves: high debt, low growth and poor operational performance. And connectivity providers are not the only type of firms facing investment issues.  


That is a fairly-common prescription for any public company to be taken out of the public markets by private equity, and many public telco assets  fit the bill. 

source: Focus Finance


One defining characteristic of infrastructure assets is their monopolistic position. We tend to forget that for most of the history of the industrialized world, much of the funding for large scale public infrastructure such as roads, canals and railroads has come from private sources of capital. And that includes telecommunications in the United States. 


source: Maria Sward 


The function of private equity also has included the rehabilitation of firms that are not performing financially. Private equity buys a public asset, restructures and then sells the asset, often within about a five-year period. 

source: Bain


Control Plane, Data Plane in 5G Networks

Separation of the data plane from the control plane is a fundamental architectural principle of cloud-native communications networks. That, in turn, also applies to the ways connectivity service providers build their 5G networks, where it comes to management and control.


source: Marc's blog 


Separation of data and control planes is the foundation of virtualized networks, and since the 5G network is, by definition, virtualized, underpins all 5G core and access networks. 


The control plane can be thought of as the signaling and routing functions of the network. The data plane is the part of the network that actually delivers and sends user data and content (sometimes known as the user plane, forwarding plane, carrier plane or bearer plane) is the part of a network that carries user traffic).


Separated control and data planes therefore underpin the role of computing resources within the connectivity providers’ network. The data plane physical network must connect every user. The control plane functions can be housed anywhere in the network where it makes sense. 


The virtualization principle also creates the value of edge computing. From a service provider perspective, control plane separation allows less capital investment in firmware and hardware, as control and processing logic and  functions can happen at fewer locations. 


In the mobile access network, that means separating radios from baseband signal processing units, allowing many fewer of the latter to be purchased and used. 


The 5G control plane also requires edge computing, though. Some service provider control logic remains near the physical network edge, but not at the absolute physical edge, or the radio transmission edge. 


In other words, the 5G network requires edge computing, simply to supply control plane functions. 

source: Reply 


Similar considerations apply to the data plane. Some applications will require very-fast processing. So there will be a necessity, not simply an advantage, to having computing take place as close to the end user device as possible (economically and computationally). 


We might debate whether edge computing is an example of “data plane” or the “control plane.” Some of us would say edge computing is “data plane,” since it is a necessary part of processing end user requests and data, in the same way that remote cloud computing requests provide data plane (end user applications) functions. 

source: Accenture 


The bottom line is that 5G networks, and presumably all to follow, will use edge computing as a principle of the control plane and data plane functions.


Many Winners and Losers from Generative AI

Perhaps there is no contradiction between low historical total factor annual productivity gains and high expected generative artificial inte...