Early on in the pandemic, Larry Irving left Washington, D.C., and took up residence in a rental property on the edge of Shenandoah National Park, in Virginia, where, he soon found, the Internet service was not sufficient to run his consulting business. So Irving would drive to the local library each day and work in his car, piggybacking onto the library’s WiFi. More than two decades earlier, as the Assistant Secretary of Commerce for Communications and Information in the Clinton Administration, Irving popularized the term “digital divide” to describe the disparity between Internet haves and have-nots. “We did a series of studies that showed that if you were poor, if you were a minority, if you were elderly, and if you were rural, you were less likely to have access to the Internet,” Irving told me. “Poverty was the constant.”
In the past twenty years, population density has favored the building of Internet infrastructure in urban areas, but there has been little economic incentive to do so in many rural parts of the country. It’s a void that the COVID-19 lockdown has laid bare, with so many Americans working from home, attending school using Zoom, relying on telehealth, registering for vaccines online, and, in many cases, saying a final goodbye to a loved one on a screen. Among the enduring images of the pandemic are pictures of children without broadband at home, logging into remote classrooms from the parking lots of fast-food restaurants or the steps of shuttered elementary schools.
Part of the problem is the absence of reliable data on how many Americans lack high-speed Internet. In a 2019 report, the Federal Communications Commission estimated that twenty-one million Americans do not have broadband Internet, which it defines as a minimum of twenty-five megabits per second download speed and three megabits per second upload speed. But that estimate is based on figures supplied by the Internet providers, who have a history of inflating the number of households they serve and the speeds they deliver. A study by researchers at Broadband Now, an industry watchdog, puts the number without broadband at forty-two million—and that does not include the estimated eighteen and a half million people who cannot afford high-speed Internet in places where it is available.
Without accurate data, efforts to bridge the digital divide are bound to fail. “The maps that show where there is service are horrible,” Senator Joe Manchin, Democrat of West Virginia, told me. “I fought with Ajit Pai”—the chair of the F.C.C. during the Trump Administration—“about this for years. He just wouldn’t come to accept that the maps were screwed up.” (Pai called Manchin’s characterization “categorically wrong,” and said that the F.C.C. had “initiated the creation of more granular, accurate maps.”) Brian Deese, the new head of the National Economic Council, told me that the Biden Administration is aware that bad data has resulted in poor Internet service, even as the companies ostensibly supplying it were receiving billions of dollars in federal subsidies. “It’s a problem we’re actively working with Congress and the F.C.C. to fix,” he told me. (As part of its December 2020 COVID-19 relief package, Congress allocated sixty-five million dollars to the F.C.C. to obtain more accurate service data.)
As a candidate, Joe Biden seemed to understand that appealing to rural voters was a political necessity. In his “Plan for Rural America,” Biden promised “to expand broadband, or wireless broadband via 5G, to every American.” As part of the effort, Biden promised twenty billion dollars to build rural-broadband infrastructure, as well as a tripling of the amount of money available to organizations, local governments, tribal groups, and corporations to wire rural communities through the United States Department of Agriculture’s Community Connect program. Shirley Bloomfield, the C.E.O. of the Rural Broadband Association, a national membership organization of eight hundred and fifty small local providers, was invited to join the campaign’s innovation and broadband-deployment committees. “Probably two weeks after the election, we were called in to meet with the F.C.C. transition team, with the National Telecommunications and Information Administration transition team, and the U.S.D.A. rural-development team to share our ideas and priorities,” she told me. “They hit the ground running.”
The federal COVID-19 relief bill, which passed with overwhelmingly bipartisan support in December, allocated billions of dollars to expand broadband access to low-income families, a billion dollars to wire tribal lands, and millions of dollars for distance learning and telemedicine. “My Republican friends feel the same as my Democrat friends,” Manchin told me. “They all want connectivity.” A case in point is his Republican colleague from West Virginia, Senator Shelley Moore Capito, who started the rural-broadband caucus in the Senate and has actively worked on improving data collection. Capito is eager to see infrastructure bills requiring fiber-optic cable to be installed alongside all new highways and roads. “This is an Administration that’s talked a lot about infrastructure development before they’d actually gotten into office,” Capito told me. “I think they’re folding all of this work that we’ve done previously and continue to do with rural broadband, into that.”
Irving told me that after Clinton left office, in 2001, the Bush Administration did little to address the digital divide. The Obama Administration, largely through its stimulus bill, sent money to states to expand broadband, but not enough to close the gap. When Trump took over, Irving said, “you still had thirty to forty million households in this country without access to good broadband—some because of affordability, some because of accessibility, but the affordability issue was pretty much ignored.”
Two years ago, the F.C.C., under Pai’s leadership, announced the Rural Digital Opportunity Fund Auction (R.D.O.F.), a twenty-billion-dollar pot of money, paid out over ten years, to bring broadband to underserved rural areas. R.D.O.F. was designed as a reverse auction, meaning that it would reward companies that submitted the lowest bids for the fastest service. The auction’s first phase, which began this past October, a month before the election, aimed to bring service to more than six million underserved homes and businesses. Bloomfield told me that she and her members were excited that the government was going to commit substantial funds to build to the last mile—those lonely outposts where low return on investment for providers has typically been a deterrent.
That changed after the auction’s winners were announced. “It was absolutely concerning,” she said. The bid of the biggest winner, LTD Broadband, a small Las Vegas-based wireless-service provider, was 1.3 billion dollars, to provide high-speed Internet to more than half a million locations in fifteen states, even though the company’s primary focus has been wireless technology, not fiber-optic broadband, and a number of broadband experts questioned whether it would be able to achieve the speed and fiber-optic capabilities on which the grant was premised. (A representative from LTD Broadband said that the company has deployed fiber-optics in the past and would be massively scaling its production capabilities to fulfill the grant obligations). Elon Musk’s SpaceX won an eight hundred and eighty-six-million-dollar bid for its satellite Internet company, Starlink, a startup that is still beta-testing its technology and whose proposed “rural” service area is about thirteen per cent urban. (The F.C.C. said that it will only award funding after a review process that determines whether the winning bidders will be able to carry out their proposals.) Indeed, when analysts looked at the areas that the R.D.O.F. winners proposed to serve, they found that the F.C.C. had classified a Pentagon parking lot,Los Angeles International Airport, and a number of upscale California neighborhoods as rural. (An F.C.C. spokesperson explained that the program has the word “rural” in its title because most of the money will go to rural areas, not because urban and suburban areas, which also need service, are excluded.)
In West Virginia, Capito was appalled to learn that Frontier Communications had won roughly seventy per cent of all the money allocated to the state—nearly two hundred and fifty million dollars. In 2017, the federal government asked the state to return close to five million dollars of grant money after it found that the state had reimbursed Frontier for “unallowable” and “unreasonable” costs. (Frontier denies that it mismanaged any funds.) When Frontier’s bid had come up for review, Capito sent a letter to the F.C.C., urging it to reject the bid. Other West Virginia groups and organizations did, too, including the West Virginia state senate and legislature, the Chamber of Commerce, and the town of Fayetteville, whose letterhead proclaims it to be “one of America’s coolest small towns,” though one with lousy Internet.
Tom Wheeler, the F.C.C. chair under President Obama, told me that, in 2016, shortly before leaving office, he and his team put out a study showing that the cheapest way to deliver broadband was for the government to pay for it up front, the way highways and airports are financed, rather than using a convoluted subsidy program. “The analysis we did showed that for about forty billion dollars, we could deliver fiber to ninety-eight per cent of all the unconnected locations in the country,” he said. (He was quick to point out that this only solved the rural problem, not the “it goes by my house but I can’t afford it problem,” which is significant.)
The experience of Slic Network Solutions, a forty-three-person operation in the far reaches of New York State which won just under a million dollars in the R.D.O.F. auction, illustrates the challenge of funding broadband piecemeal, through grants and auctions. Two years ago, Slic received thirty-two million dollars from New York State to provide service to ninety-two hundred homes in the Adirondack Park and surrounding counties, where a dismal return on investment had scared away the bigger service providers. Kevin Lynch, Slic’s chief operating officer, told me that “although there’s somewhat of a high capital cost to deploy initially, it’s a very long-term infrastructure investment. Think of the electric lines. Some of them have been there since the thirties. And, so, you know, over time, it will more than pay for itself and provide a great service to the rural areas. That’s why we’ve been big on it.”
