Monthly Archives: May 2021

The Ghost of US West

In a recent earnings call, Jeff Storey, the CEO of Lumen made it known that the company will be actively looking to sell non-core assets after sluggish revenue growth in the first quarter of this year. I speculate that the non-core assets include selling rural copper assets and customers. CEO Storey came to the business from Level 3 when his company merged with CenturyLink.

The business that Storey is trying to run is the remnants of US West, one of the seven Baby Bells created in 1983 from the divestiture of AT&T. The company was the incumbent telephone company in 14 western states ranging from Nebraska to Arizona to Washington. Until 1990 when the business consolidated under the US West name, it controlled three holding companies that retained their original telco names – Mountain Bell headquartered in Denver, Northwestern Bell headquartered in Omaha, and Pacific Northwest Bell headquartered in Seattle.

US West struggled from its inception compared to the other Baby Bells, mostly due to the huge geographic footprint of the business and the huge swaths of rural areas served by the business. In 1998 the company tried to increase shareholder value by spinning off MediaOne, which took the cable, cellular, and international business lines.

Qwest acquired US West in 2000 in a hostile takeover. Qwest was owned by Phillip Anschutz who had built a long-haul fiber business based upon using the rights-of-ways of the Southern Pacific Railroad. Qwest was formed in 1996 and benefitted from the CLEC boom of the late 1990s had a high-flying stock price. Regulators made the merged companies divest of the long-distance business as part of the merger.

The US West business was quickly renamed to the Qwest brand. The newly formed company quickly got into financial trouble due to the crash of the dot-com companies and large CLECs in 2000. Qwest had also ventured into a large partnership in Europe with the Dutch incumbent telco KPN that collapsed by 2002. Qwest got quickly into trouble with regulators for slamming – moving customers to its services without customer approval. The company was also fined $250 million by the Security and Exchange Commission for shady transactions with the telecom arm of Enron. Since Qwest had borrowed a huge amount of money to pay for US West, the business was quickly teetering on the edge of bankruptcy. In 2002 Qwest sold its directory business to help stave off bankruptcy.

Qwest limped along until it was acquired in 2010 by CenturyLink, a regulated incumbent telco from Monroe, Louisiana. CenturyLink had grown through acquisitions. For example, before buying Qwest, the company had acquired telephone lines from GTE, Verizon, Telephone USA, and smaller telcos like Madison River. CenturyLink’s final acquisition before buying Qwest was a merger with Embarq, the telco that had been built through acquisitions by Sprint.

CenturyLink acquired Qwest in a stock-for-stock cashless transaction in April 2011, which meant that CenturyLink inherited the still large remaining debt from Qwest. CenturyLink made some smart acquisitions that made it an early player in cloud computing, buying companies like Savvis, Ciber, AppFog, DataGardens, and Cognilytics.

CenturyLink acquired fast-growing Level 3 Communications in 2017 for $25 billion – although industry insiders say that Level 3 acquired CenturyLink. Within a few years, the former Level 3 executive team including Jeff Story took over the operations of the business. In September 2020, the company was rebranded as Lumen.

These various permutations of US West have struggled financially since divestiture from AT&T in 1983. The company was the slowest-growing Baby Bell after divestiture and was seriously crippled with debt after the hostile takeover by Qwest. I’ve heard the commonly spread rumors that the Level 3 executives are constantly frustrated by the inability to make changes inside the company due to the regulated nature of the core telco business.

Lumen now owns one of the biggest middle-mile fiber networks in the country, but the last mile copper is getting ancient. It wasn’t in the greatest shape at divestiture and the copper has aged 38 years since then. The question that industry watchers like me ask is if the old copper plant will be spun-off yet again to another company that will try to make it work – will the ghost of US West continue to sink everybody who touches it?

Paying for Low-income Broadband

Digital inclusion advocates for years have been looking for ways to help low-income households connect to the Internet.

The only federal program to do has been the Lifeline program that is part of the FCC’s Universal Service Fund. This fund has been allowing a $9.25 discount for either voice or broadband (only one discount per household). The discount might have felt substantial when first created in 1984, but with inflation, this discount doesn’t make a big difference in a cable company broadband bill that can easily cost $90 today. To make the program even weaker, large telcos like AT&T have opted out of this program in every state where regulators have allowed it.

The only other material option before now has been Comcast’s Internet Essentials program that provides curtailed bandwidth to low-income households for a low price. The program was started in 2011 as one of the terms to allow Comcast to purchase Time Warner Cable. The program started in 2011 by offering a 1.5 Mbps download for $9.95 per month. The speed has been increased many times since 2011 and recently was increased to 50/5 Mbps. Comcast downplayed the program for years, but eventually embraced it and was providing the service to over 8 million homes by the beginning of the pandemic.

The Federal government is back in the game and has provided $3.2 billion of low-income broadband support in the Emergency Broadband Benefit (EBB) program that was funded as part of the $1.9 trillion American Rescue Plan Act. This program provides up to $50 per month to cover broadband bills for qualified low-income homes. The main problem with this subsidy is that it has a finite end and will cease when the fund runs out of money. We’ve also seen this program be misused when Verizon required customers to upgrade to faster broadband to get the discount.

Another attempt to get low-income broadband into homes comes from the New York Legislature that has passed a bill that requires ISPs to provide a low-income connection to qualifying homes for $15 to $20 per month. This bill is being challenged in court.

So where do we go from here? Comcast recently committed to continuing the Internet Essential program for 10 more years. That’s going to benefit families in cities where Comcast provides service.

The FCC’s Universal Service Fund needs a majorly revamped funding mechanism because it’s now funded with a 30% and growing levy on Interstate telecommunications service. It’s becoming clear that this is not sustainable. Additionally, the $9.25 discount does not make broadband affordable for low-income homes.

It won’t take long to use up the new EBB fund at $50 per household per month. The fund will last six months to help 10 million homes or a year to help 5 million. This can only become permanent if Congress decides to embrace it as a needed solution, and Congress has not been forthcoming in establishing permanent subsidy programs. It could be extended, but in the long run, this is the kind of program that will always be on the chopping block with a change in administration. This program also worries me when I contemplate millions of homes losing a new broadband connection after the funding runs dry.

The New York law is not the right answer. While it may feel satisfying to force all ISPs to follow the Comcast lead and offer a low-cost broadband alternative, such a requirement could devastate smaller ISPs and could easily put them out of business. If a law like this remains in effect, we’re going to see extreme redlining as ISPs bypass low-income neighborhoods. In the long run, if this law applied to all ISPs, we might eventually see poor neighborhoods with no ISP choices.

There is no easy solution. It should be clear to everybody that the country is better off if all homes are connected. We know definitively that lack of home broadband has a direct negative impact on school achievement. Unfortunately, none of these solutions feel like the permanent answer. The best solution is being offered by Comcast, but other big ISPs haven’t voluntarily stepped up in the same way. This is a blog that doesn’t suggest a solution as a closing punchline. There is no clean universal solution if coerced low rates won’t work, and if big ISPs don’t instead step up to the plate. That leaves only government subsidies, and I don’t think anybody trusts those to be permanent.

What did the Pandemic Hide?

The growth of broadband customers has looked spectacular over the past year during the pandemic. It’s easy to chalk up higher broadband customers nationwide to the need for households to be connected during the pandemic. But as I look back on what’s happened during the last year, I can’t help but wonder if the broadband stats we are seeing are somehow overinflated.

First, look at the broadband performance of the two largest ISPs in the country. Comcast and Charter ended 2020 with 55.3 million broadband customers, which is around 52% of all of the broadband customers in the country. In 2020, Comcast added nearly 2 million new broadband customers. Charter added over 2.2 million. Combined, the companies grew broadband customers by 7.6% for the year. That’s an amazing growth rate when you consider that all of the other ISPs with more than 400,000 broadband customers collectively grew at a rate of less than 1% for last year. That kind of numerical anomaly always makes me curious. What would make these two companies do so well compared to the rest of the industry during a pandemic year?

Part of the difference can be explained by the continued drop in DSL customers. The telcos have been bleeding urban DSL customers for many years. Did that transition accelerate during the pandemic? For all of 2020, AT&T lost a net 5,000 broadband customers, but AT&T says it gained 1 million fiber customers. Verizon gained 173,000 broadband customers. CenturyLink lost 134,000, and Frontier lost 400,000. All of these companies except Frontier announced how well they were doing in getting customers on fiber, so perhaps a couple of million of the new cable customers come from the continued flight from DSL. But that doesn’t seem to explain away all of the differences because we didn’t see a huge migration from DSL in markets served by other big cable companies like Cox and Altice.

I can think of two trends that would have increased broadband connections faster than normal last year, both related to the pandemic. First are homes that added broadband so that people could work from home. It makes sense that some homes purchased a first home broadband connection for this purpose. There is a general belief that many people will continue to work from home, at least part-time, so some percentage of these connections might be permanent.

The second big trend is getting broadband into homes with students during the pandemic. Both giant cable companies had numerous press releases about how they were helping to keep students connected during the pandemic. It’s great that these two companies stepped up during the pandemic. Millions of homes got low-cost broadband for students when it was most needed, and the companies should rightfully be praised for the effort they made.

Nut I’ve been speculating for several months that we’ll see a drop-off in broadband subscriptions this fall when students return to live classes. It seems likely that many of the homes that added broadband for students will drop broadband, especially if the prices are no longer subsidized.

However, just as the pandemic is starting to slow down, we have a new government program that will boost low-income broadband connections. Congress has handed the FCC $3.2 billion for the Emergency Broadband Benefit program. This program will pay up to $50 per month towards a broadband bill for qualifying households. I have to think that Comcast and Charter, with 52% of the broadband market, are going to grab a lot of this funding. For now, this program has a finite end when the money runs out. But until it does, the broadband statistics from the biggest ISPs are likely to remain inflated due to this plan. That probably means we wait until next year to see if some percentage of the 2020 customer increase was a bubble or permanent. If it was a bubble we’ll no doubt see gloom and doom headlines about the broadband industry if subscriber numbers drop, when instead we’d just be seeing the last remnants of the pandemic pass through the industry.

This Time it’s Verizon

We occasionally get a good reminder of how poorly the large duopoly ISPs in the country treat their customers. The latest example comes from Verizon. The Washington Post reported how Verizon has been requiring customers to upgrade to more expensive data plans before being able to benefit from the $50 monthly discount offered by the Emergency Broadband Benefit (EBB) program.

The EBB program was funded with $3.2 billion from the $1.9 trillion American Rescue Plan Act. The program provides up to a $50 discount off broadband bills to qualifying households. A household qualifies by being low-income and eligible for the FCC Lifeline program or has suffered a significant drop in household income due to the pandemic.

The EBB is a temporary plan that will last until the $3.2 billion is gone, at which point participants lose the $50 discount. We won’t know how long that will be until we get a count of the participants in the program, but I’ve seen several estimates that this might last for six to nine months.

The Washington Post article cites customers who called Verizon to ask for the EBB discount and were told that they could only get the discount if they first upgrade to a more expensive package. As an example, somebody paying $60 per month will be required to upgrade to $90 per month to get the discount. This likely means that over the next year that people who get the EBB discount will probably end up spending as much or more with Verizon than if they didn’t get the EBB discount.

As the article points out, this is likely not illegal on Verizon’s part, but it goes completely against the purpose of the EBB, which is to help out households who have had hard economic times during the pandemic. Many of the subscribers asking for the discount will be the same ones who have had trouble paying rent after losing a job due to the pandemic.

There is a huge list of ways that big duopoly ISPs have mistreated customers over the years, but this particular case might be the posterchild of ISP abuse. Somebody at the company figured out a way to gain a longer-term advantage for the company as part of a pandemic relief program.

I’m also willing to bet that this will turn out to be a story of how monopoly abuses come about. It’s extremely unlikely that an idea like this started in the Verizon Boardroom. Instead, I bet that somebody down the management chain saw this as an opportunity to increase bonuses as a reward for achieving a bunch of upsells. If that sounds familiar, it is exactly what happened at Well Fargo Bank a few years ago when employees were opening extra accounts for customers as a way to make higher bonuses.

This is how monopoly abuses occur – not from the boardroom, but from employees that take advantage of the market power of the company for personal gain. That personal gain could be in the form of bonuses, or maybe just in getting recognized to gain promotions. That’s the only way to explain away some of the amazing stories that have come out over the years from Comcast customer service. The chances are high that the folks who thought up this idea are in hot water at Verizon – not because they were doing the wrong thing, but because they were dumb enough to get caught and drive a story to the front page of the Washington Post.

These periodic headlines detailing monopoly behavior are always a good reminder for smaller ISPs to be careful because employees at smaller ISPs can undertake similar behavior if they see a personal gain from cheating. Most of my clients have eliminated the temptation for shenanigans by having simple products that are always at the same prices for customers. But when an ISP is willing to negotiate rates with customers there is too much chance of bad behavior by employees.

New NTIA Grants

This is the year for unusual and unexpected broadband grant opportunities. The NTIA released a Notice of Funding Opportunity (NOFO) on May 21 for a broadband grant program it is labeling as the Broadband Infrastructure Program. The NTIA will be awarding grants for up to $288 million, with the funding provided from the $1.9 trillion American Rescue Plan Act.

This is an unusual grant program because the money is aimed entirely at public-private partnerships (PPPs). The applications must be submitted by a government entity, but the specific partner must be identified that will operate the broadband business. I can’t think of another grant program in the past that even favored PPPs, let alone one that is only available to PPPs. It’s going to be interesting to see if there are enough rural PPPs in existence to use all of the money.

The grants are holding to the firm definition that the money can only be used in places where speeds are less than 25/3 Mbps. This creates a huge dilemma if the NTIA is going to stick to the lousy FCC mapping data that incorrectly shows huge swaths of rural areas as having 25/3 Mbps broadband. One would hope that the NTIA will be open to accepting evidence that actual speeds are often far slower than what has been claimed by some telcos and WISPs. If not, it’s going to be hard to find rural areas that weren’t already covered by the RDOF grants.

The grants are like all current federal broadband grants and can’t be used where prior grants have already been awarded to an area but are not yet constructed. That’s going to create an interesting dilemma for some communities. There are some RDOF grant areas that are being heavily disputed, and which may not get awarded. The FCC also awarded grants to Viasat in last year’s incentive reverse auction and communities are rightfully upset that these places are not eligible to get fiber. There is growing concern about the pending RDOF awards made to Starlink.

The grants must propose an engineered business plan. The NTIA expects the engineering to be solid because they expect projects to be built within one year of grant awards. The NTIA can grant a one-year extension for construction in some circumstances. But this rapid construction expectation means the NTIA only wants to see applications from ‘shovel-ready’ projects. Any community thinking of pursuing these grants should be forming the needed partnership immediately.

The grant applications are due by August 17. The NTIA doesn’t expect to start making grant awards until at least November 29. The NOFO offers that the NTIA might award additional funding to approved projects if there is not enough demand for the funding.

The NTIA warns that it will likely not award money to small projects, and it expects awards to be between $5 million and $30 million. That’s understandable when you consider that the agency is going to have to process a lot of the grant requests quickly between August and November. Applicants would be wise to apply early.

While there is no statutory reason that NTIA cannot award 100% grants, they caution applicants that they will favor projects that contribute matching funds of 10% or more – the NTIA wants to see the commercial partners have some skin in the game. They also want these matching funds to be non-federal dollars, meaning the matching shouldn’t come from some other bucket of funding from the $1.9 trillion ARPA program.

This is probably the most unique federal broadband grant I can remember in that the funding is only available to public-private partnerships and no other business structure. Since the grants are only being awarded to the public member of the partnership, this also implies ownership of the network by local governments and some sort of ongoing participation in the business. It’s going to be interesting to see how partnerships are created to meet these grant requirements.

Fast Polymer Cables

Scientists and engineers are always looking for ways to speed up and more efficiently configure computing devices to maximize the flow of data. There are a lot of applications today that require the exchange of huge volumes of data in real-time.

MIT scientists have created a hair-like plastic polymer cable that can transmit data ten times faster than copper USB cables. The scientists recently reported speeds on the new cables in excess of 100 gigabits per second. The new fibers mimic the best characteristics of copper cable in that electronic signals can be conveyed directly from device to device.

https://spectrum.ieee.org/tech-talk/computing/networks/plastic-polymer-cables-that-rival-fiber-optics

Another interesting characteristic of polymer cables is that it’s possible to measure the flow of electrons through each cable from outside – something that is impossible to do with fiber optic cables. This is a key function that can be used to direct the flow of data at the chip level in fast computing devices.

If brought to market, these cables will solve several problems in the computing industry. The new fibers mimic the best feature of copper wires like USB cables are easily compatible with computer chips and other network devices. A copper Ethernet cable can connect two devices directly with no need to reformat data. Fiber cables are much faster than copper but require an intermediate device to convert light signals back into electronic signals at each device.

There are immediate uses for faster cables in applications like data centers, self-driving cars, manufacturing robots, and devices in space. The new cables would be a benefit anywhere that large amounts of data need to be transferred in real-time from device to device. Since the polymer fibers are thin, they could also this also be used to speed up data transfer between chips within devices.

The data transmission rates on the polymer cables are currently at 100 gigabits per second for a distance of around 30 centimeters. The MIT scientists believe they will be able to goose speeds to as much as a terabit while increasing transmission distances to a meter and beyond.

There is a long way to go to move a new technology from laboratory to production. There would first need to be industry standards developed and agreed upon by the iEEE. Using new kinds of cables means changing the interface into devices and chips. There are also the challenges of mass manufacturing the new cables and of integrated them into the existing supply chain.

I’m always amazed at how modern science seems to always find solutions when we need them. We are just now starting to routinely use computer applications like AI that rely on quickly moving huge amounts of data. Just a decade ago nobody would have been talking about chips that needed anything close to 100 gigabits of input or output. It’s easy to assume that computing devices somehow get faster when chips are made faster, but these new cables act as a reminder that there are numerous components required in making faster computing. Fast chips do not good if we can’t get data in and out of the chip fast enough.

Comcast Tests DOCSIS 4.0

Comcast recently conducted its first test of the DOCSIS 4.0 technology and achieved a symmetrical 4-gigabit connection. The test was enabled by a DOCSIS 4.0 chip from Broadcom. The DOCSIS 4.0 standard was released in March 2020 and this is the first test of the new standard. The DOCSIS 4.0 standard allows for a theoretical transmission of 10 Gbps downstream and 6 Gbps upstream – this first test achieved an impressive percentage of the capability of the standard.

Don’t expect this test to mean that cable companies will be offering fast symmetrical broadband any time soon. There is a long way to go from the first lab test to a product deployed in the field. Lab scientists will first work to perfect the DOCSIS 4.0 chip based upon whatever they found during the trial. It typically takes most of a year to create a new chip and it would not be surprising for Comcast to first spend several years and a few iterations to solidify the chip design. Assuming Comcast or some cable company is ready to buy a significant quantity of the new chips, it would be put into the product design cycle at a manufacturer to be integrated into the CMTS core and home cable modems.

That’s the point when cable companies will face to tough choice of pursuing the new standard. When the new technology was announced last year, most of the CTOs of the big cable companies were quoted that they didn’t foresee the implementation of the new standard for at least a decade. This is understandable since the cable companies recently made the expensive upgrade to DOCSIS 3.1.

An upgrade to DOCSIS 4.0 isn’t going to be cheap. It first means the replacement of all existing electronics in a rip-and-replace upgrade. That includes cable modems at every customer premise. DOCSIS 4.0 will require network capacity to be increased to at least 1.2 GHz. This likely means replacement of power taps and network amplifiers throughout the outside plant network.

There is also the bigger issue that the copper plant in cable networks is aging in the same manner as telco copper. There are already portions of many cable company networks that underperform today. Increasing the overall bandwidth of the network might result in the need for a lot of copper replacement. And that is going to create a pause for cable company management. While the upgrade to DOCSIS 3.1 was expensive, it’s going to cost more to upgrade again to DOCSIS 4.0. At what point does it make sense to upgrade instead to fiber rather than tackle another costly upgrade on an aging copper network?

There is then the market issue. The cable companies are enjoying an unprecedented monopoly position. Comcast and Charter together have over half of all broadband customers in the country. While there are households that are unhappy with Comcast or Charter broadband, most don’t have any competitive alternative. The FCC statistics and the screwball websites that claim that Americans have multiple broadband choices are all fiction. For the average urban or suburban family, the only option for functional broadband is the cable company.

This market power means that the cable companies are not going to rush into making upgrades to offer greater speeds just because the technology exists. Monopolists are always slow to introduce technology upgrades. Instead, the cable companies are likely to continue to increase customer speeds across the board. Both Charter and Comcast did this recently and increased download speeds (or at least the download speed they are marketing).

I expect that the early predictions that it would be a decade before we see widespread DOCSIS 4.0 were probably pretty prescient. However, a continued clamor for faster upload speeds or rapid deployment of fiber by competitors could always move up the time when the cable companies have to upgrade to DOCSIS 4.0 or fiber. But don’t let headlines like this make you think this is coming soon.

Satellite Broadband Heating Up

There is a lot of news recently about low-orbit satellite broadband. There is recent news concerning the three primary companies that will be vying in the space.

First is Jeff Bezos Project Kuiper, which is still likely to get a brand name at some point. Project Kuiper has contracted with United Launch Alliance, a joint Boeing-Lockheed Martin venture, for the first nine broadband rocket launches. It’s been speculated that these launches will carry something under 500 satellites into orbit – including the company’s first test satellites.

Project Kuiper has plans to launch 3,236 satellites and the company says it will need 578 satellites to begin offering limited service. The agreement with the FCC would require the company to launch half of the total satellites before 2026, although it appears the company intends to get to that number sooner. There have been no announced dates for the nine launches, but one would think they’ll start this year.

Project Kuiper is taking a different strategy than Starlink and is launching larger, and more capable satellites rather than swarms of cheaper disposable satellites. It will be interesting to see what this difference means in terms of customer coverage and bandwidth. The company has already been funded with $10 billion from Jeff Bezos and it seems likely the company will eventually do what’s been announced.

OneWeb is also back in the news. Eutelsat, one of the world’s largest operators of satellites has invested in a 24% stake in the company. This adds to the existing ownership by the UK government and Bharti Global, a large cellular carrier in India.

OneWeb is taking a third path and plans to launch 648 satellites which are larger and are basically floating data centers. The company recently launched 36 satellites, bringing it to a total of 182 satellites in orbit. The company says it will be able to start serving the UK, Alaska, northern Europe, Greenland, Iceland, and northern Canada after two more launches and plans to be able to serve the whole planet by the end of 2022. It’s no longer clear after the change of ownership if the company will support residential broadband or will pursue connectivity for larger users like cellular towers and corporate users.

Starlink and SpaceX are all over the news. Starlink is now taking $99 deposits from prospective customers and quickly picked up 500,000 prospective customers and a quick $50 million. There is no guarantee that any customer will be able to receive service. Starlink now has 1,300 satellites in orbit and says it will begin offering retail service by the end of 2021.

Starlink download speeds in beta tests still show a range between 50 Mbps and 150 Mbps – a great upgrade for customers using rural DSL or cellular hotspots. Elon Musk continues to say that by next year that broadband speeds will approach 300 Mbps, something that is doubted by a number of industry engineers who question the ability of the constellation to handle a significant number of customers. But there are also problems becoming apparent during the beta test. A recent article in Verge claims that Starlink can’t handle trees or impediments that block the horizon. That’s not promising for serving homes in the woods or mountains – like in my area of western North Carolina.

Starlink also won a recent regulatory battle at the FCC. The ruling is extremely technical, but the gist is that Starlink will be able to deploy some satellites in a lower orbit which will allow a lower elevation angle for customer receivers, which ought to increase the speeds somewhat. But Starlink faces another upcoming battle over the spectrum that is used by the satellite fleet to backhaul traffic to and from the earth. The battle is over spectrum between 12.2 – 12.7 GHz, which is primarily owned by Dish Networks. Dish wants to use this spectrum for terrestrial 5G, and this would greatly curtail Starlink’s backhaul capabilities. The FCC ruling warned Starlink that it may not get access to the spectrum.

Within a year or two, a lot of the hype concerning satellite broadband will be behind us as we start seeing commercial satellite fleets in operation. Of particular interest to those watching this space will be if Starlink can achieve the broadband speeds being touted once the network is under full customer load. By the looks of the applications pouring into the company, it won’t take long to find out.

Mandated Low-Income Prices

ISPs are vehemently opposed to a law recently passed in New York that requires ISPs to offer low-income broadband for prices of $15 to $20. The is a dreadful law for several reasons which I’ll discuss below. But before that, I have to first make fun of the main argument the ISPs are using to fight against the law.

An array of ISP trade associations opposing the law include the New York State Telecommunications Association, CTIA, ACA Connects, USTelecom, NTCA, and SBCA. It seems that the best argument these trade associations can muster for why this law should be invalid is that the State doesn’t have the right to impose this law because only the FCC has rate authority over broadband. The problem with that argument is that the FCC doesn’t have rate authority – or regulatory authority of almost any kind over broadband – due to heavy lobbying by these same trade associations that was able to kill Title II regulation. The group is pointing to a ruling made by FCC Commissioner Pai that says that States have no authority over broadband rates – made at the same time that Chairman Pai killed FCC authority to regulate rates. Chairman Pai’s position was that nobody should have rate authority over ISPs – which means nothing after the FCC walked away and created a regulatory vacuum. States are generally able to fill a vacuum, much like California was able to implement net neutrality after the FCC killed its own regulations on the issue.

There is a long list of reasons why ISPs hate this particular regulation. The first reason is practical. Ignore for a minute that somebody like Comcast might be able to afford this and consider instead a smaller independent fiber overbuilder. It costs on average around $1,000 to install a new customer on fiber just to build the fiber drop and install the needed electronics. It would take 50 months at a $20 monthly fee for an ISP to recover just the cost of the installation – and that still hasn’t paid a penny towards the cost of operating the business. If this law would require a small ISP to install 1,000 subsidized customers (not a big number), the ISP would have to spend $1 million (that it probably doesn’t have) and then be repaid at $20,000 per month. Anybody that knows the small ISP world knows that this scenario is untenable, and the law could sink a fledgling ISP and put them out of business. Small ISPs would have no choice but to ignore this law – the penalties for non-compliance would likely be less costly than the cost of compliance.

Localities have complained about broadband redlining for years. If New York makes ISPs wholly shoulder the cost burden to make broadband affordable, then the state will never see any more broadband investment in poor neighborhoods. ISPs will never come to areas with a substantial percentage of low-income households and the ISPs who are already in such neighborhoods will start figuring out ways to walk away from these neighborhoods. Put this law into place and in twenty years there will be true broadband deserts across New York with large neighborhoods that have no broadband options other than smartphones. Whoever wrote this law didn’t spend much energy thinking about the practical consequences.

The ISPs have an even larger concern with this law. If the State can set a rate for low-income broadband, then there is likely no reason that the State can’t also put a cap on broadband bills. What’s to stop the state from declaring that gigabit broadband can’t cost more than $50?

Even should a State decide to tackle the issue of rate regulation, it ought to do so in a deliberate and reasonable manner by weighing all sides of the rate regulation issue. Legislators are the worst possible group of folks to create this kind of regulation because they will always take the populist view without worrying about the consequences. If New York starts to regulate rates through ad hoc regulation as suggested by this bill, my advice to ISPs would be to never spend another penny in the state and take investments elsewhere. I’m trying to imagine what would happen in New York if Verizon and Charter decided to not invest another dollar. In ten years, the State might have the worst broadband networks in the country.

There is nothing good about this law because it piles the full burden of helping low-income homes on ISPs. Everybody in the industry, including the ISPs understand that the country would be better off if everybody is connected to the Internet. But digital inclusion efforts are going nowhere if ISPs are expected to solely fix the problem. The right solution will require some give by ISPs along with some subsidies from the government. There has to be a middle ground that makes sense, and we’re not going to find that middle ground through legislation like this.

Interim Treasury Rules Part 3

There is one interesting sentence in the Treasury guidance rules describing how to use broadband funding from the $350 million aimed at local governments in the American Rescue Plan Act.

Treasury also encourages recipients to prioritize support for broadband networks owned, operated by, or affiliated with local governments, non-profits, and co-operatives—providers with less pressure to turn profits and with a commitment to serving entire communities.

 Note that the key verb in that statement is ‘encourages’, meaning that the statement is not actionable and doesn’t commit that funding is to used in this manner. But this is the first time that such language has ever been incorporated in anything federal related to broadband. In the past, the language was just the opposite. Past federal grants have always been aimed at existing carriers and they always threw in a footnote mentioning that the money could also be used by entities like municipalities.

This language has to be ultimately coming from the White House – I can’t think of any other reason why it’s in the document. The new administration has said that it’s in favor of using every tool in the toolbox to get broadband everywhere, including enabling municipalities to find local broadband solutions.

In 2015, the FCC under Tom Wheeler tried to expand the ability of municipalities to provide broadband. The FCC issued an order that overturned state laws in Tennessee and North Carolina that prohibited the municipal ISPs in Chattanooga, TN, and Wilson, NC from expanding broadband outside city boundaries. The FCC was testing the waters on its authority to overturn state restrictions on municipal broadband. In a rejection of the FCC’s authority to preempt states, the FCC order was overturned by the Sixth Court of Appeals in August 2016.

The FCC is in an even weaker position today to try something similar since the Ajit Pai FCC eliminated Title II regulation of broadband. An FCC that doesn’t directly regulate broadband is in no position to challenge state broadband rules. Even if the current FCC was to restore Title II authority or something similar, it’s doubtful that the agency can challenge state laws based upon the 2016 ruling.

This means that the only challenge to state restrictions on broadband will have to come from Congress. There has been talk of adding this provision to several of the bills running through the current Congress. But Congressional action on the issue would set off a set of court challenges, which are almost automatic anytime that federal government tries to usurp state laws of any kind.

I hope that any cities or counties reading the Treasury guidelines don’t think that the Treasury has granted local governments the ability to offer retail broadband. The Treasury statement is not much more than wishful thinking for municipalities in states that don’t allow municipal broadband. A municipality that is in a state that restricts municipal participation in broadband must follow the state laws when using this latest federal funding. A city can’t directly build fiber infrastructure if the state doesn’t allow it; a municipality can’t offer broadband services where that’s prohibited.

I think municipalities probably still appreciate this gesture from Treasury because it signals a change in regulatory philosophy at the federal level. But neither the White House nor any of its agencies currently have the power to grant the right to municipalities to offer broadband. It’s an interesting sentiment in this case, but not much more.