[News] The Broadband Story Abundance Liberals Like Ezra Klein Got Wrong

The following article originally appeared on Washington Monthly:

In late March, the New York Times columnist and podcaster Ezra Klein went on the Weekly Show with Jon Stewart podcast to talk about his new book, Abundance, which Klein had co-authored with journalist Derek Thompson, then of The Atlantic. The book’s thesis is that over the years Democrats, often bowing to left-wing interest groups, have encouraged various forms of red tape—from federal environmental statutes to local zoning rules to minority contractor set-asides—that have gummed up the workings of government to the point that it is no longer possible to build things the country desperately needs, like new housing and clean energy infrastructure, in a timely and cost-effective manner. To illustrate his point, he regaled Stewart with a tragicomic tale of the government failure to expand rural broadband access.  

In late 2021, Klein explained, Congress passed and Joe Biden signed a major infrastructure law that contained $42 billion to subsidize the construction of broadband networks in rural areas that lack access to high-speed internet service. But more than three years later, not a single home had been connected by the so-called Broadband Equity, Access, and Deployment (BEAD) program. Indeed, he noted, not a dime of that $42 billion had even been spent.  

The reason, Klein went on, has to do with a “baroque” 14-stage implementation process for how BEAD funds can be distributed from the federal government to state governments, and ultimately to internet providers. To give Stewart a full appreciation for the bureaucratic nightmare this entails, he listed the steps one by one. States submit a “Letter of Intent,” a request for planning grants, a “5-year Action Plan,” an “Initial Proposal,” and a “Final Proposal,” with long federal review and approval periods in between. (This elicited an anguished “Oh my god!” from Stewart.) The federal government publishes a map of where broadband needs to be built, states challenge the map for accuracy, the feds review and approve the challenge results. (“My hair was dark when we started this process!” Stewart exclaimed.) Notice and comment periods are sprinkled in throughout. No state, Klein explained, had made it past the “Final Proposal” stage. Stewart sat in apparently stunned silence, later quipping that BEAD is not a broadband program at all, but rather an “overcomplicated Rube Goldberg machine that keeps people from getting broadband.” To Stewart’s pleas for an explanation of how this defective process came about, Klein responded simply, “This is the Biden administration’s process for its own bill—they wanted this to happen.” 

Clips of the conversation went viral—boosted by Elon Musk, who was angling for the Trump administration to call BEAD a failure and reroute the funds to his Starlink satellite internet service. And as the broader “abundance liberalism” movement has continued to gain traction among prominent Democratic politicians and donors, Klein’s narrative of the BEAD program has become a favorite anecdote of its leading evangelists. Jonathan Chait, Dylan Matthews, Josh Barro, and Matt Yglesias have all cited BEAD as an example of liberal governance failure. 

The accuracy of the story, however, has come under question. In April, Bharat Ramamurti, a former member of the Biden administration’s National Economic Council who worked on telecommunications policy, alleged that the BEAD implementation process was not a Biden or even a Democratic prerogative, but the result of a compromise with Senate Republicans during negotiations over the infrastructure law. According to Ramamurti, these Republicans had “insisted” on a cautious implementation design in part to monitor spending for waste, and in part “at the behest of large incumbent internet providers,” who wanted more opportunities to shape the program to protect their interests.

In a subsequent New York Times column, Klein admitted that he had gotten some of the facts wrong—that “portions of [BEAD’s] 14-stage process were insisted upon by congressional Republicans.” But rather than concede the broader argument, he doubled down, saying that after further talks with “various people who’d been part of the broadband program,” he discovered that “much of the process was worse than I’d known.” One official, he wrote, told him that “he’d wasted 40 to 50 percent of his time on internal government requirements he judged irrelevant to the project,” though Klein didn’t name the official or the specific requirements the official was referencing. Similarly, Klein’s coauthor, Derek Thompson, acknowledged in an interview with the journalist Mehdi Hasan that Klein initially “got some things wrong” about BEAD, but insisted that “rules we’ve put in our own way” (he didn’t specify which) had derailed the program. 

So, what’s the real story here? Was it liberal proceduralism or corporate power that incapacitated Biden’s rural broadband effort? Getting the answer right is vital for two reasons. First, the economic and political stakes of the rural broadband problem are serious. High-speed broadband is to the 21st century what electricity became in the 20th—a service so essential that lacking it means not fully participating in modern life. Yet more than 40 million Americans today live in small towns and rural communities that lack access to internet service at the speeds that are required to pursue an education, hold down a job, or run a business that can compete with firms in the big cities. This broadband “digital divide” is a major driver of regional inequality. A study by the Center on Rural Innovation found that rural counties with greater broadband utilization, typically due to faster and better broadband service, have 213 percent higher rates of business start-ups and 18 percent higher per capita income growth than comparable rural counties with lower broadband usage. And according to a survey by the Pew Research Center, 58 percent of rural residents believe that access to high-speed internet is a problem in their area, compared to only 13 percent of urban dwellers and 9 percent of suburbanites who think the same is true in their communities. Rural residents have valid reasons, in other words, for believing that the economy is rigged against them.  

Second, figuring out what precisely screwed up the BEAD program can help adjudicate a crucial debate taking place among Democrats about how best to make the party more politically competitive. Abundance liberals say the answer is for Democrats to embrace an agenda of targeted deregulation—one that would free both government agencies and private companies from picayune procedural constraints that get in the way of building needed physical infrastructure and developing promising new technologies that can solve abiding human problems. Others, including the editors of this magazine, argue that abundance liberals, in their laudable eagerness to diagnose the deficiencies of government, fail to recognize that corporate behemoths are frequently the hidden saboteurs, and that their outsized economic and political power is a direct result of past deregulatory efforts.  

To resolve this question, we spoke with nearly two dozen government officials and outside experts who were involved in the design and implementation of the BEAD program, pored over hundreds of pages of program documentation, and researched rural broadband policies of previous administrations going back to the 1990s. What we found is that while abundance liberals are certainly right that some infrastructure projects have been slowed or stalled by regulations and public engagement processes put in place by Democrats to placate progressive interest groups, that is simply not the case with Biden’s rural broadband initiative. Rather, the complexity and delays of the BEAD program and the broader failure of Washington over many years to solve the digital divide is overwhelmingly the result of telecom monopolies whose economic and political power previous administrations unleashed. And this misjudgment by abundance liberals is not a one-off mistake, but part of a pattern. 

Understanding BEAD requires a bit of background on the government’s decades-long effort to spur broadband deployment. 

When we talk about internet access, we are actually talking about two distinct, layered services. The first is data transmission, or the carriage of data from point A to point B across network infrastructure. The second is data processing, or the transformation of data from digital computer language into a more easily transmitted format at point A, and back into digital computer language at point B, using equipment like modems that sit on top of the transmission infrastructure. 

To create a competitive market for data processing, the Federal Communications Commission in the early days of the internet regulated basic transmission services—which at the time could only be provided by telephone wires—under “common carrier” rules. In practice, this meant that telephone carriers had to allow a third-party data processor (an “internet service provider,” or ISP) like AOL to connect modems and other hardware to their telephone wires. The telephone carrier then had to transmit data processed by AOL modems on the same terms as it transmitted data from modems belonging to its own ISP business, if it had one. This regulatory regime was widely considered a success: by the late 1990s, there were more than 7,000 ISPs in North America. 

The Telecommunications Act of 1996 is mostly remembered as ushering in sweeping deregulation and monopolization across the telecom and media industries. But as the Harvard Law professor Susan Crawford writes in her 2013 history of American internet policy, Captive Audience, the act actually represented continuity with the FCC’s regulations. Indeed, the legislation empowered the FCC to subject two-way “telecommunications service” providers not only to open-access requirements, but also to a broader suite of common carrier regulations. 

But this approach seemed oppressive and old-fashioned to George W. Bush’s first FCC chairman, Michael Powell, a tech enthusiast who believed fundamentally that “broadband should exist in a minimally regulated space.” By the early 2000s, new innovations had enabled cable television networks to transmit internet data, breaking the monopoly previously held by telephone wires. Other technologies, such as satellite, broadband over power lines, wireless (data transmitted via cell towers to either a fixed antenna or a mobile device), and fiber-to-the-premise (extending the fiber-optic “backbone” of transmission networks all the way to end users) were promising to join the fray. Powell feared that common carrier regulation would stifle investment in these emerging technologies. Alternatively, Crawford writes, he believed that a hands-off approach would allow them to flourish, ultimately creating a world where competition between vertically integrated transmission and internet service providers would protect consumers in the way ISP competition on top of individual transmission networks once had. “The great regulatory difficulty over the past one hundred years is, we have always had just one wire to the home,” said Powell at a 2004 conference. “We have a historic opportunity here not to repeat that world … The future is exciting, innovative, and bright.” 

To unlock this vision, Powell declined to regulate internet networks as common carriers and waged a legal battle to liberate them from the act’s “telecommunications service” classification that made such regulation possible. When a court ruled in 2003 that cable was a telecommunications service, Powell enlisted the Department of Justice to appeal; the Supreme Court ultimately ruled in Powell’s favor in a technical decision related to the FCC’s authority. With this ruling in hand, Powell was free to do as he pleased. As he campaigned for reelection in 2004, George W. Bush pledged that with Powell “clearing the underbrush of regulation”—and additional supply-side reforms Bush enacted that year, such as easing permitting requirements for providers to access federal rights-of-way—the market would deliver “universal, affordable access to broadband technology by the year 2007.” 

Of course, this vision didn’t come to pass. Powell’s deregulation effectively gave the telecoms permission to deny competing firms access to their lines, leading thousands of ISP start-ups to go out of business, as the Washington Monthly reported at the time. Moreover, in urban and suburban areas, the “intermodal” competition Powell had banked on failed to arrive. The technology behind broadband over power lines never panned out. Satellite and wireless service reached the market but proved to be poor substitutes for a wired connection (IT geeks took to calling satellite “fraudband”). Telephone carriers experimented with rolling out fiber-to-the-premise networks in affluent areas, but Wall Street insisted that they focus their investment on cornering the growing, less capital-intensive cellular service market. By the end of Bush’s second term, most consumers were left with only two choices for genuine broadband internet: their local telephone carrier or their local cable company. Under this duopoly market structure, as Nicholas Thompson reported in these pages in 2009, Americans paid far more, for worse service, than citizens of other developed nations, where forms of common carrier regulation had almost universally been adopted. 

For rural Americans, the situation was even worse. Deregulation and permitting reform were not enough to lure cable internet providers to remote, sparsely populated areas. This left telephone carriers, which had preexisting rural infrastructure thanks to long-standing universal phone service obligations, firmly entrenched as the dominant transmission providers in these areas (satellite “fraudband” and fixed wireless service had some presence). Meanwhile, lax antitrust enforcement was allowing these carriers to “cluster” their regional monopolies into increasingly large, and politically powerful, entities. By 2005, the “Baby Bell” offspring of the 1984 AT&T breakup had already reconsolidated into two behemoths: a reborn AT&T, and Verizon. CenturyLink and Frontier Communications also grew to control significant chunks of rural territory, sometimes by buying lines from AT&T and Verizon. Just like that, the Bush administration had handed a handful of carriers lucrative, unregulated monopoly fiefdoms over rural internet service provision—and the political clout to protect them. 

Before the Bush presidency was even over, it had become clear that a new approach to broadband policy was needed. In 2006, former Clinton Chief of Staff John Podesta and the Free Press founder Robert McChesney pointed out in the Monthly that a better model would be to follow what Franklin D. Roosevelt had done to bring electricity to rural America. In the 1930s and ’40s, the Rural Electrification Agency had provided low-cost loans and other support for municipalities and farmer cooperatives to set up public power utilities. Unburdened by short-term profit imperatives, public power initiatives proved a resounding success, while the threat of competition often spurred investment from private utilities. The federal government could replicate this playbook with broadband, Podesta and McChesney argued—if it first preempted a growing number of state “anti-muni” laws, passed at the behest of incumbent providers, that banned municipal networks.  

As a presidential candidate, and early in his first term, Barack Obama seemed to agree with this logic. A spokesperson for his 2008 campaign told PBS that “if private entities are not going to deliver wireless internet or broadband to an area, municipal governments should be allowed to,” suggesting Obama would preempt anti-muni laws. In his first month in office, he signed into law a major post-financial crisis stimulus that contained more than $7 billion in broadband grants and loans, much of which would go toward municipalities and cooperatives. This investment, the president declared, would allow “a small business in a rural town” to “connect and compete with their counterparts anywhere in the world.” 

Unfortunately, Obama’s promises held up little better than those of Bush. The stimulus grants were undermined by the administration’s desire for immediate results: the Commerce and Agriculture Departments, the agencies tasked with implementing the grants, insisted on funding “shovel-ready” projects, leading to rushed, ill-conceived proposals. Defaults and other misfires ensued, providing years’ worth of ammunition for industry-aligned advocates and academics to discredit the New Deal model for broadband. Meanwhile, Obama didn’t follow through with preempting state anti-muni laws, allowing them to proliferate across the country.  

The Obama administration was handed a separate opportunity to get broadband policy right. In the stimulus bill, Congress mandated that the FCC transform a long-standing program it administered to subsidize rural telephone service, the Universal Service Fund, into the new broadband-focused Connect America Fund. This could have gone a long way toward closing the digital divide by redirecting several billion dollars of federal spending every year into broadband upgrades. The question was, how would the money be spent?  

Other nations that had made significant public investments in broadband prioritized fiber-optic networks. The upfront cost of laying fiber is relatively high, but the networks are “future proof” in the sense that once the high-bandwidth wire is in the ground, it can be scaled up to provide ever-faster speeds by cheaply switching out electronic components (cable broadband is also future proof, although to a lesser extent). In other words, a one-time investment in fiber can yield true broadband internet service for decades, whereas other technologies are expensive to upgrade or, in the case of satellite, need to be replaced altogether after a few years. 

But in the development of the plan, two factors discouraged the FCC from prioritizing fiber. The first was lobbying from incumbent providers. Telephone companies were well positioned to deploy fiber in rural areas, but they preferred to squeeze all the juice they could out of their legacy telephone wire networks rather than cannibalizing these sunk-cost investments. Satellite and fixed wireless providers, meanwhile, wanted a slice of the subsidy pie—and did not want the government to fund fiber competitors to their low-quality offerings. 

The second factor was a flawed, consultant-produced cost model commissioned by the FCC, which greatly overestimated the price tag of laying fiber in rural America. Jonathan Chambers, who would help implement the Connect America Fund as the FCC’s chief of policy analysis from 2012 to 2016, told us that the model projected the costs of building “greenfield” networks from scratch, rather than the more common “brownfield” approach of leveraging preexisting poles, ducts, or other infrastructure. “They calculated a cost of like, $150 to $180 billion,” Chambers said. “They said, ‘Well, my God, we don’t have that kind of budget.’” 

Thus, Julius Genachowski, Obama’s first FCC chair, had a conundrum. Subsidizing satellite, fixed wireless, and telephone wire was clearly not a long-term solution to the digital divide. But Genachowski was a business-friendly moderate who had already been engaged in another battle with industry over an Obama campaign pledge to prevent internet providers from privileging their own content applications. With incumbents opposing a fiber-first approach—and a study suggesting that it would take decades to implement—Genachowski took a technologically neutral approach to the new Connect America Fund, pledging support over the following decade for any networks capable of providing download speeds of at least 4 megabits per second (Mpbs). This speed threshold could be met with modest upgrades to telephone networks, and by satellite and fixed wireless service. 

Four Mpbs service allows a user to browse the web, send emails, and stream standard definition video. These were the essential needs as of 2010, but far short of where the internet was heading—the plan itself set a target for most Americans to receive service at 100 Mpbs download speeds by 2020. When consumer advocates and members of Congress raised these concerns, Genachowski responded that 4 Mpbs was just an initial speed threshold, and could be scaled up as the FCC implemented the subsidies over time.  

Nevertheless, the direction had been set: Through the Connect America Fund, rural Americans would receive second-class service from incumbent providers. Between 2012 and 2015, the FCC offered the 10 largest telephone companies more than $10 billion to upgrade their networks. Between 2016 and 2019, the FCC gave $15 billion to smaller, rural-only telephone carriers to do the same. In 2018, the FCC held a $1.5 billion auction for alternative providers to bid on locations that the large telephone companies had declined to serve, allowing fixed wireless and satellite providers to get a piece of the action. 

By the close of the 2010s, most of the networks supported by the Connect America Fund were already obsolete. As Donald Trump’s FCC crafted the $20 billion Rural Digital Opportunity Fund in 2020, its signature program under the Connect America Fund umbrella, it found it had to re-subsidize many of the same locations subsidized just a few years prior. “It was scandalous, what the commission did,” Jonathan Chambers would tell Broadband Breakfast of the Obama-era implementation of the fund. “It was graft.” 

The Trump FCC’s iteration of the Connect America Fund would be an improvement in one critical respect. Rather than offer money to incumbents and auction off what they declined, it ran an auction from the start. The third-largest winner in the auction was a consortium of over 90 rural electric cooperatives led by Chambers himself, who had cofounded a company that helps cooperative energy utilities leverage their infrastructure to deploy fiber broadband. The consortium secured more than $1 billion by bidding as much as 50 percent below the FCC’s asking price on more than 600,000 locations, validating Chambers’s hypothesis that its cost model had been way off the mark. 

But at the same time, the Trump FCC repeated many mistakes of the past. While the Rural Digital Opportunity Fund primarily financed fiber, it also put billions toward short-term fixed wireless and satellite deployments—including a new satellite entrant whose infrastructure came with better service quality but equally short life spans: Elon Musk’s Starlink. And as with Obama’s original stimulus grants, haste to deployment undermined many of the projects that received funding. Hoping to get money out the door amid Trump’s reelection campaign, the FCC awarded the subsidies without asking for much proof that bidders had the financial means or technological capability to follow through on their commitments. The result was widespread defaults that continue to roll in to this day. 

In total, the federal government spent $50 billion on rural broadband over the 2010s—more than enough money to lay fiber to every rural home and business in the nation, according to Chambers’s estimate. And yet by 2020, as many as 42 million Americans still lacked broadband. 

The rural broadband programs of the 2010s were products of what the telecommunications scholar Christopher Ali has called “the politics of good enough.” Politicians and regulators wanted to do something about the digital divide. But their deference to industry, and desire for immediate results, prevented them from approaching the problem rationally. 

The COVID-19 pandemic looked like it might finally upend this political calculus. “We reached a pressure point where we couldn’t play games anymore,” Harold Feld, a longtime public interest advocate at Public Knowledge, told us. Suddenly, members of Congress were overwhelmed by “their constituents calling them up and saying, ‘My kid is doing homework in a parking lot.’” The door was open for a real solution. 

In March 2021, when Joe Biden unveiled his vision for one of the first major legislative pushes of his young presidency—a massive infrastructure package that would ultimately become the Infrastructure Investment and Jobs Act, or IIJA—it contained a remarkably ambitious broadband plan. It was a plan that finally seemed to grasp the urgency of closing the digital divide, and the correct historical precedent for doing so. “With the 1936 Rural Electrification Act, the federal government made a historic investment in bringing electricity to nearly every home and farm in America, and millions of families and our economy reaped the benefits,” read an announcement from the White House. “Broadband internet is the new electricity.” 

Biden’s plan called for a $100 billion investment focused on “future proof” technology. It promised to promote competition, including by “lifting barriers that prevent municipally-owned or affiliated providers and rural electric co-ops from competing on an even playing field with private providers”—a clear reference to preempting state anti-muni laws—and “prioritiz[ing] support” for these networks. Finally, the plan promised to “reduce the cost of broadband internet service” while noting that “providing subsidies to cover the cost of overpriced internet service is not the right long-term solution,” hinting at a move toward rate regulation. “When I say ‘affordable,’ I mean it,” Biden declared in a speech in Pittsburgh the day the plan was released. Through these measures, Biden vowed to bring broadband to “every single American” at long last.  

For incumbent providers, Biden’s plan was nothing short of an existential threat. Big telecom stocks tumbled on the announcement. Providers and trade associations released a flurry of statements urging the new president to be reasonable. “A fiber first policy might inadvertently grow a rural mobility digital divide,” warned one. A “misguided” push to fund municipal networks would prevent continued private investment from “get[ting] the job done,” warned others. Michael Powell himself, the architect of internet deregulation under Bush and now the president of the National Cable & Telecommunications Association, cautioned against “discarding decades of successful policy.”  

Congress took up the task of converting Biden’s plan into legislation in the early summer of 2021. The broadband portion of the IIJA was hammered out by a bipartisan working group of senators from heavily rural states, with Susan Collins of Maine and Jean Shaheen of New Hampshire serving as lead Republican and Democratic negotiators. Behind the scenes, an epic lobbying campaign was underway. 

Across a long summer of negotiations, the key provisions of Biden’s plan were pared back one by one. The $100 billion figure dropped to match a Republican counteroffer of $65 billion, with only $42 billion allocated to the new subsidy program, now known as the BEAD program (the rest would support preexisting initiatives). Forty-two billion dollars was still a historic investment, sufficient to connect all unserved areas of the country. But it would leave almost nothing left over to fund the deployment of new networks in “underserved” areas with existing but inadequate service. This aligned with incumbents’ goal of avoiding competition to their existing infrastructure. Meanwhile, Biden’s mandate to fund future proof networks was cast aside when Congress set a speed threshold allowing fixed wireless networks, and Elon Musk’s Starlink, to qualify for subsidies. The proposals to preempt anti-muni laws, and to give municipalities priority in applying for subsidies, disappeared altogether. When the dust settled on negotiations in August, one disappointed consumer advocate called the BEAD program a “Shadow of Its Former Self.”  

A different Senate compromise was ultimately responsible for slowing the pace of BEAD’s implementation. In early stages of negotiations, the White House had proposed to run a centralized, nationwide subsidy program out of the federal government, according to several former Biden administration officials and congressional staffers with knowledge of negotiations. This proposal—and the other key elements of Biden’s original plan—was modeled on legislation first introduced by Representative James Clyburn in 2020, which would have tapped the Commerce Department to award funds to providers. But Senate negotiators instead opted for a federalist design. The final text of the infrastructure law ordered the Commerce Department—specifically, a bureau within the department called the National Telecommunications and Information Administration, or NTIA—to award grants to states, which would then design their own subsidy programs within guidelines set forth by the NTIA, and ultimately allocate the funds out of newly created “broadband offices.” 

With those provisions, the Senate overrode Biden’s streamlined vision, and created a whole new layer of bureaucracy: Rather than directly awarding subsidies to providers, the NTIA would have to manage 56 separate broadband programs (one in each state, plus several territories). Government sources told us that the initial impetus for implementing BEAD through the states came from Republicans and rural-state Democrats. It was a good outcome for providers, which maintained extensive lobbying operations in statehouses. It also provided a way for Republicans to defeat a White House proposal to force subsidized networks to provide a basic internet plan for low-income consumers costing $30 a month. This proposal was a top priority for the White House; one former senior White House official said that “at one point, we had [it] signed off and in text by Susan Collins, because we made the case over and over again.” But with the states in play, Collins was able to force a compromise. The final text of the infrastructure law allowed states to propose their own definitions of a “low-cost service option,” while explicitly prohibiting the NTIA from “regulation of rates.” 

Consumer groups also supported state involvement in BEAD, not wanting to repeat the hastily deployed, default-prone programs of the Obama and Trump years. “I mean, who wants to repeat that nonsense when you’re talking about $42 billion?” said Gigi Sohn, a cofounder of Public Knowledge and former Biden FCC nominee. 

The upshot, however, was that BEAD would be a much slower program to implement. No one was happier about this than telecom lobbyists, who were cheerfully gearing up for the coming state-by-state battle for funding. “In general,” reported the IT trade publication Light Reading, “the message from top telecom trade associations to their members is: Don’t worry. We’ve got this.” 

Right up front, BEAD’s federalist design added an extra step to the implementation process: the divvying up of a $42.5 billion pie between the states. A key piece of context about BEAD—one that has been missing from the discourse surrounding the program—is that this one step would account for nearly half of the total time between the enactment of the infrastructure law and the end of Biden’s term.  

To allocate funds fairly, the federal government needed to know how many underserved locations each state needed to connect. The problem was that no accurate and comprehensive data set existed. Thanks to more than a decade of industry lobbying, the FCC’s map of broadband availability counted coverage at the level of census blocks, based entirely on provider self-reported data—allowing incumbents to keep subsidized competitors at arm’s length. In early 2020, Congress finally intervened, enacting legislation to force the FCC to create a granular, address-level map, and implement a system for the public to challenge provider coverage claims. But Trump FCC Chairman Ajit Pai sat on his hands for the remainder of his term, claiming that the FCC could not start work on the map until Congress appropriated funds explicitly for that purpose. (The funds were appropriated in December, but Pai was by then looking ahead to a private-sector career—first in private equity and later as the president of a major wireless provider trade association.) By the time the infrastructure bill was being negotiated, the Biden FCC had just begun work on the new map, and staff predicted that it would not be ready until “probably next year.”

If speed to deployment had been the top priority, Congress could have allowed the NTIA to make initial allocations based on a rough approximation of each state’s needs. Instead, the law was written to explicitly require the NTIA to wait for the new map. 

As with the decision to run BEAD through the states in the first place, the political dynamics behind this move were mixed, people with knowledge of the negotiations told us. All senators wanted to ensure that their own state received its fair share of what remained, even after all the watering-down in the Senate, a “once-in-a-generation” federal investment. But Republicans also worried about the possibility of any state receiving a disproportionately large budget. Their motivation for cutting BEAD down to $42.5 billion in the first place had been to allow the program to cover unserved locations, while minimizing leftover funds that might be used for competitive deployments. But if certain states received too much money, the calculation would be thrown off, opening the door for providers’ worst nightmare: the “overbuilding” of incumbent networks. Thus, basing the state-by-state allocation on the new map “became table stakes for Republican members,” a former senior Commerce Department official with knowledge of infrastructure law negotiations told us. “You’ve got to use the maps, you’ve got to serve all the unserved locations before you do anything else.” 

From November 2021 through September 2022, the new map was caught up in a series of legal challenges filed by a losing bidder for the contract to build the map’s underlying “fabric” of broadband-serviceable locations. When the “pre-product,” yet-to-be-challenged version of the map finally arrived in November 2022, it was riddled with errors. In many states, the pre-product map overstated broadband coverage—an expected outcome, since at this stage it was still based on provider-reported data. In other states, the map fabric mistakenly included so many sheds, barns, and other uninhabited structures that on balance it erred in the opposite direction, overstating the number of locations that needed to be served. Seeing the extent of the mess, senators of both parties called for the NTIA to push back its “aggressive” June 2023 allocation deadline, to allow for an extended challenge period. Future Republican Senate Majority Leader John Thune even cosponsored legislation to codify the extension, in hopes that funding would go “to areas that are truly unserved.” The NTIA declined to extend the deadline, and allocated the funds in late June. 

By then, a large chunk of the 14-step process Ezra Klein would later describe to Jon Stewart was already complete. States had applied for and received planning grants from the NTIA, used the planning grants to create new broadband offices, and taken care of preliminary planning tasks. The NTIA had released its guidelines to structure state program proposals. From the day allocations were made, the NTIA gave state offices exactly six months—until late December 2023—to submit initial proposals for their programs, and one year after that to finalize them with the NTIA. By November 2024 at the latest, states would be able to start getting money out the door. 

If you put abundance-aligned punditry on BEAD under a microscope, you’ll find that it shifts between multiple narratives to explain the slowness of the program. Ezra Klein has emphasized the Democratic fetish for pointless process. But others in his orbit peg the delays on a separate preoccupation of abundance liberals that Klein has dubbed “everything-bagel liberalism”: the Democratic penchant for loading up infrastructure projects with burdensome requirements related to labor, climate, and social justice goals. “One thing the abundance movement asks Democrats to do is to scrape the toppings off the ‘everything bagel,’” wrote Josh Barro in June. “Don’t assign climate goals to your rural broadband project, et cetera.” In May, Matt Yglesias published a guest essay by a former Biden political operative, who repeated the abundance conventional wisdom that “in an attempt to satisfy a narrow ideological faction,” the Biden administration had saddled BEAD with “rigid, impractical rules, sacrificing effective governance for ideological purity.”  

A close look at how the NTIA’s everything-bagel-y state requirements played out in the real world reveals that they were not major time sinks. While states had to document steps taken to ensure that minority- and women-owned businesses would be “recruited, used, and retained when possible,” complying with this was as simple as advertising that BEAD was happening through identity-based nonprofit organizations (the “Houston Hispanic Chamber of Commerce,” the “Women’s Business Center of Utah”). While states had to ensure that subsidized providers “use strong labor standards and protections,” the minimum requirement here was simply asking subsidy applicants to submit a written plan for ensuring compliance with labor and employment laws. In awarding grants, states had the option to prioritize applications from providers who committed to additional labor and wage provisions, but there was no federal requirement to do so. Finally, a requirement that states conduct “an assessment of climate threats” served to ensure that states prioritized funding technologies resilient against local weather patterns—it did not have anything to do with fighting climate change. “Resilience is important, right?” said Veneeth Iyengar, executive director of Louisiana’s ConnectLA broadband office. “Hurricane season is coming soon, and over 90 percent of our infrastructure is gonna be buried.” 

In reporting this story, the Monthly spoke with five leaders of state broadband offices, as well as outside experts who consulted on state proposals. While they expressed various frustrations with the NTIA’s management style—confusing or delayed guidance on technical requirements was a big one—all dismissed the idea that liberal requirements were a significant drag on completing their proposals. They emphasized that most of their time was spent figuring out how to divide unserved areas into parcels (“project areas”) that would make sense to providers as network building blocks, and how to design a scoring rubric for applications that would result in commitments to serve each parcel at a competitive price—in other words, the essential business of parlaying their budgets into universal broadband coverage. “If you want to know the biggest lift of designing the proposal, it was trying to figure out how to do the project areas,” said Eric Frederick, Michigan’s chief connectivity officer. Frederick explained that his office developed a “hexagonal grid system” rather than relying on oddly shaped census blocks, allowing providers to propose more efficient network designs. Compared to this technical challenge, writing a few paragraphs on how his office had advertised BEAD to diverse populations “was not a lift for us.” 

Sascha Meinrath, a Penn State telecommunications professor who consulted on several state proposals, concurred. “Doing an entire regional network, you know, diagramming and all that, is easily 90 percent of the time,” Meinrath said. “Having done many of these successfully, I can tell you that if that’s the case [that a state is struggling to comply with the everything-bagel requirements], they’re definitely doing it wrong.” 

Nor was compliance with everything-bagel requirements a barrier to states receiving final NTIA approval on their program designs. If any particular requirement stood out as a limiting factor here, it was the requirement for states to define a low-cost plan that subsidized networks would have to offer to poor consumers. (Lest this be dismissed as a nonessential social justice priority, it’s worth noting that affordability is a key driver of the digital divide. There are more Americans who forgo broadband because they cannot afford it than there are Americans who do not have any access at all.) 

Recall that during infrastructure law negotiations, a compromise between the White House and Senate Republicans had given states flexibility over what counted as “low cost.” Several states took this flexibility to an extreme, proposing in their initial program designs to leave the definition of low cost entirely up to providers. When the NTIA ordered these states to define low cost as an exact price or formula, several states—including South Carolina, Virginia, and Georgia—held firm, accusing the NTIA of illegal rate regulation.  

After a prolonged back-and-forth negotiation, another compromise was reached, with the NTIA allowing states to define the low-cost option within a range of acceptable minimum prices. But in the case of Virginia, the standoff lasted over a year, and inspired a Politico feature later cited by Vox as evidence of a generalized problem with BEAD bureaucracy, rather than a fight centered on affordability. Texas, which fought the NTIA over its low-cost proposal as well as requirements that implicitly restricted states’ ability to subsidize wireless providers, was the very last state to receive NTIA approval, right up against the November 2024 deadline. By that point, eight states were already in the process of committing funds to broadband projects; Louisiana had finished. 

For all the messiness of its implementation design, in early 2025 the BEAD program was shaping up to be a success. Three states had finished running their processes for awarding grants, and the results were promising: Provider participation had been robust, and in each state, more than 80 percent of awards were going toward fiber projects. Unserved areas, as determined by genuinely accurate maps, would finally receive future proof networks. 

Then, immediately following Donald Trump’s inauguration, the president signed an executive order that froze BEAD funds. The administration’s cynical rationale echoed the criticisms the abundance camp has made in earnest: “woke mandates” and “government red tape,” Commerce Secretary Howard Lutnick explained, had caused “delays and waste.” To rectify this, the Administration was delaying BEAD intentionally while it “revamp[ed]” the program. 

In June, the NTIA released updated program guidelines that shifted subsidies to Musk’s Starlink satellite service by forcing states to award grants to the lowest-cost providers that meet a minimal set of requirements. The change sets BEAD back to where it was in mid-2023. All states will need to submit, and have approved, updated program proposals; states that have already awarded subsidies to providers must rescind them. Analysts now predict that money won’t get out the door until 2026.  

While Starlink satellites have some advantages (if your fiber power goes out in an emergency you can still get a satellite signal as long as you have a home generator), they have greater disadvantages (slow signal during peak traffic, latency issues). Even worse, while fiber can last for decades with minimal upkeep, satellites need to be replaced every five to seven years, thus requiring billions of dollars more in ongoing federal subsidies. The new pro-Starlink guidelines give Trump leverage over Musk in their on-again, off-again feud. If the bulk of BEAD funds ultimately go through to Musk’s company, a program that was meant to close the digital divide for good will become just the latest in a series of shortsighted, wasteful investments.  

The digital divide should have been closed long ago. Indeed, it would never have opened in the first place had Washington dealt with broadband the way other rich democracies did, or the way FDR handled electricity back in the 1930s: essentially, by treating broadband as a public good. Instead, the advent of broadband coincided with a bipartisan swing toward deregulatory policymaking. Deregulation, in turn, empowered monopolies to capture the new broadband industry and use their economic and political clout to beat back nearly every attempt at sensible reform. That, not some liberal fetish with procedures and everything-bagel requirements, is why implementation of the BEAD program took so long and why, over decades, we’ve made so little progress in shrinking the digital divide despite the billions of taxpayer dollars we’ve thrown at the problem. 

Abundance liberals aren’t wrong that excessive rules and regulations pushed by progressive interest groups can, in some cases, slow down or increase the cost of building housing, infrastructure, and the like. But as we have seen, the BEAD program is not one of those cases. The slow and convoluted nature of BEAD is better understood as an example of what the Johns Hopkins University political scientist Steve Teles has called American “kludgeocracy.” A “kludge” is computer coding jargon meaning a “a clumsy but temporarily effective solution to a particular fault or problem.” U.S. policymakers, Teles argues, increasingly create kludgy government because of the tension between liberals who want to use federal power to address national problems and conservatives who want to siphon federal largess through private corporations and the states. The result is unnecessarily complicated federal programs that don’t work, that citizens can’t understand, and that “redistribute resources upward to the wealthy and the organized at the expense of the poorer and less organized.” 

It would be one thing if BEAD were the only Biden program abundance liberals have misunderstood and mischaracterized. But it’s not. For instance, they accused the administration of unwisely burdening its flagship legislation to reshore semiconductor manufacturing, the CHIPS Act, with everything-bagel requirements such as onsite daycare and training pathways for the disadvantaged. But as Joel Dodge reported in the Washington Monthly, those rules didn’t seem to have bothered the chip manufacturing firms, whose applications for funding outstripped the program’s budget by nearly two to one. Moreover, executives at the companies have said the legislation’s training and other requirements match their own strategies for opening up pipelines for hard-to-find talent. 

Time after time, Klein and other abundance thinkers blame liberal proceduralism for policy disasters that, on further inspection, turn out to be partly or entirely the result of corporate power. The United States lacks high-quality nationwide passenger rail service not just because of the soaring expense and snail-like pace of California’s high speed rail project—a special obsession of abundance pundits. The bigger reason is that hedge fund–controlled freight rail monopolies own nearly all existing U.S. rail infrastructure and refuse to give Amtrak trains reasonable right of way—and politicians in Washington are too afraid to take those corporations on. America lacks sufficient electricity transmission lines to carry renewable energy from rural areas where it’s produced to metro areas where it’s needed not just because of federal environmental laws, as abundance liberals argue. It’s also because investor-owned utilities use their control over the grid to block those new lines to protect the profits from their own fossil fuel–driven generation facilities. American health care is increasingly expensive not because of bottlenecks in the supply of doctors, as abundance liberals insist, but because the whole health care sector, from hospitals to pharmaceuticals to insurance, is now controlled by private monopolies.  

The story Ezra Klein told Jon Stewart, of the Biden administration deliberately bogging down its own broadband program in needless complexity, went viral in part because it fits an increasingly popular abundance liberal vision of what’s wrong with liberal governance and what Democrats need to do to win back power. The story told here, of how telecom monopolies are behind the failure of government to solve the digital divide, suggests a different strategy for Democrats, and has the advantage of being true.

Article link here.