Category Archives: The Industry

Why Customers Choose FWA

It’s been interesting to watch cable companies downplaying FWA cellular wireless. For example, in September, Comcast President Mike Cavanaugh said that FWA wireless is a ‘near term’ issue that is competing for the lower end of the market. CEO Brian Roberts was quoted this year about competing against FWA saying, “Three companies are all simultaneously within a short period of time are all offering a home connectivity product by their own admission a lower speed, more easily congested network.”

And yet, the carriers selling FWA continue to sell at astounding numbers. AT&T, T-Mobile, and Verizon have consistently added 900,000 customers per quarter since the second quarter of 2022. The big cable companies have been fighting back by significantly lowering the prices of their slowest bandwidth products, and this seems to be stemming the losses due to FWA. But those lower prices come with a cost with lower margins and average revenues per customer.

I think that what has been missing from the discussion of FWA of how the technology compares to the alternatives. Consider the following table that shows average speed tests for a 12-month period in three rural counties for all broadband technologies. There is nothing unusual about these counties – they are just three places where I happened to recently do some analysis. Each county has a county seat and a few towns with cable broadband and some fiber, but rural areas are largely still not served with any fast broadband.   It’s not hard to understand why FWA is so competitive in rural counties. It’s generally faster than fixed wireless and Starlink, and with generally a lower prices. FWA is being priced at roughly the same level as DSL in many markets.

The big limiting factor for FWA in rural markets is the broadband footprint and good speeds like those shown in the above table are only available within a few miles of cell towers. In all three counties, the FWA providers cover only roughly one fourth of the geographic footprint of the county.

Comparative speed tests are always interesting. Each county is served by a different cable company, and yet each is delivering almost the same average speeds – likely because each is operating similar DOCSIS 3.1 networks.

There are some noticeable difference in these counties that require local knowledge to explain. For example, fiber speed tests are lower in County 2 due to a fiber provider that offers a very affordable 100 Mbps fiber product that pulls down the average speed. FWA speeds are also slower in County 2 due to households still using cellular hotspots from cell sites that haven’t been upgraded to FWA.

Starlink speeds are consistent with the national average numbers I’ve been seeing – but Starlink has the most erratic variance in speed tests with a range of tests between a few Mbps and several hundred Mbps. In the three counties, the speeds on fixed wireless (from WISPs) are relatively slow since the WISPs have not upgraded to faster radios. There are counties where WISP speeds are much faster. The speed that might surprise some folks is DSL. As the copper networks have emptied of customers, the remaining customers are seeing faster speeds than just a few years earlier.

Grant Funding for Government-Owned Networks

The State of New York recently awarded $140 million in grants to support publicly-owned open-access networks. These projects will cover more than 60,000 homes with broadband.

The funded projects include $30 million for fiber to pass 4,000 locations in Schoharie County, $26 million for fiber to pass 6,600 locations in Cayuga and Cortland Counties, $30 million for fiber to pass 14,000 locations in the City of Jamestown, $30 million for fiber and a wireless tower to pass 22,000 locations in Sullivan County, $13 million for fiber and fixed wireless to pass 1,500 locations in Franklin County, and $11 million for fixed wireless to serve 11,000 locations in Orleans County.

The grant funding comes from $228 million provided to New York by the Capital Projects Fund that was created by the American Rescue Plan. That program was administered by the U.S. Treasury and sent money directly to states to make broadband and related infrastructure grants. The grants are specifically being awarded by New York’s ConnectALL’s Municipal Infrastructure Program. These new grants will construct 1,200 miles of fiber and wireless assets to cover 60,000 locations with broadband. This program, in total, has funded over 2,000 miles of fiber that covers 87,000 locations.

The projects are all being touted as public-private partnerships because the local governments will own the infrastructure, and ISPs will provide service to customers. This arrangement is generally described as open-access, because the goal of the network owner is to bring multiple ISPs to a market. In an open-access network, no ISP has a technical advantage since they share a network and ISPs must compete for customers with price and service.

Every federal infrastructure grant I’ve worked with lists municipalities as grant candidates. Some, like BEAD, even stress the goal of making grants to local governments. And yet, broadband infrastructure grants to local governments are rare.

There are several reasons for this. The primary issue might be the timeline. Unless a local government has already done all of the homework, planning, and engineering to know if a grant makes sense and if a business plan is feasible, then it’s hard to submit a grant in what is usually a fairly short time window. A second reason is local governments typically use bond funding to pay for any matching requirements. Bond funding can be slow to obtain and complicated, particularly for non-traditional uses like funding a broadband network. It’s also hard to sync the timing of grants and bond issues since bonds are not guaranteed until the day the bonds are sold, while bank loans can be arranged contingent on winning a grant.

Probably the main reason why it’s hard for a municipality to get a broadband grant is that grant scoring rules generally heavily reward past performance as an ISP. It’s not easy for any startup to get grant funding, not just a municipal startup. Most grant rules are heavily biased towards not making a bad grant award since there are past horror stories of grants made to entities that weren’t capable of fulfilling the grant.

New York eliminated most of these roadblocks. This particular grant program is aimed specifically at local governments. The grant program gave local governments plenty of time to find ISP partners and plan for the grant filing.

These particular grants are interesting because, while the grants are made to local governments, in the long-term, most of the profits on an open-access networks go the ISPs that operate on the networks. You would think it would be harder for opponents of municipal broadband to oppose projects that ultimately benefit for-profit ISPs – but many even oppose these public-private partnerships.

Starlink in the News

There is a lot of speculation that Starlink is positioned to get a lot more federal subsidy from the BEAD grant program. There are a few things that have to happen for that to come to pass, but that is not the only news about Starlink these days.

Starlink announced in September that it reached four million customers worldwide. What is most impressive about that announcement is the rate of growth, with the company just hitting the three million customer mark in May of 2024. The company served two million customers at the end of 2022, so the rate of growth is on a steep upward curve. The company currently has over 6,700 working satellites, up from 5,400 at the end of 2023. The company still plans to grow the first-generation constellation to 12,000 satellites. This growth puts the company on track to hit $6.6 billion in revenue for the year, which means the company will be able to internally fund its continued efforts to improve its reusable rockets.

At the end of November, the FCC’s Space Bureau granted the SpaceX application to construct, deploy, and operate a constellation of the next generation of satellites, which the company is calling its Gen2 Starlink constellation. The FCC authorized SpaceX to operate the Gen2 satellites at altitudes of 340, 345, 350, and 360 kilometers. The FCC also gave permission to connect to the new satellites using Ku-, Ka-, E-, and V-band frequencies. Starlink says it still hopes to eventually reach 29,988 Gen2 satellites, up from the 7,500 approved by the FCC so far.

In the application that requested the changes, Starlink said that “small-but-meaningful updates” can boost broadband speeds to 1 Gbps. Starlink’s website says that current actual speeds vary between 25 Mbps and 220 Mbps, with a majority of users experiencing speeds over 100 Mbps. Current upload speeds are typically between 5 and 20 Mbps. Current latency ranges between 25 and 60 milliseconds, with 100+ milliseconds in a few remote locations.

The company says the faster speeds would come from several changes. The company plans to use larger Starlink satellites, which are so big that only Starship can launch them. SpaceX has also requested to lower the altitudes of existing satellites, requesting to lower satellites at 525, 530, 535 kilometers to 475, 480, and 485 kilometers, respectively.” Another request to the FCC would change the elevation angles for satellites operating between 400 and 500 kilometers from 25 degrees to 20. This would increase the connection times to earth stations.

Finally, the FCC approved the request by SpaceX and T-Mobile to offer supplemental mobile coverage from space. The FCC ruling said this “will put an end to mobile dead zones.” For now, the approval extends to basic connection for texts or emergency communications using slices of T-Mobile’s spectrum in areas the mobile carrier’s terrestrial network can’t reach.

There is no guarantee that the FCC will approve everything Starlink is asking for. Companies like AT&T and Verizon have said that some of Starlink’s requests for frequency would interfere with and degrade service from terrestrial mobile networks.

One bit of negative news is that Starlink has reintroduced a wait list in some markets. Is this something that will be relieved as more satellites are launched, or will the rapid customer growth outstrip network capacity. This is the same kind of issue that every ISP that grows quickly faces.

Elon Musk’s close ties to the administration bodes well for the company to get much of its wish list. The company had a very good 2024, and the future looks even brighter.

My Predictions for 2025

Disruption of Federal Grants. It seems almost inevitable that Congress is going to pull back some or all future funding for the Digital Equity Grants that are part of the BEAD program. Some in Congress are already warning the NTIA not to award the current grants that are under review.

The BEAD grant program will change. Congress is likely to reverse some of the BEAD provisions that Senators have been complaining about, like the BEAD requirement that requires a low-rate option. However, BEAD is ultimately designed to be a state grant program, and a lot of States are going to fight hard against trying to direct funds away from fiber. With that said, it’s likely that a lot more BEAD funding will go to satellite than earlier estimates. I predict that the change of administration and a swap out of folks at NTIA is going to result in at least a six month delay in the grant process.

The FCC Will Stay the Course. The new FCC will not change the agency as much as you might expect from a change in administration. New Chairman Carr will act quickly to reverse the current FCC rulings on net neutrality and discrimination. But otherwise, there won’t be a lot of revisiting of other recent decisions. Assuming that Chairman Carr will tackle what he addressed in Project 2025, the FCC will spend a lot of energy trying to free up new 5G spectrum and investigating issues associated with Section 230 and content moderation.

Job Well Done? I predict at some point that the FCC and/or the NTIA will declare at some point this year that the rural broadband problem has largely been solved, relieving the federal government of any obligation to fund any more broadband infrastructure.

FWA Will have Another Strong Year. Some industry analysts have written off FWA cellular broadband as a temporary flash in the pan. I predict that T-Mobile, Verizon, and AT&T will continue to collectively add 900,000+ customers per quarter again this year. Any increased inflation in the economy will drive the FWA numbers even higher.

Universal Service Fund Will Change. I predict that the Supreme Court is going to rule that Congress erred when it gave the FCC the authority to operate the USF and to establish fees to fund it. That’s going to force Congress to scramble to revamp the popular program. Congress will be forced to fix the funding issues. I predict Congress will create a tax that will be charged against a larger base that includes large users of the Internet like Google, Microsoft, Meta, and others. All of the changes to USF will probably mean that the launch of the $9 billion 5G Fund for Rural America will be delayed or shelved for the year.

The Mad Scramble to Buy Fiber Businesses Will Continue. There is still a glut of investment capital looking for a place to land, and a lot of that money is going to be aimed at buying existing fiber-based ISPs.

RDOF Troubles. I don’t think we’re done with RDOF defaults. This might be further exacerbated by any movement by the administration to claw back RDOF funding that hasn’t resulted in infrastructure.

Cable Companies Will Tame Their Losses. While large cable companies will continue to collectively lose customers, the rate of losses will slow as the companies focus on holding their market share. The large cable companies collectively lost 265,000 customers in the third quarter of this year. However, Comcast and Charter both said they would have had small gains except for the one-time losses due to the end of ACP.

FWA Wins 3Q 2024

The table below shows the broadband additions for the publicly traded ISPs in the country in the third quarter of this year. Missing from this list are large ISPs that don’t publish customer counts, like Cox,  Mediacom, Windstream, Brightspeed, and Google Fiber. However, this list represents roughly 90% of the broadband customers in the country.

To the dismay of other ISPs, the three FWA cellular carriers continued to perform well and collectively picked up 913,000 net customers in the quarter, just 20,000 fewer than in the second quarter.

The big cable companies continue to lose customers. However, both Comcast and Charter reported that they would have had small customer gains for the quarter except for losses due to the end of ACP. We’re liable to hear more about the impact of ACP with the year-end customer numbers.

Telcos and fiber overbuilders on the list had a collective small overall gain of 39,000 customers. That number disguises the continued fast growth of fiber customers and the continued loss of DSL customers. Lumen continues to underperform the rest of the telcos and lost 57,000 broadband customers in the quarter.  Frontier continues to be the fastest growing of the large telcos with 1.6% customer growth in the quarter.

Also not included on this list are the fiber networks of Lumos and Metronet. T-Mobile bought a partnership share of these two companies and it’s not clear if we’re going to get details of their future growth.

 

Broadband Usage 3Q 2024

OpenVault recently published its Broadband Insights Report for the end of the third quarter of 2024. OpenVault is documenting the continued growth in broadband usage by U.S. households.

I think one of the most useful statistics from OpenVault is the average monthly broadband usage per customers in gigabytes. Below is the trend in average monthly U.S. download and upload volumes since the third quarter of 2020. These are the average amount of data used combined for all residential and small business customers. Over the last four years the average monthly download and upload usage has increased roughly 53%.

The average U.S. broadband customer used 35 more downloaded gigabytes and 5 uploaded gigabits per month than a year earlier. This means continued pressure on broadband networks because if we assume roughly 120 million broadband subscribers, this equates to over 4.8 billion more gigabytes of data used each month than a year earlier.

This table raises an interesting question if the growth in broadband usage is slowing down. It would be rash to draw that conclusion by only comparing the third quarters of 2023 and 2024, but a slowdown would be obvious over the next several quarters.

The growth in what OpenVault calls power users is even more dramatic than overall bandwidth growth. Below is the percentage of broadband customers who use more than 1 terabyte of data per month and those using more than 2 terabytes. These numbers show the potential harm created when ISPs place data caps on monthly usage.OpenVault always includes other interesting statistics in its quarterly reports:

  • The report shows that the average rural customer uses only slightly less average broadband than urban households.
  • Median household broadband usage was 389.3 gigabytes – half of homes use more broadband than the median, and half use less. The higher overall average is explained by the large number of power users.

The report includes a section that shows there is no longer a strong correlation between household incomes and data consumption as was seen in past years. Interestingly, in the third quarter, the group with the highest average download usage was households with incomes under $50,000.

 

Ten Years Ago

As a country we have a very short memory when it comes to broadband. Anybody in their late 40s or older clearly remembers twenty-five years ago when the predominant form of broadband was dial-up and we were seeing the first DSL and cable modem trials in the market. But I think most people have forgotten about the state of broadband just ten years ago.

I admit I have the same time bias. I was cleaning out some old bookmarks, and I ran across several articles from ten years that surprised me.  The big difference between 2014 and now is broadband speeds. Consider the following map that was published in the Washington Post in early 2014. This shows the average broadband download speed in the U.S. was 18.2 Mbps. The maps shows the parts of the country where speeds were faster or slower than average.

To put these speeds on this map into context, the FCC adopted the updated definition of broadband of 25/3 Mbps in 2015 – updated from 4/1 Mbps. At the time, there were critics who said that 25/3 Mbps was a ridiculously high definition, which can be understood when seeing the large parts of the country on this map where average speeds were under 10 Mbps. The map is an interesting way to see where fiber had already been widely built in 2014, mostly in Verizon FiOS markets in the Northeast and some of the PUDs in Washington. Many other smaller areas had fiber, but are not large enough to be seen on a map at this scale.

In 2014, the national broadband penetration rate was around 80% of households, up from 20% of homes ten years earlier. 2014 is the year when the 18-34 age group spent more time online than watching television (the first this had ever happened for any age group).

Video streaming was becoming a big deal, and Netflix had 48 million customers by the end of 2014. In looking back at blogs and articles, there was widespread complaints about pixelated video streams. Livestreaming was still in its infancy and the Superbowl was livestreamed in 2014 by NBC along with a Verizon mobile app.

Video conferencing was also in its infancy, and before 2014 most videoconferencing required specialized hardware. Skype became the predominant software-based videoconferencing platform, in 2014, but was mostly used for business meetings. Apple users had been using FaceTime for one-on-one video conferencing since 2010.

Ten years ago is also when the cable company first started to lose customers. The year ending March 31, 2013 was the first 12-month period where the big cable companies collectively lost net cable TV customers. At the end of 2013 there were over 8 million homes that had cut the cord, but the cable companies had continued to add new customers to replace those who were leaving.

Apple sold 40% of cellphones in 2014 and Samsung had a 21% market share. BlackBerry still held a 5% market share.

How many of you remember the broadband speed you were buying in 2014? In 2013 I was served by a WISP that delivered about 5 Mbps. In 2014 I moved and subscribed to Comcast, but I can’t recall the speed they provided – my fuzzy memory says it was 30 Mbps.

The bottom line from this look back is a recognition of the extraordinary strides we’ve made with broadband speeds in just ten years. Cable companies now routinely offer gigabit speeds and millions of new fiber passings are being built every year. FWA cellular is bringing pockets of 100+ Mbps all around the country. New fixed wireless radios now have big bandwidth capabilities. I don’t think anybody in 2014 could have predicted where we are today with broadband.

Stealing Copper

I recently saw that AT&T is offering a $10,000 reward to anybody who provides information that leads to the arrest and conviction of people stealing copper wiring. The particular announcement is related to a recent theft of copper in South Dallas, Texas – but there are numerous other thefts.

This is not a small problem and the estimated value of stolen telephone copper in between $1.5 and $2 billion annually. This isn’t just a telecom issue, and the U.S. Department of Energy estimates that theft of copper electrical wiring costs around $1 billion per year.

The theft of copper telephone and electric wires is very much a function of metal prices, and thefts increase any time metal prices climb. The current price for raw copper is $4.46 per pound. The value fluctuates with demand, like most raw materials. In September 2022, the price hit a low of $3.24 per pound.

Some states and cities have taken measures to combat metal theft, such as requiring scrap yards to keep detailed records of metal scrap purchases.

Copper theft has been a problem for many years. The National Insurance Crime Bureau said that there were 33,775 insurance claims for stolen metal between 2010 and 2012, and 96% of those thefts was of copper. At that time, the states with the most thefts were Texas, Georgia, California, and North Carolina.

In searching the web, there have been copper thefts that have caused major problems. For example, in April 2008, five tornado warning sirens in Jackson, Mississippi, did not sound after thieves had stripped the copper feeding the sirens.

I remember when I was visiting an independent telephone company in south Texas in the late 1970s, that sophisticated thieves had stolen over 20 miles of copper wiring over a weekend. The thieves were using unmarked utility trucks and reeled the copper off the poles into large spools. At that time, the price of copper was $1.20 per pound, up from just over $0.50 per pound just a few years earlier. Folks with long-term memories might recall that the spike in copper prices in the late 1970s also led folks to melt down a huge number of all-copper pennies.

You might think this is a diminishing problem since the use of copper is decreasing as telcos upgrade from DSL on copper to fiber. However, companies like AT&T and other telcos overlash a lot of fiber onto existing copper lines – a process that saves a lot of time and money. It’s a big problem if thieves pull down fiber to get to the supporting copper because the fiber repairs in this situation will not be quick.

Copyright Infringement – a New Worry for ISPs

In decisions that should trouble every ISP, multiple Courts have ruled that ISPs are liable if they don’t disconnect customers accused of copyright infringement.

The U.S. Court of Appeals for the 5th Circuit ruled against Grande Communications, a subsidiary of Astound Broadband. The courts sided with three record companies, Universal, Warner, and Sony that Grande had failed to disconnect customers from broadband who engaged in copyright infringement by downloading illegal copies of music. Grande might have been singled out because it had a firm policy since 2010 that it wouldn’t disconnect customers over the infringement issue. The appeals court upheld a lower court ruling that Grande is liable for copyright infringement. The appeals court said it would consider lowering the original award of $46.8 million.

This is the second major lawsuit on the issue. In 2018 the major record labels sued Cox Communications over its copyright policies. The labels accused Cox of refusing to disconnect customers who repeatedly broke copyright rules by downloading music without paying for it. In 2019, a court in Virginia found Cox liable for both contributory and vicarious copyright infringement and awarded the music labels an astounding $1 billion in damages. Cox appealed, and the Fourth Circuit U.S. Court of Appeals upheld the charge of contributory infringement but reversed the charges for vicarious infringement and vacated the $1 billion in damages.

Cox asked the Supreme Court in August to decide whether the 4th Circuit erred in deciding that an ISP can be held liable for copyright infringement without proof that the company fostered or promoted copyright infringement. The record labels are asking the Supreme Court to reinstate the original $1 billion award.

This has to concern all ISP, because if these two cases are resolved in favor of the record industry, then all ISPs are vulnerable. Altice USA, Frontier Communications, Lumen, and Verizon filed a brief with the Supreme Court saying that the 4th Circuit ruling imperils the future of the Internet by making ISPs liable for huge damages if they don’t carry out mass Internet evictions.

Folks might have a visceral reaction thinking that copyright offenders should be punished. There is some needed context to fully understand the issue. Complaints of copyright infringement are rarely made directly by record companies or others who hold copyrights, There is an entire industry of companies that makes a living by issuing takedown requests for infringements of copyrighted materials. These companies get paid by issuing huge numbers of takedown requests. Social media companies are inundated with these takedown requests every day to remove posts that link to copyrighted music, movies, and other materials.

The takedown process is completely one-sided and there is no appeal for a broadband customer who is unjustly accused of bad behavior. The music companies expect ISPs to cut off subscribers after only a few claimed violations of copyright.

This is also troubling to broadband customers. Any home with teenagers will have to worry if teens will download copies of games, movies, or music. People could hit a link on social media that downloads copyrighted material without even realizing they did something wrong. The bad behavior doesn’t even have to be done by a family member. Losing an essential broadband connection because teens, roommates, or visitors violated copyright laws seems like an extreme penalty. If ISPs start cutting customers dead for violating copyrights, I have to imagine that people are going to be a lot more cautious against giving visitors or even family members the WiFi password.

Copyright holders want ISPs to act as judge, jury, and executioner and unilaterally punish customers without a trial or hearing by taking away their Internet access. Like many other problems in the industry, the only real fix for this is to have Congress update or replace the Digital Millennium Copyright Act (DMCA), which was adopted in the 1990s when we were all still using dial-up access.

The recording industry has a legitimate gripe, but their complaint is against the folks who steal copyrighted materials. They should be required to pursue a law enforcement solution, like is done for folks who violate other laws. The solution is not to turn the intermediate ISP into a policemen on the issue. This goes down an ugly and slippery slope. What’s next, forcing ISPs to turn off broadband for customers who break myriad other laws?

Veterinary Telemedicine

One of the newest uses of broadband that many millions of us discovered during the pandemic was telemedicine visits to avoid time-consuming live visits to a doctor or therapist office. Very quietly, there has also been an explosion of veterinary telemedicine visits.

Some states have laws for veterinarians that don’t allow them to dispense medicine over the phone or telemedicine connection for a for a pet they haven’t previously seen. However, many vets are able to prescribe medicine for a pet they’ve recently seen, and some states are more open to allowing vets to prescribe medicines during an online visit.

If your vet conducts telemedicine visits, then virtual visits can be a real benefit compared to dragging a reluctant pet to the vet for a routine matter. I’m pretty certain that my cat Baxter believes that we’re trying to kill him every time he has to go into a carrier for a trip to the vet. Telemedicine visits can reduce pet and pet-owner anxiety for routine visits that don’t require an examination. Vets say that many cat owners never bring their pets to see a vet because of the anxiety created for the pet.

Since our pets can’t talk to us, pet owners are often not sure if their pet is having a problem that requires a vet visit. One interesting and popular area of online veterinary practice is tele-triaging. This is a service where a vet will answer basic questions about your pet. Is their behavior a sign of a problem? Is the houseplant they just ate poisonous? Is your pet sick enough to justify a nighttime run to an emergency pet clinic?

Vets say that they diagnosis many problems by telemedicine with the participation of the pet owner. The vet can do things like see a pet’s gums, count the respirations per minute, or watch a pet walk to see if there are any obvious problems. Vets say that they’re finding that doing visits this way has the side benefit of bringing pet owners into the process of understanding their pet better.

Veterinary telemedicine visits are marketed in a variety of ways. Some visits are analogous to human telemedicine using pay per visit. Other veterinarians sell annual plans that allow pet owners to contact them more often. And, as you might expect, there is now pet insurance that can help to offset the cost of telemedicine visits.

Another interesting aspect of using broadband for pet care is using broadband connections to talk to experts. For example, Chewy, the large online pet supply company offers customers who use their autoship program free televisits with a pet expert who can discuss a wide range of behavioral or dietary questions concerning pets.

Broadband is becoming so embedded in most people’s lives that at some point we’ll start to make a list of things that are not available online.