Connecting the unconnected in rural America

A majority of Americans believe the COVID-19 pandemic has made the internet “essential.” Once a “nice to have” service in to enjoy streaming videos or playing games, broadband connections are now “must have.” Reaching online services to work and learn from home, visit the doctor, and even order toilet paper has never been more important.

But there are up to 42 million Americans for whom this essential network is not available, and millions more for whom it is available but unaffordable.

“There are up to 42 million Americans for whom this essential network is not available, and millions more for whom it is available but unaffordable.”

The Biden FCC is inheriting a two-part broadband challenge: an accessibility gap of those (largely in rural areas) for whom there is no network, and an adoption gap (more prevalent in urban areas) of those unable to afford the network. This paper addresses the accessibility problem and how was aggravated by the Trump FCC’s substitution of PR for progress and the mismanagement of the most promising effort to connect unserved Americans. The next paper in this series will address the adoption gap. One day before leaving office – and only nine months after its previous “annual” report – the Trump Federal Communications Commission (FCC) took one last turn at the political spin machine with a self-congratulatory analysis that proclaimed “significant progress” in expanding access to the internet. The report pegged the number of Americans lacking access to high-speed internet connections decreasing to 14.5 million. An independent study of the previous regularly-scheduled report found the FCC was consistently underreporting reality; that up to 42 million Americans were unserved by broadband.

Speak Loudly and Spin a Small Stick

“Since my first day as Chairman of the FCC, my number one priority has been closing the digital divide and bringing the benefits of the Internet age to all Americans,” Trump FCC Chairman Ajit Pai proclaimed. In a 14-page list of “FCC Accomplishments Under Chairman Ajit Pai” that he ordered be placed on the FCC website, “Bridging the Digital Divide” consumes the first four pages. It is a triumph of self-promotion, but a laundry list devoid of significant measurable results.

The Biden FCC’s challenge begins with correcting the charades that inflated internet coverage by deflating the definition of broadband and overstated the number of homes connected.

Defining broadband down

Any assessment begins with what is being counted. In this case, the Trump FCC’s spin was to define “broadband” unrealistically. In 2015 the Obama FCC – over the objection of then-commissioner Pai – increased the network speed definition for “broadband” to 25 megabits per second (mbps) into the home and 3 mbps out of the home. It was adequate for the time, but the simple evolution of the internet in the intervening years, including the proliferation of in-home internet devices – even before COVID – subsequently rendered those speeds inadequate.[1]

The Trump FCC, realizing that as broadband speeds went up, coverage calculations went down, refused to update the old definition. Mignon Clyburn, the sole Democrat on the Commission when the decision was made, called this out, wisely identifying what would become manifest during COVID as multiple family members tried to get online simultaneously. The Trump FCC’s decision, she objected, “does not even consider that multiple devices are likely utilizing a single fixed connection, or the multiple uses of a mobile device.”

The coverage charade

There is no national address-by-address database of internet coverage. Instead, the FCC relies upon data voluntarily provided by the industry on the agency’s Form 477. That process is fatally flawed because it looks at census blocks rather than street addresses. Under this approach, if there is one internet-accessible household in the census block, the entire area is deemed to have broadband access.

Telecom providers argued that requiring them to collect more accurate data was unduly burdensome. That argument may have worked in the days when the internet was a “nice to have” service; but with broadband now an essential service, such flawed data is unacceptable. For its entire four years, the Trump FCC failed to resolve this mapping problem. Finally, Congress passed a law in 2020 ordering the FCC to reform its data collection. The Trump FCC failed to meet the congressionally-established deadline to adopt procedures for the mapping effort. It is not unreasonable to expect that when the real facts replace the Trump-era charade, the number of unserved Americans will increase. By slow-rolling that result, the Trump FCC avoided such a reckoning on its watch.

There is an old management maxim: “if you can measure it, you can manage it.” An early challenge of the Biden FCC will be to measure broadband penetration based on the facts rather than the spin.

RDOF: A Creative Solution Undermined by Poor Administration

Decades ago, the FCC adopted a strategy of subsidizing rural telephone companies for the higher-than-normal costs created by lower-than-normal population densities. By subsidizing rural telecommunications carriers, the program provided rural customers with services comparable in cost and quality to their urban counterparts.[2]

As the internet age developed, the FCC refocused the program from telephone service to broadband. By one estimate, the period from 2013 to 2018 saw multiple federal efforts provide $22 billion in support to rural telecommunications companies.[3] Yet the number of Americans without access to a broadband internet connection remains unacceptably high.

Allocation via reverse auction is a good idea. Poor management by the Trump FCC, however, has limited its effectiveness.

In an effort to overhaul an underperforming subsidy system, the Obama FCC developed a reverse auction plan for the distribution of broadband subsidies. To avoid the shortcomings of previous efforts, the plan allowed providers to bid for unserved areas and allocating subsidies to the company committing to provide broadband service for the lowest subsidy. In 2018 the Trump FCC held a reverse auction for what was then called the Connect America Fund (the CAF-II auction). In 2020 the Trump FCC renamed the reverse auction concept the Rural Digital Opportunity Fund (RDOF).

Allocation via reverse auction is a good idea. Poor management by the Trump FCC, however, has limited its effectiveness. The Biden FCC has an opportunity to fix some of that poor management and thus deliver more broadband quicker.

A budget not spent is broadband not built

“I’m thrilled at the success of this auction,” Chairman Pai crowed about the RDOF results, “our strategy worked.” In fact, the strategy failed and as a result fewer unserved areas will be receiving broadband than would have in a well-designed auction.

Instead of an auction designed to make its $16 billion budget serve as many areas as possible, the Trump FCC built an auction that would spend as little as possible. In this case $6.8 billion of the $16 billion budgeted was not put to work. The Chairman applauded the “savings.”

While saving money would seem a good thing, in this situation every dollar unspent translates into a rural area unserved. There is a high probability that an auction that prioritizes a race-to-the-bottom for subsidies will produce a similar race-to-the-bottom in service. The Trump FCC turned an auction intended to prioritize delivery of quality service to unserved areas into an auction that disincentivized delivery in favor of a quick win, even if the bidders and their technology were unproven. As one observer noted, such a policy “cheated huge numbers of people out of getting fiber” and the future proof high-speed connections that optical fiber offers.[4]

As a strategy to add to the Chairman’s “List of Accomplishments,” the RDOF auction was a success. As a method to deliver quality broadband service to the most people as quickly as possible, RDOF missed the mark and left a mess for the Biden FCC to clean up.

Failing the smell test

When something looks too good to be true, it often is., The rules the Trump FCC established for RDOF invited mischief and malfeasance. Consider just a few of the warning signals:

• The largest auction winner is a small Nevada company that defaulted on some of its winning bids in the previous CAF-II auction. Nevertheless, the FCC accepted its new promises and gave the firm $1.32 billion to build gigabit fiber to over half a million locations spread over 15 states, even though it had no track record of being able to accomplish something of this scale.
• The auction was structured to incentivize gigabit network speeds, but did not specify, for instance, that fiber optic cable was the best way to achieve that goal. An amazing 85 percent of the winners promised to deliver such speed within 10 years. But a 10 ten year wait to see if unproven technology works hardly solves the short term access problem.
• Winning bids were often pennies on the dollar compared to previous estimates of what would be necessary. The frequency and sheer magnitude of this should have triggered warning signals. Did the auction winners discover a new technology that had gone undiscovered by the unsubsidized companies?
• Many of the winners proposed to provide gigabit service via wireless technology even though such high speeds to multiple locations over vast rural distances has been previously unsuccessful. Telecom consultant Doug Dawson explained this “means the FCC believes that fixed wireless technology is the functional equivalent of fiber,” despite the fact its point-to-multipoint architecture “can’t be used to deliver giant bandwidth to more than a few customers – and it’s not really designed to deliver gigabit download, and certainly not a symmetrical [e., upload from the home)]gigabit.” Even more of a warning: if wireless technology can provide such quality service at the bid costs, why hasn’t it been doing so in more populated areas where the return would be greater?
• The lack of good maps came home to roost. The title of one study on RDOF winners says it all: “Fiber to the Clubhouse: Pai Subsidizes Broadband to the Rich.” According to this analysis, among the unserved areas receiving public subsidies are a luxury resort, an area adjacent to an exclusive golf course and fancy homes, and the Singapore Airlines and Korean Air facilities at the Los Angeles airport.

The Biden FCC inheritance

In order to speed the process to its declaration of victory, the Trump FCC bucked any meaningful due diligence on bidders until after the auction (and, coincidentally, after the presidential election). The Trump FCC could have decided, for instance, that any technology not currently available on a commercial basis at the speeds and the capacity required should simply not be eligible for the auction. Instead, the principal check on bidders’ qualifications and the feasibility of their technology was simply the completion of the so-called “short-form” in which the bidders indicated interest in participating in the auction.

The lack of due diligence allowed companies to bid for billions of dollars without any meaningful vetting of whether their technology would work in the areas they were bidding, or any validation of operational or technical capabilities to deliver on their bids. Following the auction, a bipartisan-bicameral letter from 160 senators and representatives focused on this problem, telling the FCC to increase its scrutiny of RDOF winners. “We urge you to thoroughly vet the winning bidders to ensure they are capable of deploying and delivering the services they committed to providing.”

The Biden FCC will be responsible for the “long form” process that collects and assesses the information the Trump FCC should have collected and assessed in the first place. The letter from congress that identified the work the Trump FCC did not do asked the agency to “redouble its efforts to review the long-form applications…to validate that each provider has the technical, financial, managerial, operational skills, capabilities, and resources to deliver services that they have pledged…”

“By shirking its responsibilities over such obvious matters, the Trump FCC not only rushed out an under-performing product, but also left behind for the Biden FCC a political trap built on a game of chicken.”

By shirking its responsibilities over such obvious matters, the Trump FCC not only rushed out an under-performing product, but also left behind for the Biden FCC a political trap built on a game of chicken.

If the Biden FCC demands and inspects the technical details of the winning bidders – such as whether wireless can provide gigabit speeds to the promised areas, or wether a small fixed wireless company that has never deployed fiber-to-the-home can now reliably do so across hundreds of thousands of homes in multiple states – it risks triggering defaults on the winners’ bids before a dime has been put to work. Alternatively, if it proceeds without such investigation it risks wasting subsidy dollars on networks that don’t perform – an answer that, because of the buildout schedule, could be years off.

The Trump FCC created a situation in which a bidder could over promise to win the bid and then play chicken with the FCC to force the agency to renegotiate to more favorable terms. The Trump FCC knew of this problem because it caved to such a strategy in the CAF-II auction. When a bidder in that auction could not deliver on what it promised, the FCC retroactively changed the rules. This established a “bid what it takes now and negotiate a settlement later” precedent that by all indications was adopted as a primary strategy by many RDOF participants.

Moving Broadband Deployment from a Soundbite to Reality

The haunting challenge the Biden FCC faces with the RDOF legacy is that Americans could be cheated out of meaningful, future proofed broadband service. If subsidies are paid for technologies that cannot work or that perform below promises, potentially millions of Americans will lose. If selecting inexperienced companies or unproven technology over companies that bid based on reality and experience does not work, American can lose again. If the federal government ends up spending public monies on transitional technologies that are better than today, but below what should be built for long-term performance, America can lose yet again.

Because the previous FCC failed to develop a process to validate bidders and technology, it now falls to the Biden FCC to fix the Trump FCC’s RDOF mismanagement in the “long form” process that will soon begin. It will not be an easy process.

As it cleans up the Trump FCC’s debacle, the Biden FCC will fortunately have its own opportunity to deploy the unspent funds to build out unserved areas. The RDOF was budgeted to spend $16 billion but only spent $9.2 billion. The Biden FCC thus has almost $7 billion to allocate correctly. Add this residual to the $4.4 billion that has previously been identified for the next round, and there could be $11 billion for the effective support of rural broadband.

Why any new program should follow a fiber-first strategy is illustrated by this chart illustrating how once it is installed, optical fiber is future proofed with ever-expanding capacity.

Everyone knows of Moore’s Law and the growth of processing power of microchips, but a similar (and not unrelated) acceleration happened with optical fiber as well. When the Trump FCC decided to make RDOF “technology neutral” they hewed to Republican orthodoxy but ignored technological reality. The Biden FCC has the opportunity to run its own auction in a manner that promotes high-speed, future-proofed results by embracing a Moore’s Law-like technology that will keep Americans up to speed long into the future.

[1] The amount of bandwidth or speed needed varies by household. Using Broadband Now’s Bandwidth Calculator, a family of four, each with their own mobile device and laptop, sharing three TVs and two gaming consoles, two tablets and a smart home device (e.g., Alexa) requires 179 megabits per second (mbps) service.

[2] To fund this, the FCC requires carriers to add a surcharge to every telephone bill (the “Universal Service Fee”) and the ability to pass that fee on to subscribers.

[3] I was Chairman of the FCC during many of those years. The failure to more rapidly expand broadband was one of my greatest disappointments.

[4] Without a doubt, there are areas where alternatives to fiber should be utilized; but those should be subsidized only if fiber is impossible.