Why Elon Musk Is No Longer Getting $886M from the FCC

Note: This is part seven in our series. Be sure to check out our earlier posts:

Part 1: Fiber to the Clubhouse: Pai Subsidizes Broadband for the Rich

Part 2: Broadband Boondoggle: Ajit Pai’s $886M Gift to Elon Musk

Part 3: Space-X Broadband: Coming to an Empty Traffic Island Near You

Part 4: Ajit Pai's Broadband Legacy: Haste and Waste

Part 5: Plan Fail: How Pai Imperiled a Rural Town's Fiber Network

Part 6: Good News: Mass. Muni Fiber Spared from FCC RDOF Blunder

---

Even the world’s richest person can’t get everything he wants.

Earlier this month the FCC told Starlink — Elon Musk’s new satellite-broadband provider — that the agency was no longer going to give the ISP the nearly $900 million in rural-deployment subsidies provisionally granted in the waning days of the Trump administration.

In the same ruling, the leadership of the Biden FCC also canceled the awards provisionally granted to LTD Broadband, a small fixed-wireless startup that somehow managed to win even more than Starlink: $1.3 billion. Those two entities were among the biggest potential recipients of the total $9.2 billion of deployment subsidies from the FCC’s Rural Digital Opportunity Fund (“RDOF”). In a brief note explaining the change in plans, however, the FCC stated that “LTD and Starlink are not reasonably capable of complying with the Commission’s requirements” for investing this money in new broadband service.

This action came 20 months after Free Press first sounded the alarm that the Trump FCC was about to waste billions in these scarce broadband-deployment subsidies by giving free money to Starlink, LTD and other companies to deploy broadband in areas like storage tanks and traffic medians that should not qualify for support.

This month’s decision means that so far the Biden FCC has canceled nearly $2.5 billion in winning awards — more than 25 percent of the program’s total — and that amount is likely to rise once the agency completes its “long form application” review process. And while it might not sound like a success to pull back so much funding provisionally awarded during the final month before the White House and FCC changed hands, it’s exactly the right approach to scrutinize these investments carefully.

It’s too early to say how well the Biden FCC has cleaned up former Chairman Ajit Pai’s RDOF mistakes, but there are plenty of lessons to glean from this debacle.

Making a mess on the way out the door

Shortly after the 2020 election, Trump-appointed FCC Chairman Ajit Pai issued the latest in a long string of over-the-top, self-congratulatory press releases. Pai was marking the conclusion of the bidding rounds of the RDOF reverse auction, a newly constituted $9.2-billion program that offers ISPs subsidies to deploy broadband in rural areas that are too remote for carriers to profitably serve without government support. Pai boasted that the mere completion of the initial bidding-and-selection process “show[ed] that [his] strategy worked.” Pai’s efforts to win the press release certainly worked, as it garnered plenty of positive coverage.

But as was so often the case with Pai, behind the facade of self-praise was an unfortunate truth: The RDOF, far from being an “incredible success,” was actually a mess.

Literally minutes after Pai issued his bombastic press release, Free Press had already discovered numerous problems with the RDOF awards. Two days later we published the initial blog post in this series, becoming the first to sound the alarm that the FCC was about to waste billions of dollars collected via a regressive user-fee system.

Among the issues we noted were awards of more than $700 million to carriers in urban areas; awards for deployment to uninhabited highway medians, golf courses and cul-de-sac virges; subsidies for already served locations such as numerous international airports, big-box retail stores, universities and luxury resorts; awards to ISPs with zero experience, which were nonetheless promising gigabit-level service at suspiciously low bids; awards that would have subsidized for-profit companies overbuilding in rural areas that municipal fiber providers already served; and awards to Elon Musk’s Starlink, a new satellite-broadband company that certainly did not need public support.

Our watchdog efforts prompted plenty of action. Cable-broadband provider Charter — which was slated to receive millions in RDOF subsidies in areas that we identified as already served by municipal fiber providers — told the FCC it realized it had won subsidies in many places that never should have been eligible for support.

Our work also raised plenty of policymakers’ eyebrows, including those of then-interim FCC Chairwoman Jessica Rosenworcel, who made cleaning up Pai’s RDOF mess a top priority. It’s clear that the Biden FCC has learned important lessons from this entire process. Hopefully these lessons will inform other policymakers — particularly those in all 50 U.S. states who are deciding right now how to spend $42.5 billion in rural-broadband subsidies that Congress authorized last year in the infrastructure bill’s Broadband Equity, Access, and Deployment (or “BEAD”) program.

The long road to reform

When we first raised the alarm about the RDOF’s initial results, many reporters and congressional staffers asked us “what can the FCC do about it now?”

The ideal move would have been to cancel all awards for urban areas and redo the auction in rural areas after a serious vetting of all bidding ISPs, using the most recent (and supposedly improved) broadband map produced pursuant to the requirements of the Broadband DATA Act. But the Biden FCC had to act meticulously, within the narrow confines of the existing RDOF and auction rules, to avoid lengthy court battles.

Fortunately that auction process has a step called the “long form application review,” where provisional-subsidy winners submit detailed information regarding their financial qualifications and technical plans. The FCC has plenty of latitude in this review process, and can use it to toss out awards it never should have provisionally authorized in the first place. Yes, it is ridiculous that this step comes after the auction, but so too are many things about the way Washington operates.

While the FCC was engaged in this process, it sent letters to dozens of RDOF winners. They included lengthy lists of urban areas that the agency said were already served, and asked these ISPs to review the lists to “assess whether existing service in these areas will affect your ability to meet all [RDOF] program requirements and deployment milestones.” If an ISP decided it didn’t want the subsidies for those areas, the FCC said it would entertain requests to waive default penalties otherwise imposed on winning bidders that failed to meet their RDOF obligations in those already served places.

This is apparently how cautious the Biden FCC felt it needed to be to proceed. It had a list of urban areas that it determined were already served, and instead of just canceling those RDOF awards, it let the subsidy winners decide if they still wanted the free money. It’s not clear exactly how the FCC came up with these lists of already served urban areas. The lists included many that Free Press had identified — but there were plenty of urban areas and other questionable locations we’d uncovered in our research that were not on the FCC’s after-the-fact lists either, including some of the golf courses and Target parking lots we noted.

Nevertheless, the Biden FCC’s work didn’t stop with these letters. The agency’s recent decision to bounce LTD Broadband and Starlink from the program altogether has prevented billions of dollars in waste, which will benefit people who actually reside in unserved rural areas.

If the areas in which these two companies won RDOF support still show up as unserved in those forthcoming new FCC broadband maps, then the people living and working in those places will likely see broadband deployment funded by the $42.5-billion BEAD program. Plenty of these locations hopefully won’t show up as unserved in the new maps, though, because they’re clearly either uninhabited, or located in urban areas where broadband-deployment subsidies are not needed.

And, of course, if anyone actually lives in these areas, they may be able to sign up for Starlink service right now (or get on the waiting list, which presumably will be cleared before the arrival of the six-year RDOF buildout deadline that Starlink would have been obligated to meet had the FCC not rejected its application).

Apparently, a company with a business plan like Starlink can already serve at least some customers in remote areas without subsidies — which Musk often claims to disdain anyway, even while pocketing them. So the FCC’s decision could be a great opportunity for Starlink to put its own money where its mouth so often is. But Starlink’s results — with data suggesting it is routinely falling short of the upload speeds required of RDOF recipients — show the wisdom of the FCC analyzing the company’s technical bona fides and pulling the plug on these awards.

Trust the process, but verify the result

There are plenty of lessons to draw from the Trump FCC’s RDOF debacle.

Chief among them is that broadband-availability maps must be a starting point — not an end point — in determining where deployment subsidies should go. Maps are only as good as the data that are fed into them, and by necessity that information is mostly coming from the ISPs themselves. There’s just too much room for error, even if everyone attempts to get it right. While the FCC’s existing mapping process has rightly received years of criticism, our research shows that the potential for understating broadband deployment on these maps is as real as the potential for overstating coverage.

Indeed, there are already signs that the glorified “fabric” map reforms aren’t the magic solution that so many hoped they were. (These are maps that are supposed to accurately identify all serviceable locations, which are then to be combined with the ISP’s own maps of their service areas.) This isn’t surprising, as all of the previously released “improved” state-level maps based on similar methodologies have plenty of glaring errors.

For example, the New York State map has bizarre results, such as declaring parts of a dense urban shopping mall as unserved while stores mere feet away have fiber. Or Georgia’s map, which shows Atlanta Hartsfield International — the nation’s busiest airport — as unserved.

It all goes to show that no matter how understandably skeptical people may be of ISPs’ coverage claims, a map with more areas marked unserved is not necessarily more accurate just because it fits that skeptical prediction.

Partisan deadlock threatens future progress

Chairwoman Rosenworcel’s meticulous RDOF review process is a stark contrast to former Chairman Pai’s rush job. Pai’s series of self-congratulatory RDOF press releases garnered plenty of uncritical press during the 2020 election season, while reporters largely ignored Rosenworcel’s concerns about potential problems.

It makes one wonder, if the 2020 election had had a different outcome and Pai had continued as FCC chairman, would he have acknowledged his mistakes and acted to prevent wasteful spending of scarce subsidies? In all likelihood he wouldn’t have: When confronted with our research, he responded with the churlish remark “if we’re getting [criticism] from both directions . . . we probably struck the right balance.”

And Commissioner Brendan Carr had a bizarre and transparently political reaction to the FCC’s decision to cancel Starlink’s $900-million award. Carr’s Trumpian approach harms the public interest, which makes the Senate’s failure to seat Gigi Sohn as the fifth FCC commissioner all the more frustrating. Chairwoman Rosenworcel has managed to do quite a lot of good without a full Commission, but the agency could do even more like this with a potential third vote to overcome cynical partisan resistance.