Manufacturing Returning to the U.S.

The other day I watched an online announcement by Nokia of a partnership with Sanmina in Pleasant Prairie, Wisconsin to rehab and expand an existing factory. The factory will create over two hundred new jobs and will manufacture fiber electronics like OLTs and ONTs that are used for fiber-to-the-premise. Vice President Kamala Harris was on hand for the announcement since the impetus to build a factory in the country was partially driven by Buy America provisions in the upcoming BEAD grants.

Nokia is not the only fiber-related manufacturer to expand production in the U.S. Corning announced the construction of a new fiber optic cable plant near Hickory, North Carolina. CommScope is building a new factory in Catawba, North Carolina.  Prysmian announced the conversion of a factory in Jackson, Tennessee from building copper cables to fiber cables.

A recent press release from the U.S. Department of the Treasury documents the big burst of investments in new factories. This is being funded, at least in part by infrastructure spending that came from the Infrastructure Investment and Jobs Act (IIJA), the Inflation Reduction Act (IRA), and the CHIPS Act.

The following chart comes from that Treasury press release and shows how 2023 spending for manufacturing facilities has doubled the average spending for 2005 – 2022. Most of the new spending is on computers, electrical, and electronic factories. The Treasury press release notes that 18 new chipmaking factories were started in the country in 2021 and 2022. But since the announcement of the CHIPs Act there are over 50 new chipmaking facilities underway.

This can only be good news for the broadband industry. First, it increases the chance to buy American electronics as part of fulfilling grants. But the real benefit is over the longer run. This means that a lot of U.S. electronics manufacturing will be able to rely on U.S. factories manned by U.S. employees.

I’m sure many of you join me in being dismayed for decades as U.S. manufacturing jobs were shifted overseas. We’ve seen a steady erosion of good-paying factory jobs and a decrease in households in the middle class.

Many of these new and repurposed factories don’t require as many new workers as older factories due to automation. But every new U.S. manufacturing job created is a win for the economy. This is a needed shot in the arm for the economy. We can’t run an economy where everybody is doing service jobs – although it looked at one time like that is where we were headed.

Who Still Has Landlines?

I’m a regular reader of the Washington Post, having started to read it as a teenager. The newspaper has a great series of articles called the Department of Data, written by Andrew Van Dam. His articles look at questions asked by readers for which there are statistics available to answer the question. He recently answered the question of who still has landline telephones. The article is behind a paywall, but here is a link.

Van Dam found the answer in the National Health Interview Survey that is conducted by the National Center for Health Statistics. It may seem odd for this government survey to contain a few questions about telephones, but over the years, the survey has shown a correlation between having a landline and overall health. According to the survey, people who cut the cord and only use cell phones are more likely to engage in risky behavior. They are more likely to binge drink, more likely to smoke, and more likely to go without health insurance. The folks who give the survey don’t know why that is – it’s just a statistical trend that has held true for many years.

Here are some of the statistics about landline telephone service and cell phone usage from the latest NCHS:

  • 27% of homes still have a landline / 73% of homes only use cell phones.
  • 2% of homes have a landline and no cell phone.
  • Only 1% of homes have no landline or cell phone.
  • 34% of homeowners still have a landline. Only 15% of renters have a landline.
  • Landline usage is correlated with age. 88% of adults between 25-29 only use a cell phone. Only 47% of those over 65 only use a cell phone.
  • Interestingly, there is not much difference in cell phone usage based on level of education.
  • There is some correlation between household income and cell phone usage. 78% of homes that are below the poverty line only use cell phones while 72% of those making twice the rate of poverty or higher use only cell phones.
  • There is virtually no difference in the percentage of homes that use only cellphones between urban and rural areas.
  • There is a big geographic difference in households that only have cell phones. Nearly 80% of homes only use cell phones in Idaho, Oklahoma, Mississippi, Wyoming, and New Mexico. The states with the lowest percentage of homes that only use cell phones are New York, Maryland, Massachusetts, and New Jersey. The whole northeast has fewer homes that rely on only cell phones than the rest of the country.
  • There is no correlation between still having a landline and having a computer or tablet in the home.

Van Dam speculates that the northeast has the highest percentage of landlines because Verizon built FiOS fiber networks back before the giant drop in landline subscriptions. He thinks it’s likely that people who have used Verizon FiOS for a long time have never bothered to drop the landline service.

I always find it interesting when ISPs choose to offer broadband and no telephone service. It’s really easy these days to layer on VoIP service, and it’s an easy margin with little headaches. I think many ISPs will be surprised to find that over one-fourth of homes still are willing to pay for a landline.

Who’s In Charge of Broadband?

On July 24, the FCC authorized a new subsidy program, Enhanced A-CAM (Alternate Connect America Cost Model). This program will extend subsidies to small, regulated telephone companies at a cost of about $1.27 billion per year for ten years. The subsidy will be paid from the FCC’s Universal Service Fund.

The funding requires recipients to deploy voice plus broadband with speeds of at least 100/20 Mbps to 100% of the areas covered by the subsidy within four years. The order is technology neutral, so telcos could elect to meet this requirement with fiber or with licensed fixed wireless technology.

According to Mike Conlow, this order will bring broadband to almost 583,000 unserved or underserved locations that are already covered by the NTIA’s BEAD grant footprint. Today’s blog talks about the absurdity of the FCC making this announcement only weeks after the NTIA announced the distribution of the $42.5 billion in BEAD funds to states. This means that two U.S. agencies both announced funding to cover the identical half-million locations within a month of each other.

Think about what this means. A state that has some of these A-CAM locations was allocated BEAD grant money to bring broadband to these areas. The FCC order is then directly funding to build broadband to the same passings. This means that a state that has a lot of unserved and underserved A-CAM passings is getting a funding windfall. Conlow estimated that this double funding is bringing a funding windfall of $180 million to Nebraska – the state with the most unserved and underserved A-CAM locations. The downside of this is that if Nebraska and other states are getting a windfall from the FCC decision, then other states are receiving less BEAD funding than they would have if these locations had been excluded from BEAD before the NTIA allocated the $42.5 billion.

The FCC’s A-CAM order was released only three weeks after the NTIA announced the BEAD allocations to states. There is no way that the FCC didn’t do this deliberately. The FCC could have asked the NTIA to take these locations out of the BEAD process so that the $42.5 billion would have been allocated fairly.

Two years ago, the Biden administration directed the FCC, the NTIA, and the USDA to coordinate everything associated with federal funding for broadband. The FCC’s actions with this decision are the exact opposite of coordination.

I speculate that the FCC did this to reclaim relevance in the discussion of who is helping America solve the rural broadband gap. The FCC has taken a lot of criticism in recent years for botching the RDOF funding process and handing out wasted billions to the big telcos in the CAF II subsidies. The FCC was also largely cut out of the biggest effort ever with BEAD grants to solve the rural broadband gap, and that had to sting. The FCC can now say to the folks living in the A-CAM areas that it provided the funding to bring better broadband instead of the NTIA. I’m picturing FCC ribbon cuttings for projects that launch fiber in these areas. I can’t think of any other reason that this order would have been released so soon after the NTIA announcements of BEAD funding for each state.

The NTIA should react to this announcement by reallocating the BEAD funding to states because for every state that got a windfall like Nebraska from the FCC’s A-CAM order, other states received less BEAD funding. Unfortunately, reopening the allocation process could open a can of worms, so that likely won’t happen.

In my mind, the FCC has become a loose cannon due to its control of the Universal Service Fund. The USF for all practical purposes is a big slush fund that gives the FCC the ability to tackle anything it wants, outside of any control by Congress or the White House. After this announcement, it wouldn’t shock me to see the FCC announce another round of RDOF funding in the middle of the BEAD grant process next year.

Preempting Local Government

In May the House Energy and Commerce Committee marked up nineteen pieces of telecom-related legislation, which means the bills can move forward to the full House for a vote. Today’s blog looks at one bill in particular because it represents what I’m seeing as a new trend of actions taken by big ISPs to preempt the authority of local governments.

The bill is H.R. 3557, the American Broadband Deployment Act of 2023. This legislation would preempt a host of current rights of local governments to manage public rights-of-way for telecom infrastructure. This applies to both wireless infrastructure like towers, but also to landline infrastructure like fiber, huts, and cabinets.

The legislation includes a long list of changes aimed at taking local government out of the business of controlling telecom infrastructure deployment.

  • The bill establishes a 60-day shot clock for local governments to consider requests for rights-of-way. If the local government doesn’t approve a request within that time, the request is ‘deemed’ to be approved. Further, once a project is deemed to be approved, the carrier or ISP can proceed with construction without further notice to the local government. It appears that would give builders the ability to bypass local inspections, traffic control regulations, etc.
  • The legislation would give ISPs and carriers the ability to install facilities anywhere they choose and bypass local zoning rules. This would also eliminate local requirements to hide, conceal, or disguise infrastructure in historic neighborhoods.
  • The legislation imposes a complicated formula for calculating and justifying any fees, with the overall goal of greatly lowering fees.
  • One of the most intrusive changes is that the legislation would place all disputes at the FCC rather than in local courts. This would force local governments to battle disputes in D.C. rather than locally. This would upset a 35-year long truce between Congress and local governments that allows disputes on local-related issues to be heard in local courts.
  • Eliminates cable franchise renewals and eliminates the ability of local governments to require rules such as an ISP having to serve the whole community, the local government requiring PEG channels, or the local government requiring customer service standards.
  • The biggest killer is that the law would give holders of franchise agreements the ability to cancel the agreement without losing any rights-of-ways included in the agreement. This would also kill local franchise fees, a major source of revenue for many governments. Perhaps the most severe provision is that franchise contract holders can eliminate any contract provisions they deem to be commercially infeasible.

There is a mountain of bills in the House this year, and there is no way to know the chances of this coming for a vote. However, there are several telecom bills that have bipartisan support, and bills like this one could be attached to such bills. This includes a bill that would renew the FCC’s authority to hold spectrum auctions and a bill that would stop federal grant funding for broadband infrastructure from being taxable.

To me, this bill is part of an ongoing effort of cellular carriers, cable companies, and big telcos to restrict the ability of local governments to affect the construction of infrastructure. These big companies have already been successful in recent years in eliminating regulation. The cellular companies already got relief from the Ajit Pai FCC that made it a lot easier to place cell sites. This law would codify that change so that a future FCC can’t change it. The large ISPs were successful in getting the Ajit Pai FCC to eliminate most broadband regulations.

This bill is going for a home run to eliminate local regulations these big companies don’t like. I’ve written recently about regulatory capture, and this is an ultimate example of changing the laws to get what the big monopoly providers want. This law would eliminate franchise fees, allow carriers to put infrastructure anywhere they want, and pay low fees in doing so.

More proof of the degree of regulatory capture in the telecom market is that there are no equivalent efforts to change local government control of other kinds of infrastructure like roads, factories, buildings, etc. This bill is the ultimate example of the biggest companies in telecom flexing their power and influence to bypass some of the last vestiges of regulation.

New Battery Technology

The world is growing increasingly dependent on good batteries. It’s clear that using the new 5G spectrum drains cellphone batteries faster. Everybody has heard horror stories of lithium batteries from lawnmowers or weed eaters catching fire. Flying with lithium batters is a growing challenge. People with electric cars want better range without having to recharge. The best way to capture and use alternate forms of power is to store electricity in big batteries. The increasing demand for batteries is happening at the same time that trade wars for the raw materials used for batteries are heating up through tariffs and trade restrictions.

Luckily there is a huge amount of research underway to look for batteries that last longer, charge faster, and are made from more readily available minerals.

Zinc-manganese oxide batteries. Researchers at the Department of Energy’s Northwest National Laboratory have developed a technology that can produce high-energy density batteries out of zinc and magnesium. These are readily available minerals that could be used to create low-cost storage batteries.

Scientists have experimented with Zinc-manganese batteries since the 1990s, but they could never find a way to allow batteries to be recharged more than a few times due to the deterioration of the manganese electrode. They have found a technique that reduces and even replenishes the electrode and have created batteries that can be recharged over 5,000 times. This technology creates the larger batteries used for electric storage in solar systems, vehicles, and power plants.

Organosilicon Electrolyte Batteries. Scientists at the University of Wisconsin were searching for an alternative to lithium batteries to avoid the danger of the electrolyte catching fire. Professors Robert Hamers and Robert West developed an organosilicon electrolyte material that can greatly reduce the possibility of fires when added to current Li-ion batteries. The electrolytes also add significantly to battery life.

Gold Nanowire Gel Electrolyte Batteries. Scientists at the University of California, Irvine, have been experimenting with gels as the main filler in batteries since gets are generally not as combustible as liquids. They had also been experimenting with using nanowires as the diode, but the tiny wires were too delicate and quickly wore out. They recently found that they could use gold nanowires covered with dioxide along with an electrolyte gel. This combination has resulted in a battery that can be recharged 200,000 times, compared to 6,000 times for most good batteries.

TankTwo String Cell Batteries.  One of the biggest problems with batteries is the length of time it takes to recharge. The company TankTwo has developed a technique to build batteries in tiny modular compartments. These are tiny cells with a plastic coating and a conductive outer coating that can self-arrange within the battery. At an electric car charging station, the tiny cells would be sucked out from the battery housing and replaced with fully charged cells – reducing the recharging process to only minutes. The charging station can recharge deleted cells at times when electricity is the cheapest.

NanoBolt Lithium Tungsten Batteries. Researchers at N1 Technologies have developed a battery structure that allows for greater energy storage and faster recharging. They have added tungsten and carbon nanotubes into lithium batteries that bond to a copper anode substrate to build up a web-like structure. This web forms a much greater surface area for charging and discharging electricity.

Toyota Solid-state Batteries. Toyota recently announced it is introducing a new solid-state lithium-iron-phosphate battery as a replacement for the lithium-ion batteries currently used for its electric vehicles. These batters are lighter, cost less, and recharge faster. Toyota claims a range of 621 miles per charge. They say the battery can be fully recharged in ten minutes. By comparison, the best Tesla battery is good for about half the distance and can take a half-charge in fifteen minutes.

Cybersecurity for Schools

FCC Chairwoman Jessica Rosenworcel recently asked the other FCC Commissioners to support a proposal to spend $200 million over three years to bolster school cybersecurity. Rosenworcel plans to issue a Notice for Proposed Rulemaking (NRPM) soon for her proposal. The NPRM will set off a round of public comments and then a ruling if a majority of the Commissioners agree with the final set of rule changes.

There seems to be some need for better school security systems. According to Emsisoft, a New Zealand-based anti-viral and anti-malware company, there were ransomware attacks on 44 U.S. universities and colleges and 45 on school districts in 2022. That was up slightly over 88 attacks in 2021. According to Emsisoft, school IT networks are a popular target since they have less security and staff with less training than corporations.

This announcement immediately raised the question for me of why the FCC is considering this. The U.S. Department of Education has a 2023 budget of $79.6 billion. I can’t help but wonder why school and university cybersecurity is not the responsibility of the USDE, state governments, or local school systems rather than the FCC.

Rosenworcel is proposing that this effort get funded from the Universal Service Fund, specifically the recently launched Learn Without Limits program that is part of the E-Rate program that subsidizes broadband connections for schools with a high percentage of low-income students. According to Rosenworcel’s press release, this could be done without undermining E-Rate’s primary mission of promoting digital equity for schools.

The E-Rate program is perhaps the most popular program at the FCC since it helps poor school districts afford gigabit broadband connections. I can see why the FCC wants to ride that wave of popularity. Rosenworcel has made other interesting proposals recently that would also come from the E-Rate program.

For example, Rosenworcel recommended that E-Rate be used to provide mobile hotspots on school buses. That seems to be an extension of bringing broadband to schools, to bring broadband to students who have long bus rides. She’s also recommended that E-Rate be used to provide Wi-Fi hotspots for students and library patrons. This also extends broadband to students but seems to be in competition with the funding from the Infrastructure Investment and Jobs Act, which is providing billions of dollars for digital equity that would also provide money for hotspots.

The main reason this raises an issue for me is that the Universal Service Fund is funded with an ever-increasing fee burden on voice lines and interstate broadband services. There has been widespread unhappiness with the FCC USF fees. There doesn’t seem to be any appetite at the FCC to let the size of the Universal Service Fund shrink when it makes sense. Instead, the FCC keeps finding new ways to spend the pot of money.

While cybersecurity for schools seems like an important function, cybersecurity is not broadband. If the FCC can sink money into cybersecurity in this manner, then what’s next – money for training for school system IT employees? I’m sure I’ll get some negative comments about my position, but I am not against somebody helping schools with cybersecurity issues. I just can’t see why this is the responsibility of the FCC.

Slowdown of Cellular Expansion

The broadband industry has always been cyclical. The industry has repeatedly gone through periods of booms and busts that have typically been exaggerated by the manufacturers of telecom equipment. When something new comes along, vendors jump on the new idea and drive up expectations for future sales. The stock prices of the vendors rise on the announced future expectations. But inevitably, the wave of enthusiasm comes back to earth, and the market returns to normal and vendor stock prices drop.

We’re now seeing the beginnings of the end of the boom of the big cellular upgrades to 5G. One indicator that the boom is slowing is that Ericsson and Nokia both recently lowered expectations for future equipment sales, and the stock of both companies instantly dipped around 10%.

For the last four years, the cellular industry has been in a boom as the big cellular carriers upgraded around 70% of their cell sites nationwide while also building new small cell sites. These upgrades meant huge sales for Ericsson and Nokia. It meant a big boom for tower climbers and crews who work on upgrading new cell sites. It has also meant a boom in fiber construction when carriers like Verizon and AT&T constructed fiber to replace costly leased transport for cell sites.

The improvement to the nationwide cellular networks has been impressive. The median cellular download speed nationwide measured by Ookla in 2017 was 22.6 Mbps, and at the end of 2022 had climbed to 193.7 Mbps. Most people think that fast cellular speeds are primarily for the benefit of customers. While this is an important issue, faster speeds are even more important for the best functioning of cell sites. Faster speeds mean a given customer uses the spectrum resources for a shorter time, thus freeing the network for other customers. Faster speeds alone have stretched the capability of cell sites to be able to handle a lot more traffic.

A slow-down of 5G construction will have a lot of repercussions around the industry. It will most immediately negatively affect firms and crews who have been working on upgrading cell sites for the last several years.

But there is an upside for the industry as a whole since some of the technicians who have been working on cellular projects can transition to the giant workload currently coming from building fiber. This won’t help technicians who only climb towers, but many of the other technicians already have fiber experience in their background.

These boom and bust cycles raise some interesting questions for the industry. The ones most harmed by the busts are the smaller construction and support companies that gear up to meet a specific industry demand – and these are usually the first ones cut when that demand slows.

I have to wonder what will happen to all of the cell sites that haven’t been upgraded. A lot of the remaining cell sites are rural, and I still see a lot of rural cell sites where carriers have not upgraded to FWA broadband. I recently cited the CEO of T-Mobile who described how the company rates rural markets. His rating system hinted that upgrades might not be coming soon for markets that the company rates low where the population is scattered.

I’ve worked in a dozen counties recently where 30% or more of residents told us on surveys that cellular coverage doesn’t work at their homes. This blog has largely concentrated on the lack of good broadband, but it’s just as devastating for a community when cell phones don’t function well. I’m not sure that DC policymakers fully grasp the hardships that come from lack of cellular coverage. One of my blogs earlier this year talked about a family killed by a tornado since they couldn’t be reached by cellular or broadband to warn about the coming storm. That’s an extreme example of problems that come from lack of cellular coverage – but the bigger tragedy comes in folks that can’t communicate in ways that the rest of us take for granted.

U.S. Worldwide Broadband Ranking

The latest worldwide broadband speed test comparison has been issued by cable.co.uk. This compares broadband speeds in 220 companies is a good way to track the advance of broadband speeds around the world while also seeing how the U.S. stacks up. The comparisons were compiled using 1.3 billion speed tests from M-Lab.

Worldwide broadband speeds continue to climb. The average worldwide broadband speed in 2023 was 93.6 Mbps, up from 72.1 Mbps in 2021 – and only 9.1 Mbps in 2018. That’s over a tenfold increase in worldwide average broadband speeds in five years. Only 34 countries had average speeds faster than the worldwide average, with the United Kingdom coming in 34th with a speed of 93.6 Mbps.

The fastest average broadband speed comes from Jersey, an island nation off the coast of France, with an average speed of 264.5 Mbps. The next fastest countries are Liechtenstein, Macau, Iceland, Gibraltar, Andorra, and Luxembourg. It’s not hard to notice that these are all small countries where it’s much easier to bring fast broadband to everybody.

At the bottom of the list are Afghanistan and Yemen, with an average broadband speed of 1.7 Mbps. There are 48 countries on the list in 2023 with average broadband speeds under 10 Mbps. This is down from 67 countries in 2022 and 94 countries in 2021.

The U.S. is twelfth on the list, with an average speed of 136.5 Mbps. We often forget how much broadband speeds have improved in the country. In 2018 the U.S. ranked 20th with an average broadband speed of 25.9 Mbps. In 2018, the fastest broadband was in Singapore at 60.4 Mbps. Singapore today is 23rd with an average broadband speed of 101.8 Mbps.

One of the most interesting items on the list is the average speed in China of 12.7 Mbps – 156th on the list. We’ve heard from vendors and politicians for years how we can’t afford to lose the broadband battle with China. However, this speed seems unrealistically low. The China speed raises an interesting point about the study. The statistics include speed tests from every broadband technology from each country. In this country, that means combining speed tests from fiber, cable companies, fixed wireless, and cellular hot spots. The U.S. average speed comes from over 373 million speed tests. In China the statistic comes from only 323,000 speed tests. It’s fairly obvious that the China statistic doesn’t represent a very big slice of all of the broadband market in China. That might be due to a simple reason, such as people in China don’t use the M-Lab speed test.

As might be expected, the speed tests show that the fastest major countries are in North America, Europe and Asia. Canada is immediately below the U.S. with an average speed of 136.1 Mbps.

The African country with the fastest average speed is Rwanda, ranked 99, with a speed of 39.9 Mbps. But ten of the bottom twenty slowest speeds on the list are from sub-Saharan Africa. The fastest country in South America is Uruguay, ranked 20th with an average speed of 111.5 Mbps.

The Growth of Upload Usage

I’ve written a number of blogs about the growth of download broadband usage. I recently looked at the growth trend for upload broadband usage and found that upload usage has been growing faster than download usage.

The statistics in the following table come from OpenVault, which has been tracking broadband usage statistics each quarter. The numbers represent the national average monthly usage of broadband for households at the end of the second quarter of each year until before the pandemic. Just like with download usage, there was a big burst in upload usage at the onset of the pandemic as people were sent home. People instantly needed upload links to communicate back to the office or the school. But even since the pandemic, the overall trend shows upload usage growing faster overall than download usage.

Upload Annual Download Annual
Mbps Growth Mbps Growth
2Q 2019 15 265
2Q 2020 23 56% 357 35%
2Q 2021 28 22% 405 13%
2Q 2022 31 11% 460 14%
2Q 2023 36 15% 498 8%

There are a lot of possible explanations for the growth of upload usage:

  • The pandemic trained the whole country to communicate by video conference. This has grown to become a routine practice. I use video conferencing at least a few hours per day, and often a lot more.
  • Over the last five years, a lot of the routine software we use migrated to the cloud. As a common example, Microsoft Office 365 has migrated the Microsoft suite of products to store and save in the cloud. Opening or modifying spreadsheets, Word Documents, or PowerPoints now uses upload bandwidth.
  • There is also widespread use today of collaboration software where multiple people can work on documents, spreadsheets, and graphics at the same time.
  • It’s hard to imagine anybody with a lot of files that doesn’t back them up in Dropbox or the many other storage systems.
  • There is a lot of hidden machine-to-machine traffic where software automatically and routinely connects to the outside world. A few years ago, a Washington Post reporter left his computer running during a month-long vacation and found that his home had generated almost a gigabyte of upload traffic in his absence.
  • It’s now a video-driven world, and people share videos as easily as we used to share pictures.
  • A major portion of gaming has moved to the cloud.
  • We are using a lot more security cameras. There has been a proliferation of doorbell cameras installed as well as inside cameras to check on pets, kids, and babysitters. People routinely check the cameras remotely.

It seems unlikely that upload usage will ever catch up to download usage for most homes. Most people consume more video and other content than they generate. But the volume of average upload usage is still significant. I doubt that anybody a decade ago would have predicted that the average U.S. home would be uploading 36 gigabytes each month.

There doesn’t seem to be any reason on the horizon why the growth won’t continue. More people are sharing videos and other content. We’re slowly creeping towards having early versions of telepresence and virtual reality, which will likely mean a huge bump up in upload usage for many homes. Does anybody care to make a prediction of the average amount of upload usage a decade from now?

Grant Funds are Still Taxable

In October 2022, I wrote a blog about a bipartisan attempt to exempt broadband grant funding from being taxable income. Unfortunately, Congress has still not moved this legislation forward. Any company pursuing any federal, and most state grants need to be aware of the tax implications.

In 2022 there was an attempt to push this through during the lame-duck session between the old and new Congress being seated. When that failed, there was a bipartisan bill introduced in both the House and Senate in February 2023, and those bills are still languishing.

On the House side, Representatives Jimmy Panetta (D-CA) and Mike Kelly (R-PA) introduced H.R. 889 – The Broadband Grant Tax Treatment Act. On the Senate side, Senators Mark Warner (D-VA) and Jerry Moran (R-KS) introduced an identical bill, S.341. Both bills would amend the federal tax code so that grants received from the Infrastructure Investment and Jobs Act and the American Rescue Plan Act will not be considered to be taxable income.

This would cover any funding that ultimately derived from ARPA funding, and I have to assume that it covers both state and local government grant awards made using ARPA funding. This would also presumably exempt ReConnect and NTIA grants that were funded through those two federal laws. But it might not exempt ReConnect funds that were directly funded in an Agricultural bill. The tax exemption would be retroactive for eligible grants awarded in 2021 and 2022.

In the past, the IRS had the authority to excuse grants from being taxable, and the agency excused the taxes on grants made from the NTIA’s 2009 BTOP and BIP grants. But the IRS lost that authority in the 2018 Tax Cuts and Jobs Act. It now takes specific action from Congress to forgive tax on grant income.

Most grant revenue is taxable. This makes sense because the majority of government grants are made to cover salaries for people like researchers – such grants are taxed as personal income like any other source of payroll. But there is a good argument to be made that grants for infrastructure are different. The whole point of the current broadband grants is to build infrastructure. A huge portion of the IIJA funding goes towards the labor of building a broadband network, a road, a bridge, a dam – and those payrolls from the grant funding are fully taxable. Most of the materials used in the grants are supposed to be made in America, and the profits from selling those materials is also taxable.

Fully taxing the infrastructure grant money and then also taxing the payrolls of the folks who build the grant-funded project feels like double taxation to me. We don’t tax corporations on gross sales. Corporations only pay taxes on profits, but most of the tax from sales to corporations comes from the taxes on the salaries paid to employees.

An argument can be made that grant revenue used for infrastructure isn’t taxable – but it involves an understanding of accounting to understand this. Corporations don’t get to immediately write-off the cost of building an asset. The cost of an asset, like a fiber network, is recognized as a tax deduction over time through depreciation. If a corporation depreciates a fiber network over thirty years, then it claims a little piece of the fiber asset as an expense for each of the thirty years.

But saying that an ISP will eventually get the money back ignores the practical impact of making infrastructure grants taxable. Grant money is given to ISPs as they build the network. If it takes three years to build a grant-funded fiber route, then the corporation gets the grant award spread over those three years. In each of those years the grant money is considered to be income. Assuming the corporation is already profitable, it will have to pay a 21% federal tax each year on the grant award. States with a state income tax will also expect the tax payment.

For sake of simplicity, let’s assume the total tax liability is 25%. If an ISP accepts a $40 million BEAD grant, it’s going to owe $10 million in taxes. Over the thirty years of deprecation its tax liability will theoretically get this money back, but the company will be required to write a big check to the tax authorities over each of the first three years. Note that taxes are more complex than this and that’s a simplified explanation.

That is a huge penalty for an ISP for taking grant funding. The company will have already made a 25% matching contribution to the grant project, which would be roughly be $13.3 million. In this simplified example, the ISP will have had to come up with $23.3 million to accept a $40 million grant.

This have huge implications for an ISP. First, while banks might fund the 25% grant matching, they are going to leery about funding the tax liability. ISPs in general don’t carry a lot of cash, so coming up with the tax liability is a big problem. This also makes the project a lot less profitable. My math shows that most rural grant projects are only minimally profitable even before considering this extra tax. BEAD grant areas, by definition, are sparsely populated, meaning there is not a huge amount of revenue generated. I honestly have not looked at a rural broadband project that could absorb this extra tax cost.

This extra taxation makes it a lot harder to justify taking a broadband grant. ISPs consider grant projects because they add customers and increase economy of scale. But nobody wants to take on a grant project that is a cash loser.

I can’t find any news about the topic. The House and Senate web sites show the bills still stuck in the Finance Committees. Perhaps this issue has been rolled into some other piece of legislation, but the folks I know who follow this haven’t heard about anything like that. If this legislation is still stuck in limbo, ISPs need to reach out to their congresspeople to reiterate the dire results of keeping these grants as taxable income. If this is not solved, it’s one more reason that many ISPs won’t be able to take BEAD grants. Only those ready to take on the extra tax liability can justify it – and most ISPs I know can’t afford this extra cost.