Are Broadband Cost and Benefit Out of Alignment?

"Better broadband" is no less a "desirable thing" than better roads, bridges, elecrical grids and resource management programs that actually could work. So with decarbonization, elimination of disease, clean drinking water and sanitation, social equity or quality education. We always have multiple problems to solve.

But opportunity costs always exist. More of one thing means less of the others. Occasionally, it might be helpful to evaluate our priorities. We desire many positive outcomes, but our resources lag our ambitions. So cost and beneift always are legitimate questions.

Virtually all government programs to close the digital divide are touted as important because--it is is argued--broadband leads to economic growth. In fact, careful reports only use the word correlation, not “causation” when discussing broadband and economic growth. 

Often, even correlation cannot be shown, as is the case with much “foreign aid.” Still, correlation often does exist, and for obvious reasons. Some regions and industries are fast growing. It should come as no surprise that areas such as Silicon Valley show both high growth and high broadband availability. 

But it never is clear which holds: high growth leads to wealth; and wealth leads to excess spending power; which leads to demand for quality broadband, good restaurants and other outcomes associated with areas of high income. 

To use a phrase, perhaps high economic growth, high wealth, high educational levels, some industries and lots of younger people lead to quality broadband demand and supply, rather than broadband causing those outcomes. 

Will quality broadband really boost economic growth in sparsely settled rural areas far from urban centers, already losing residents and already suffering from low economic growth? 

To be sure, there is virtual agreement that universal broadband is a good thing, as universal telephone service, universal mobile service, electricity, education or medical care are  considered good things. 

But there still is no actual evidence--possible correlation, but not causal proof--that broadband access--or better quality broadband access--actually does “cause” economic development. 

Nor, for that matter, are many government reports actually clear about the differences between supply and demand issues; consumer choices and supplier business choices; or the ways people actually use the internet as it relates to potential economic benefit.

Reports often confuse “people who choose not to buy a product” with “inability to buy.” The former is a matter of consumer choice; the latter a supply chain issue. It is one thing to say “few people buy gigabit internet access.”

But that does not mean they “cannot buy.” They may choose to buy a different product, such as access at 200 Mbps. 

And while the internet can be used to conduct homework or conduct work,  it mostly gets used to “watch TV,” or engage in social media. 

Few--if any--really believe watching TV or engaging with non-business social media has a positive impact on economic growth in a direct sense, important though it is as a driver of income for influencers, advertisers, writers, directors, actors, studios, streaming services and TV networks. 

Beyond all that, improving broadband involves opportunity costs: other uses for that capital that are not undertaken because we spend the money on broadband. In a broad sense, all public policy choices involve such trade offs: things we cannot do because we chose to do something else.

Quality broadband is a good thing, don’t get me wrong. But it might not have nearly the economic upside people often wish to believe.

What Impact on Subscriptions from New Federal Broadband Program?

The actual impact of the 2021 infrastructure bill passed by the Congress will hinge on the rules governing disbursement of the funds. But the funds are sure to decrease the cost of adding new infrastructure and increase demand as well.

It also is worthwhile noting that not every household “does not have access” to broadband for reasons of supply. In Canada, for example, six percent of homes do not buy fixed network broadband. Some 63 percent report they do not wish to buy. Some 39 percent claimed they could buy, but that cost was the issue. 

Note the non-existent percentage of “non-buyers” who say they “cannot buy” because the service is not available. In the U.S. market, it often is estimated that about 1.5 percent to two percent of homes literally do not have any fixed network service available to them. 

The point is that most buyers unable to use “broadband” (defined as a minimum of 25 Mbps) actually do have internet access from one or more providers, though speeds are not what most of us prefer. 

That is important as we look at the potential impact of new U.S. federal government support for broadband. 

source: Statistics Canada 

Though money can be spent on digital literacy, equipment support or other training, most observers are likely waiting to see what impact can be made on quality of service. Connecting the “unconnected” also is important, for the last one to two percent of homes. 

But quality of service is the bigger issue for most. 

So how might the money could affect a single state, such as Hawaii. Assume total spending is about $300 million. What could that do? Much depends on the mix of demand and supply incentives, and none of that is clear yet. 

Assume, for the sake of argument, half is used to stimulate demand, and half is used to stimulate supply. At a subsidy rate of $30 per month, the fund means $360 in annual benefits for eligible households. That arguably means 416,667 annual subsidy accounts. Over a decade, that would mean 41,667 households per year could save $360 in access fees. 

That might actually exceed actual demand. 

Assume10 percent of people cannot buy internet access. Some might not buy because there are no facilities or cost is deemed too high. Those are easier problems to solve than gaining adoption by households that see no value to computers or internet access, or who see mobile broadband as a reasonable substitute product. 

According to Broadband Hui, some 55,000 households in Hawai‘i (nearly 12 percent) do not buy internet access services. That does not mean they cannot, they simply do not. That might be a demand issue more than a supply issue.

Perhaps seven percent of those homes simply do not wish to buy internet access, or about 3850 households. That implies 51,150 households that either cannot currently buy internet access; have access but not at 25 Mbps or choose to use mobile access instead. 

Mobile-only households might be 15 percent to 20 percent of all homes. Use the lower figure. At 15 percent, that implies 8250 households prefer to use mobile access in place of fixed access. So the potential pool of customers for the $30 subsidies might be 42,900. 

It is not clear that all those households would qualify for, or use, the $30 subsidies, though. Nor is it clear that the new subsidy actually increases adoption very much. 

There are other assistance programs already in place. For example,98,000 households might qualify for the $30 subsidy as recipients of support from the Supplemental Nutrition Assistance Program (SNAP). Medicaid recipients also are eligible. So are students receiving Pell grants. 

But most internet service providers already were providing low-cost internet access for low-income customers,supplying 25 Mbps access at prices between $5 and $15 a month.

The point is that low-cost 25 Mbps access already has been available to households at prices between $5 and $15 a month from all the major suppliers. It is not clear how many low-income households that want internet access were not already buying. 

source: High Speed Internet.com 

The point is that the $30 subsidies are more likely to shift demand from some service plans to others than to cause non-subscribers to become subscribers, since the subsidies can be used to defray consumption of any tier of service. 

In that sense, allocating half the estimated $300 million to service subsidies might have relatively slight impact. The bigger problem might be lack of any service, and inability to buy any grade of service faster than 25 Mbps. 

So there is likely a public policy argument for spending most of the broadband funds on infrastructure improvements, rather than additional service subsidies.

In U.S., Federal Funding Changes FTTH, Mobile, Fixed Wireless, Cable Business Case

Facilities-based access network business models are changing, and higher levels of government spending to bring down costs for networks in rural and other high-cost areas is among the reasons. The 2021 passage of an “infrastructure” bill by the U.S. Congress will reduce costs in several ways. 

The bill includes $42.45 billion in grants to states for broadband projects, which can range from network deployment to data collection to help determine areas that lack service. Not all of that money will build infrastructure, to be sure. But much will. And the plan allocates at least $100 million in funds to every state, with lesser amounts to U.S. territories. 

source: S&P Global Market Intelligence 

If the money is not wasted, the cost of adding new internet access facilities should fall. Also, additional locations could find they are upgraded to  increase connection speed. 

There also are provisions that will stimulate demand. The bill allocates $14.2 billion for consumption subsidies. To be sure, there have been subsidy provisions before. But the new bill widens eligibility for such subsidies. 

Included in the bill is a $30-a-month voucher to low-income Americans to pay for internet service.

This program replaces the temporary $50-a-month Emergency Broadband Benefit program that was part of efforts to sustain the economy during the period when work and schooling were mostly shut down,  offering less money monthly, but increasing the number of those eligible.

Another demand stimulation effort is the allocation of $2.75 billion for digital inclusion and equity projects, such as improving digital literacy or online skills for seniors.

Additionally, $2 billion was allocated  for rural broadband construction by U.S. Department of Agriculture, as well as another $2 billion for a Tribal Broadband Connectivity Program run by the National Telecommunications and Information Administration (NTIA). Both ideally will help create new facilities. 

The bill also allocated $1 billion to build "middle mile" infrastructure to connect internet service providers to internet access points.

Finally, $600 million was authorized for bonds to finance broadband deployment projects in rural areas.

All that amounts to stimulating demand and supply of internet access, ideally. Undoubtedly some of the money will be wasted. 

But that much additional demand and supply stimulation is going to change business models for the better in many cases, directly lowering the cost of building facilities and defraying consumption as well. 

That is among the reasons many telcos are boosting their spending on fiber to home projects, AT&T, Lumen Technologies and Frontier Communications among them, and why other firms such as T-Mobile and Verizon are dramatically investing in fixed wireless on a national basis. 

Business models are better on the demand side, while the cost of such facilities is lower on the supply side.

"Time" is an Important Variable for FTTH

Almost 40 years ago, an engineering vice president at a major connectivity provider quipped that “fiber is the future….and always will be.” The humor lay in the fact that deploying the “best network” requires a complicated assessment of what constitutes “best” for a particular contestant, at a particular time, with a particular combination of assets and constraints. 

We might note that the argument for fiber to the home as the ultimate solution has been “correct” for at least 50 years. But it also has not been the “best for my business today” for that same length of time, for every provider and in every geography, given the existing cost and demand curves. 

Today, the analysis is even more complicated by the change in demand. Where the business cases might once have been built on revenues from internet access, video entertainment services and voice, increasingly the fixed network business case is driven by consumer broadband and mobility plus enterprise use cases. 

"Cost per location" is one key input. But so is "expected revenue." "Cost per passing" and "cost per customer" as well as "revenue per passing" and "revenue per customer" also matter when competitive conditions prevail. The reason is that a great percentage of invested capital will be stranded, generating no revenue.

source: Semantic Scholar

The payback model necessarily extends beyond FTTH to mobility platform support and enterprise and business communications demand. The traditional arguments about lower operating cost remain. 

Evaluators might agree, in principle, that fiber to the home is the ultimate “best” solution in some cases, while also insisting other choices continue to make financial sense in the immediate time frame and for some business models. 

Rarely, if ever, in the access portion of the connectivity or computing businesses is there one single solution that works “best” for all use cases and requirements. Instead, architects and business managers have to balance numerous values and costs.

Among them, “time” is an important consideration, even if not shown in the formal cost and performance analyses. Basically, this dimension boils down to the time value of money. 

Making 10 Gbps internet access speeds available “right now” when demand is at far lower levels can be the wrong business decision. Generally, internet service providers want to match performance to customer demand and willingness to pay. Raw performance is not the only issue. 

Platform choices often boil down to “what works for the next decade, in the context of our fundamental business model choices?”

In other words, it can make sense to choose a less-capable platform now because it boosts revenue upside and reduces risk, even if that platform is not the “ultimate” solution. 

Lumen Technologies now estimates a cost less than $1,000 per passing for FTTH, in a 16-state territory that is about 70 percent urban and suburban, after the sale of former CenturyLink assets in 20 states, for example, about half what such investments might have cost two decades ago, and perhaps a third of what might have been necessary 40 years ago. 

But what makes sense for Lumen or many independent internet service providers does not make sense for Starlink, Comcast, many rural ISPs, T-Mobile or even Verizon and AT&T, in some instances. Starlink’s value is based on applications suited to constellations of low earth orbit satellites. Comcast can still rely on hybrid fiber coax as a mainstay, if not the sole platform. 

And demand is better matched to facilities cost in many rural, mountainous and hilly or heavily forested areas using some platform other than FTTH. 

T-Mobile will focus on both mobile and fixed wireless. Verizon, especially, will rely on 5G fixed wireless outside its fixed network footprint. 

The point is that there is no contradiction between the belief that “optical fiber to the home is the ultimate solution” and the countervailing arguments that other platforms make more sense in the shorter term, in many geographies, by ISPs with different business models, capital investment constraints or business models. 

-------------------------------------

PTC'22, New Broadband Funding, PTC Impact on Locals

With PTC’22 just a week or so away, program committee members Joe Weinman and Gary Kim talk about this year’s program with Burt Lum of Bytemarks Cafe. 

They discuss highlights of the conference program, changes in PTC mission over the last 40 years, impact of new federal broadband funding on Hawaii, and the impact of PTC on the local economy. 

You can listen to the full podcast here.  

Burt Lum

Joe Weinman

Gary Kim

Home Broadband Prices have Dropped in Real Terms, Over the Last 2.5 Decades

U.S. home broadband inflation-adjusted costs have declined since the mid-1990s, according toan analysis of U.S. Consumer Price Index data. That will often not be obvious when observers consider only “current” prices for home broadband, and compare them to past “retail” prices. 

Comparing prices internationally over time is even harder, since there additionally are currency issues and general cost of living differences between nations.

Two primary forces are at work: price inflation over time and changes in quality or performance and features of the “same” products over time. Still, despite the oft-heard complaint that home broadband prices are "too high," they actually have dropped over two decades.

Consider U.S. prices.

According to the U.S. Bureau of Labor Statistics, prices for internet services and electronic information providers are 21 percent lower in 2021 versus 1997, for example. 

Other communication and computing related prices also have fallen, often in both stated and “real” terms after adjusting for inflation. 

According to the U.S. Bureau of Labor Statistics, prices for communication are 22  percent lower in 2021 versus 1993.

Between 1993 and 2021: Communication experienced an average inflation rate of -0.88 percent per year. 

Compared to the overall inflation rate of 2.26 percent during this same period, inflation for communication was significantly lower.

Also, according to the U.S. Bureau of Labor Statistics, prices for computer software and accessories were 74 percent lower in 2021, compared to 1997. Information technology, hardware and services are 92.64% lower in 2021 versus 1988. 

Prices for information and information processing are 27 percent  lower in 2021 versus 1993. 

Hedonic adjustments also are exceedingly common in computing and information products over time. The obvious examples are the price and performance of dial-up internet access in 1995 and broadband in 2021; the cost of computing operations; data storage or bandwidth. 

Hedonic qualIty adjustment is a method used by economists to adjust prices whenever the characteristics of the products included in the consumer price index change because of innovation. Hedonic quality adjustment also is used when older products are improved and become new products. 

That often has been the case for computing products, televisions, consumer electronics and--dare we note--broadband internet access services. 

Hedonically adjusted price indices for broadband internet access in the U.S. market then looks like this:

source: Bureau of Labor Statistics 

In other words, dial-up internet access and gigabit broadband are not the same product. 10 Mbps broadband is not the same product as 100 Mbps or 500 Mbps service. 

The same trend holds for mobile phone service, phones and other consumer electronics gear. The value and “quality” of a mobile phone subscription in 2000 is not the same as the 2020 value. Nor are the capabilities of a mobile phone the same in 2020 as was true in 2000. 

Frontier Fiber Model Shows How Models Have Changed

In recent investor presentations, Frontier Communications has made three points about its prospects for revenue growth based on optical fiber deployments: the number of consumer broadband accounts; the number of businesses within 250 feet of existing fiber assets and the number of cell towers within one mile of Frontier fiber assets. 

Recent presentations also have shown fiber-to-home home broadband average revenue per user of about $63. 

source: Frontier Communications 

For those of you who have followed the business model for home broadband over the past few decades, that number might seem quite surprising. The late 1990s justification for FTTH was the ability to sell subscription TV as well as faster internet access. Keep in mind that “faster” at that point was 10 Mbps. 

About 2000 the “average” U.S. cable TV bill was estimated to be about $32 a month, according to CordCutting.com. It actually is hard to remember what home internet access actually cost between 1995 and 2000, in large part because most people were buying dial-up services in 1995, while broadband subscriptions did not reach parity with dial-up until about 2005. 

In August 2000, only 4.4 percent of U.S. households had a home broadband connection, by some estimates.  

But a dial-up cost in the $10 a month to $15 a month range is close enough for 1995. Consumer broadband with speeds in the less-than-1.5 Mbps region cost more than that, perhaps in the $30 a month range by 2005. 

The point is that a telco customer with a voice line generating $30 a month, plus internet plus video could have been worth about $100 a month in revenue. Ignoring for the moment the issue of market share in each of the services (compared with a competing cable TV company), the potential revenue was as much as $100 a month for an FTTH-passed consumer location.

Now Frontier says ARPU for an FTTH customer is about $63 a month. Assume that figure includes some amount of voice revenue and zero video revenue. The change in revenue expectation (not adjusted for inflation) per potential customer is roughly 40 percent lower than might have been the case in 1995. 

The biggest change is the assumption that future revenue will be driven principally by one service--home broadband--not three potential sources. 

That was impossible to foresee in 1995. Absent the potential upside of video, and being charitable about the future of fixed network voice, most executives would have argued that the upgrade to FTTH would never make financial sense. 

So lots of assumptions have changed. Among them, in Frontier’s case, is the expectation that business customer revenues from cell tower backhaul and business broadband and services will underpin the FTTH network business model. 

The mid-1990s expectation of higher revenues from video and internet was only partially validated. Linear video no longer figures into the model, though over-the-top streaming might, for some access providers. 

Revenue contribution from voice arguably has been far worse than initially expected. 

So consumer revenue is principally driven by home broadband. Overall payback models for FTTH now lean on business customer revenues and backhaul. 

It is part of the change in FTTH business models in the U.S. market.