ICLE Issue Brief

The Permission Layer: Clearing the Way for AI-Era Broadband

Executive Summary

AI and other next-generation technologies require more than data centers, computing power, and electricity. They also depend on communications networks installed underground, strung across utility poles, and built through corridors controlled by federal, state, local, and private gatekeepers.

Broadband networks require large upfront investments that providers may never recover if demand, technology, or regulation changes. They also create social benefits that providers cannot fully capture. Search, negotiation, permitting, and enforcement costs can therefore make valuable projects uneconomic, especially in rural and underserved communities.

Permitting compounds these risks. A single route may require federal reviews, state highway approvals, local permits or franchises, private crossing agreements, and access to thousands of poles. Each gatekeeper controls part of the project, but none bears its cumulative cost. Uncertain delays can matter more than fees because they postpone revenue, shorten an asset’s useful life, and create opportunities for hold-up.

Reform should make approvals faster and more predictable while preserving legitimate environmental, cultural, public-safety, and land-management interests. Policymakers should streamline federal reviews, impose firm shot clocks with deemed-granted remedies, limit fees to reasonable government costs, modernize cable franchising, and strengthen pole-access and cost-allocation rules. One-touch make-ready and dig-once policies can further reduce construction costs.

Recent agency actions and court decisions advance many of these goals, but future officials or courts could reverse them. Congress should codify their durable core. Stable rules would reduce investment risk across all projects and deliver the greatest gains where deployment remains hardest to finance.

I.               Introduction

Previous papers in this series have argued that technological leadership depends less on who invents the next breakthrough than on which jurisdictions build the full, low-friction infrastructure stack—spectrum, broadband, fiber, satellite, edge computing, data centers, and AI accelerators—that allows innovation ecosystems to form, cluster, and endure.[1] At its foundation, AI still relies on physical communications networks buried underground and strung along utility poles. But before providers can bury cable or mount an antenna, they must secure a web of permits and regulatory approvals.[2]

Broadband networks must cross land controlled by federal, state, and local authorities. Deployment typically requires zoning approvals, construction permits, and permission to occupy public rights-of-way.[3] When permitting is slow, inconsistent, or costly—or access to rights-of-way is delayed or priced unpredictably—deployment becomes more expensive and risky. Those barriers also postpone the point at which providers can begin recovering their investments.[4]

Access barriers can constrain new construction as much as a lack of funding. Even a provider ready to invest cannot build without timely, reasonably priced access to the corridors its network must follow.[5] Streamlined permitting and fair, predictable access to public rights-of-way will therefore help determine where—and how quickly—the infrastructure needed for next-generation applications gets built.

II.             The Broadband Investment Challenge

Debate over the infrastructure needed for artificial intelligence (AI) has focused largely on the data centers where models are trained and run and the electricity required to power them.[6] That focus is understandable. The scale of modern computing has strained electrical grids, renewed interest in additional generating capacity, and made data-center siting a major political issue.

Compute and electricity are, however, only part of the infrastructure these technologies require. AI and the immersive applications built on it cannot operate in isolation inside server farms.[7] They must exchange data with users and devices in real time, wherever those users happen to be. That requires high-capacity, low-latency communications networks capable of carrying large volumes of data with minimal delay. AI, augmented reality (AR), and virtual reality (VR) will therefore require substantial investment in broadband and related communications infrastructure.[8]

Broadband investment is not assured. Networks require large upfront expenditures that providers cannot recover once committed to a particular route or technology. Providers also capture only part of the social value their networks create. Search, negotiation, permitting, and enforcement costs can therefore make socially valuable projects uneconomic, while control over a single segment can give one gatekeeper the power to hold up an entire build.

Providers must invest amid uncertain demand, technological change, and regulatory risk. The prospect of obsolescence or new rules that reduce returns raises financing costs and discourages marginal projects, especially in rural and low-income communities. The following subsections examine these constraints and explain why predictable rules and well-designed incentives are essential to broadband deployment.

A.             The Economics of Broadband Investment

Broadband providers face significant barriers to entering and expanding within a market.[9] Chief among them are the high fixed costs that providers must incur before serving a single customer.[10] Recovering those costs depends on attracting enough subscribers, an uncertain prospect in a competitive market.[11]

Much of that investment also becomes a sunk cost—an expense that cannot be recovered if the project fails. Providers cannot easily remove fiber from an unsuccessful build and redeploy it elsewhere. They cannot recover the labor and regulatory costs at all. If subscriptions fall short, providers have few ways to limit their losses.

Economies of scale can offset these risks.[12] A provider can spread a network’s fixed costs across a larger customer base, while customers with a greater willingness and ability to pay can support higher revenue. Projects serving large or relatively affluent populations therefore tend to carry less risk than projects in rural or low-income communities.

Competition across different kinds of technologies, often called intermodal competition, adds another layer of uncertainty. Broadband markets once allowed one or two providers to capture large shares because consumers had few alternatives. Today, providers using different technologies compete across price, speed, reliability, coverage, and other dimensions.[13]

Home broadband may reach consumers through coaxial cable, fiber optics, fixed wireless service delivered over mobile networks, or low-Earth-orbit (LEO) satellites. Mobile broadband can offer enough capacity and performance for some consumers to forgo a fixed home connection altogether. At the same time, traditionally fixed providers have added mobile services so customers can remain connected outside the home. Because these technologies have different costs and infrastructure needs, multiple competing networks may prove more efficient than a single set of pipes.[14]

Competition and deployment both depend on continued network investment. In a competitive capital market, investment flows toward projects whose expected return, adjusted for risk, exceeds their cost.[15] Broadband providers calculate that return primarily from monthly subscription fees, supplemented by advertising and ancillary services.[16]

Broadband also creates substantial value that providers cannot capture through those revenue streams.[17] These spillover benefits—known as positive externalities—reach millions of people and many sectors, while providers bear the concentrated cost of deployment. A project is socially worthwhile when its private revenue and broader social benefits together exceed its cost.[18] Yet a rational provider will not build if its own expected revenue cannot cover that cost. It cannot charge for every increase in a farmhouse’s value, improvement in a student’s grades, or savings in a rural hospital’s budget.

Communities lose those benefits when broadband investment falls short. Real-estate prices can help measure broadband’s value because they incorporate local access to education, health care, employment, and other amenities.

Students without reliable home internet perform worse on standardized tests than their connected peers, which can reduce college enrollment and degree attainment.^[19] Telemedicine can lower patients’ costs and make rural living practical for more people. It also helps rural health care facilities reach more patients and connect with specialists at major medical centers without employing them full time. Patients, in turn, can obtain care without traveling to a facility. Broadband also allows higher-earning remote workers to move to lower-cost rural areas, bringing income and spending into those communities.

These positive externalities create an underinvestment problem. Because providers cannot capture broadband’s full social value, expected private returns may not justify deployment even when a project would benefit society overall.[20]

AI, AR, and VR will intensify that problem. These technologies require high-capacity connections with minimal delay, reliable fiber links connecting local networks to the broader internet, processing located closer to users, and access to large-scale computing resources. Many applications must run AI models and exchange data continuously in real time, placing simultaneous demands on communications networks.

The benefits of these applications will extend far beyond what any one provider can monetize. Without policies that encourage investment, the communities least likely to attract private capital may also lack the infrastructure that AI, AR, and VR require. That outcome would widen the gap between served and unserved regions just as advanced connectivity becomes more important to economic competitiveness. Policymakers seeking broadband’s broader benefits must therefore give providers stronger incentives to invest in unserved and underserved areas.

B.             Transaction Costs and the Broadband Holdout Problem

In a world without transaction costs, parties would bargain toward an efficient allocation of resources regardless of who initially held the relevant property rights.^[21] Applied to broadband, this principle suggests that a fiber provider will secure access to a utility pole, strip of land, or railroad crossing whenever it values that access more than the owner values excluding it. The parties will strike a deal, and the network will get built.

In practice, transaction costs—the costs of identifying the right parties, negotiating an agreement, and enforcing it—can make broadband deployment prohibitively expensive. These costs generally take three forms:

• Search and information costs. Providers must identify who owns or controls each asset along a proposed route. That may require researching the ownership of thousands of utility poles divided among electric utilities and legacy telephone companies, tracing easements across private land, or determining which government agency controls a particular right-of-way.
• Bargaining and decision costs. Providers incur legal fees, administrative expenses, and delays while negotiating access. A single project may require agreements with municipalities, private landowners, homeowners associations, and railroads, each subject to different legal requirements and each able to demand payment for its consent.
• Policing and enforcement costs. Providers must monitor compliance and enforce their contractual and legal rights. These costs include policing pole-attachment agreements, protecting easement boundaries against encroachment, and litigating disputes over “fair and reasonable” rates.

Asset specificity can magnify each of these costs. The term describes investments tailored to a particular project that cannot easily be redirected elsewhere. Suppose a provider plans a 10-mile fiber route and obtains every approval except the one needed for the final mile. The owner controlling that segment can threaten the entire project and demand a payment far above the value of the access itself. The provider must either pay, redesign the route, or abandon investments already made along the other 9 miles.

For communications networks, transaction costs can exceed the price of the equipment itself. A mile of fiber-optic cable containing 144 strands costs roughly $5,000 to $8,000.[22] Make-ready work—the work needed to prepare a utility pole for a new attachment—can cost $600 to $6,000 per pole. Replacing a pole that cannot accommodate another attachment can cost $5,000 to $15,000.[23] Permitting and access to public rights-of-way can add $20,000 to $40,000 per mile, while railroad-crossing fees can make an otherwise viable project uneconomic.[24] Major projects also require substantial spending on construction crews, traffic control, and coordination among the parties working in public rights-of-way.

Costs accumulate when many parties hold vetoes over a single linear project and each seeks the greatest possible payment for its approval. Each gatekeeper considers the value of its own permission, not the cumulative burden on the network. The result resembles a linear anticommons—a project that requires consent from a long chain of independent parties, any one of which can block it.

These pressures can also push providers toward technologies that are less suitable for a particular community but easier to deploy. A fixed wireless project, for example, may require one tower-site approval, while a fiber build may require permission to attach equipment to 5,000 poles. Avoiding that chain of approvals can determine the technology a provider chooses, even if another option would deliver better service to consumers.

Government policy can raise or lower these transaction costs. When regulators can change rules abruptly or apply them subjectively, providers must account for the risk of an unfavorable policy shift.[25] That uncertainty weighs especially heavily on broadband projects because they require large upfront investments, take years to repay, and consist largely of sunk costs.[26]

A decision by the Federal Communications Commission (FCC) to subject broadband to common-carrier, utility-style regulation could reduce the expected return on an existing investment. So could changes by the National Telecommunications and Information Administration (NTIA) that tighten performance benchmarks under the Broadband Equity, Access, and Deployment (BEAD) Program. Providers must price these risks into a project even if the rules never change. Regulatory uncertainty can therefore sharply increase deployment costs and deter investments that would otherwise proceed.

C.             Demand, Obsolescence, and Regulatory Risk

Alongside high deployment costs, broadband providers face uncertainty about future demand. The FCC currently uses download speeds of 100 megabits per second and upload speeds of 20 megabits per second as its benchmark for fixed broadband.[27] That benchmark supports many current applications, but individual needs vary widely. A household with three children streaming separate programs requires more capacity than a single occupant who uses the internet mainly for email. Providers must design networks around expected use across their customer base, not the needs of any one household.

That task becomes harder as technology and consumer behavior change. AI, AR, and VR could sharply alter usage patterns and increase demand for network capacity. Providers can draw on lessons from the deployment of fourth-generation (4G) wireless networks and other technological shifts, but historical data cannot eliminate uncertainty about future demand.

Providers must also choose among technologies whose relative costs and capabilities will continue to change. Advances in LEO satellite service, sixth-generation (6G) wireless networks, or other technologies could undermine the economics of infrastructure built today. Physical assets could then become stranded, meaning they would no longer generate enough revenue to repay the debt used to finance them.

Investors account for this risk of obsolescence. If they cannot be confident that an asset will remain commercially useful for 10 years, they will demand a faster return of capital. That shorter recovery period can lead to higher prices, delayed construction, or no deployment in marginal markets. Regulation can compound the problem. Rules that calculate returns from historical costs over long recovery periods may fail to account for infrastructure that becomes obsolete before investors recover their capital.

Providers also face the risk of regulatory hold-up.[28] Before construction begins, providers retain leverage because capital is mobile. If one jurisdiction imposes excessive burdens, they can invest elsewhere. Governments may seek to attract investment by offering subsidies, regulatory stability, forbearance, or favorable franchise terms.

Once a provider has built the network, the balance of power shifts. The provider cannot move its fiber, recover most of its costs, or credibly threaten to leave. Regulators can then change the terms governing the investment by imposing rate controls, requiring competitors to receive access to the infrastructure, or adopting other rules that reduce the provider’s expected return.

This mismatch between regulatory assumptions and investment realities can produce underinvestment. Regulators may view broadband as a mature monopoly utility. In competitive markets, though, broadband deployment often carries risks closer to those of a technology startup than a traditional water or telephone company. Policy that ignores those risks will discourage investment precisely where uncertain demand already makes deployment hardest to justify.

III.           The Costs and Risks of Permitting

Broadband deployment requires costly, long-term investments that providers cannot easily recover or redirect. Regulators seeking to encourage deployment should therefore reduce avoidable costs and risks. Permitting offers a prime opportunity for reform.

A permit is more than a line item in a project budget. An outside party with discretion—and often an effective veto—controls both its price and its timing. Providers must identify the relevant authorities, navigate overlapping review processes, negotiate access and conditions, and comply with ongoing obligations. Each step adds transaction costs. Uncertain timelines, subjective standards, and litigation also delay revenue, shorten an asset’s useful economic life, and raise the return investors require. When a route needs approval from multiple gatekeepers, any one of them can hold up the entire build. Permitting can therefore become both one of a broadband project’s largest transaction costs and one of its greatest sources of investment risk.

A.             How Permitting Raises Transaction Costs

Permitting maps closely onto the three categories of transaction costs identified in Section II.B: finding the relevant decision-makers, securing their approval, and complying with or enforcing the resulting agreements.

The first category consists of search and information costs.[29] A rural broadband route may cross federal land administered by the Bureau of Land Management (BLM) or the U.S. Forest Service. That can trigger a right-of-way grant, environmental review under the National Environmental Policy Act (NEPA), historic-preservation review under Section 106 of the National Historic Preservation Act (NHPA), and tribal consultation.

The same route may follow or cross a state highway right-of-way, then pass through a patchwork of county and municipal permitting regimes as it approaches consumers. It may also require access to a railroad crossing controlled by a private carrier or to easements across private land. Before applying for approval, the provider must identify who controls each segment, which law governs, and which overlapping review processes apply. In areas where the federal government owns much of the land, a rural route may remain within the federal permitting system for most of its length.

The second category consists of bargaining and decision costs. Once a provider identifies the relevant authority or landowner, it must secure the right to enter the right-of-way and build. Each entity may impose its own applications, fees, engineering studies, surveys, and substantive standards. The provider must satisfy each one separately.

Many of these decisions involve discretion rather than a ministerial review based on an objective checklist. Providers must therefore do more than pay a posted fee. They may need to negotiate terms, commission studies, redesign routes, and accept project-specific conditions.[30] Multiplied across every jurisdiction along a route, these requirements create a substantial administrative burden before workers install a single strand of fiber.

The third category consists of policing and enforcement costs. After securing a permit, providers must comply with restoration requirements, seasonal work windows, inspection regimes, and other conditions. They must monitor those obligations and may need to challenge denials or disputed conditions through administrative proceedings or litigation.

Each authority and landowner along a route controls an independent point at which the project can stall. Each may seek to maximize the value of its own permission without accounting for the cumulative burden on the deployment.[31] A railroad may charge far more for a perpendicular crossing than the cost of the materials disturbed. A single county may withhold a right-of-way permit. Either can stop a project that has secured every other approval.

This dynamic helps explain why providers sometimes choose technologies that require a single approval over those that require thousands. Yet even that single approval may trigger NEPA and Section 106 review when a project crosses federal land. The holdout risk also grows as construction progresses. The last party controlling an unapproved segment gains bargaining power because the provider cannot readily redeploy the capital already sunk into the rest of the route.

On many routes, obtaining permission costs more than the hardware that permission allows the provider to install. The process meant to authorize construction can become costlier than construction itself.

B.             Delay, Discretion, and Litigation Risk

Permit fees account for only part of the risk that broadband providers must consider when investing in infrastructure. Delays, discretionary standards, and litigation can impose even greater costs.

Delay matters because broadband projects require large upfront investments in assets that providers cannot readily redeploy. Project economics depend heavily on when revenue begins relative to when costs accrue. Permitting pushes revenue further into the future, while financing and development costs begin accumulating before construction starts.

Providers must carry those costs through a review process of uncertain length. Each month of delay also consumes part of an asset’s useful economic life while the risk of technological obsolescence continues to grow. A 10-year asset that spends three years in review and litigation must repay its debt from seven years of revenue rather than 10. If a disruptive technology emerges in year five, the delay will have consumed much of the period during which the asset remained competitive. Permitting delays thus compress the recovery period on which the investment depends.

Discretion adds another layer of risk. Permitting standards often turn on open-ended questions: What constitutes a “reasonable period of time”? Does a denial rest on “substantial evidence”? Does a project qualify for a categorical exclusion from full environmental review?[32] A provider can incorporate a fixed fee into its budget. It cannot price a discretionary decision with the same confidence. Instead, it must estimate a range of possible outcomes and require a higher expected return to compensate for that uncertainty.

Completing agency review does not eliminate the risk. NEPA is the most frequently litigated federal environmental statute.[33] Research indicates that national advocacy organizations, rather than directly affected local parties, have filed most NEPA suits in recent years.[34] Litigants can challenge a categorical exclusion or a finding of no significant impact, adding years to a project that has already cleared agency review.

The Supreme Court narrowed that exposure last year in Seven County Infrastructure Coalition v. Eagle County.[35] The Court directed lower courts to give agencies substantial deference when reviewing the scope of their environmental analyses. It also held that NEPA generally does not require agencies to analyze attenuated upstream or downstream effects tied to separate projects in different places or at different times. The decision should narrow both the required analysis and the available grounds for litigation. But it did not eliminate NEPA review, its associated delays, or the possibility that a court could enjoin a project after an agency approves it. Investors must still account for those risks before committing capital.

Taken together, permitting adds fragmented veto points to projects that already require large, sunk investments exposed to technological obsolescence. It then makes both the duration and outcome of the process uncertain. Providers price that risk into every project, even when no delay or denial ultimately occurs. A project that proceeds on schedule still costs more ex ante because investors had to account for the possibility that it would not.

These burdens do not fall evenly. Delay and denial risk raise the hurdle rate—the minimum expected return needed to justify an investment—and eliminate marginal projects first. The effects therefore fall hardest in communities where deployment promises substantial social value but weak private returns. Streamlining permitting can help close that gap by allowing more of broadband’s diffuse social benefits to support actual investment.

IV.          The Broadband Permitting Stack

Section III explained how permitting operates as both a transaction cost and a source of investment risk. This section maps those burdens across federal, state, and local governments, as well as the separate regime governing pole attachments.

Although the requirements differ, their costs recur in familiar forms: application fees, recurring rents, in-kind obligations, consultant expenses, bonding and insurance requirements, compliance costs, and, above all, the carrying cost of delay. Federal environmental, historic-preservation, tribal, water, habitat, and land-use reviews may overlap with state highway-access and environmental rules. Local franchising, right-of-way, siting, street-cut, and inspection requirements add further approvals. Aerial deployments must also navigate pole-access, make-ready, and replacement processes controlled by utilities and existing attachers.

These layers compound rather than merely accumulate. Independent decision-makers control sequential veto points without bearing the project’s total cost. The resulting delays and uncertainty raise the return needed to justify investment and threaten marginal projects first—especially rural and underserved builds with limited private returns but substantial social value.

A.             Federal Environmental and Land-Use Permitting

A broadband project may require federal approval because it receives federal funding or because a federal agency must authorize part of it. The most consequential federal process is often review under NEPA, which requires agencies to assess the environmental effects of major federal actions.[36] For broadband projects, that duty may arise when a build receives federal support, including a BEAD subgrant, or when the FCC’s licensing authority makes a wireless deployment a federal action.[37]

The required analysis varies with the project’s likely effects. Agencies may apply a categorical exclusion to routine work requiring minimal documentation, prepare an environmental assessment to determine whether significant effects are likely, or require a full environmental-impact statement. Providers may need to hire consultants and wait for agency review, but uncertainty and litigation often impose the larger costs.[38]

Congress imposed deadlines and document-length limits in 2023, and the Supreme Court further narrowed NEPA exposure in Seven County Infrastructure Coalition v. Eagle County.[39] The Court directed lower courts to defer substantially to agencies and rejected demands that they analyze attenuated upstream and downstream effects tied to separate projects. These changes should limit the scope of review and the grounds for litigation. Neither eliminated NEPA’s underlying requirements or the delays they can create.

Section 106 of the NHPA adds a separate historic-preservation review. It requires agencies to assess a project’s effects on properties listed in, or eligible for, the National Register of Historic Places and to consult the relevant state or tribal historic preservation officer.[40] A single consultation can take six months or longer. In areas rich in historic or cultural resources, that review can determine the schedule for the entire build.[41]

Projects affecting tribal lands or cultural resources may require broader consultation. Section 106 reaches beyond tribes whose present-day lands contain the project. Agencies must also consult any tribe that attaches religious or cultural significance to an affected property, including ancestral sites far from current reservation boundaries.^[42] For wireless facilities, the FCC’s notification process may require consultation with many tribes over one site.

This process serves genuine interests in tribal sovereignty and cultural preservation. It also adds per-site fees and another uncertain scheduling variable. Providers and some policymakers contend that some tribes have used review fees partly to generate revenue rather than solely to fund oversight, allowing revenue to rise with the pace of deployment.[43] Whatever the purpose of a particular charge, providers must include both its amount and its timing in the cost of a build.

Routes crossing wetlands, streams, or navigable waters face another layer of federal review, administered primarily by the U.S. Army Corps of Engineers (Corps).[44] The Corps offers nationwide permits for routine linear crossings that remain within defined limits on the area and degree of disturbance.[45] This streamlined process can keep the burden modest.

Costs and delays rise sharply when a crossing exceeds those limits and requires an individual permit. That process may demand a formal wetland delineation, public notice, and a review lasting more than a year. Section 7 of the Endangered Species Act can add consultation with the U.S. Fish and Wildlife Service, biological surveys, and seasonal work restrictions.[46] Those restrictions may affect a project’s schedule more than its budget by halting construction during breeding or migration periods.

As with NEPA review, uncertainty often costs more than the permit fee itself. Providers must account for the possibility of an individual permit, extended consultation, or a seasonal restriction at every crossing, including those that ultimately qualify for streamlined treatment.

Projects crossing federal land face a separate right-of-way process. This is especially common in the western United States, where the federal government owns large expanses of land. The Federal Land Policy and Management Act authorizes the BLM to grant rights-of-way across public lands.[47] The Forest Service, National Park Service, and Department of Defense administer separate regimes for their holdings.

These approvals can generate two fee streams: cost-recovery charges that reimburse agencies for processing and monitoring, and rental fees based partly on the population a facility serves. A 2024 BLM rule streamlined the communications-use program, required electronic filing, and established a 270-day deadline for decisions.[48] Those reforms implicitly acknowledged that the earlier process imposed delays substantial enough to deter deployment.

B.             State Permitting and Regulatory Fragmentation

The most common state-level requirement is permission to occupy or cross state highway rights-of-way, typically through an encroachment or utility-accommodation permit.[49] Because each state establishes its own regime, a provider operating across several states must navigate distinct fee schedules, technical standards, applications, and review timelines. This fragmentation raises search and bargaining costs before the provider pays a permit fee.

Many states also require a carrier to obtain a certificate of public convenience and necessity, or to register with the utility commission, before operating or occupying the right-of-way. A related designation, eligible-telecommunications-carrier status, is a precondition for most subsidy programs and carries obligations of its own.[50]

State law may duplicate or displace federal requirements. Roughly 16 states have enacted their own environmental-review statutes, often called “little NEPAs.”[51] California’s Environmental Quality Act is the best-known example.[52] These laws can add a parallel track of environmental study and litigation to projects already subject to NEPA.

In roughly 20 states, Section 224’s reverse-preemption provision displaces the federal pole-attachment regime. State rules therefore govern one of the largest infrastructure-access costs across much of the country. Many states have also replaced local cable franchises with statewide video-franchise systems.[53]

Other state requirements include dig-once policies that coordinate conduit installation, one-call rules requiring excavators to locate buried utilities, and pavement-restoration standards. These measures add procedural steps, but thoughtful design can reduce overall costs by coordinating construction, preventing damage, and avoiding repeated excavation.

C.             Local Franchising and Siting Controls

At the municipal level, cable operators must obtain a local franchise under Title VI of the Communications Act. Federal law caps the franchise fee at 5% of gross revenue.[54] Localities have also required in-kind contributions, including public-access channels, capacity on institutional networks serving public facilities, free service to public buildings, and buildout commitments. The FCC’s 2019 order counts many of these contributions toward the 5% ceiling, but disputes over additional obligations persist.[55]

Negotiating and renewing franchises can itself delay deployment. Reform proposals have therefore paired franchising requirements with a shot clock—a deadline for government action—and a deemed-granted remedy that automatically approves an application when the locality misses that deadline.[56]

Noncable providers generally use permits rather than franchises to access local rights-of-way.[57] Section 253 bars state and local requirements that prohibit or effectively prohibit telecommunications service. It also preserves local authority to manage public rights-of-way and require fair and reasonable compensation.[58] The FCC has generally treated fees closely tied to a locality’s actual costs as the most appropriate measure under that framework. Enforcement remains uneven, and some localities continue to impose above-cost or gross-revenue fees.

Above-ground facilities—including cabinets, equipment huts, small-cell antennas, and towers—may require conditional-use permits, variances, site-plan and aesthetic review, public hearings, and compliance with height and spacing limits. Some jurisdictions also require providers to place facilities underground. Section 332(c)(7) preserves substantial local authority over wireless siting while prohibiting decisions that ban or effectively ban wireless service.[59] FCC shot clocks limit the time for review, but the process can still take months or years.

Providers must obtain additional public-works approvals before trenching or boring. These may cover street cuts, lane closures, traffic control, and pavement restoration. Rules addressing pavement damage and moratoria on cutting newly paved streets can add large and highly variable costs.[60] Building and electrical permits and inspections may apply as well. Some localities also impose temporary siting moratoria that consume much or all of the applicable shot-clock period.[61]

Wireless facilities operate under a more developed federal timing framework. Under Section 332(c)(7), localities presumptively have 90 days to act on collocation applications (which seek to place equipment on existing structures) and 150 days to review new sites.[62] Missing those deadlines constitutes a “failure to act,” but it does not automatically approve the application. The provider must seek relief in court.

The FCC’s 2018 Small Cell Order shortened the review periods to 60 days for collocations and 90 days for new small wireless facilities. It also established presumptively reasonable fee levels.[63] Section 6409(a) requires localities to approve eligible modifications that do not substantially change the physical dimensions of an existing tower or base station. FCC rules apply a 60-day shot clock and a self-executing deemed-granted remedy to those requests.[64]

These provisions are among the strongest timing protections in the current permitting system. Their existence highlights how little comparable discipline governs many other approvals. Even here, localities can pause or restart the clocks by declaring applications incomplete or requiring supplemental studies.[65]

D.            Pole Attachments and Make-Ready Costs

To deploy aerial fiber, a provider must attach its lines to utility poles it usually does not own. Federal law, rather than private bargaining alone, largely governs the terms of that access. Section 224 of the Communications Act requires pole-attachment rates, terms, and conditions to be “just and reasonable.” The FCC administers that standard through formula-based rates in states that have not certified their own regulatory regimes.[66]

This framework aims to promote investment and facilitate deployment, but it does not cover poles owned by municipalities or rural electric cooperatives.[67] Providers seeking to use those poles must negotiate without a federal regulatory backstop.

The permit or rental rate represents only part of the cost. The larger and less predictable expense often comes from make-ready work: the rearrangement, relocation, or reinforcement of existing equipment needed to create space for a new line.[68] This work can cost several hundred to several thousand dollars per pole. It may also take months, because each company with equipment already on the pole must move its facilities in sequence.[69]

When a pole is too short or weak to support another attachment, make-ready work may require replacing it at a cost of roughly $5,000 to $15,000.[70] Utilities have historically charged the entire replacement cost to the new attacher, even when the pole had deteriorated, neared the end of its useful life, or otherwise required replacement.[71]

The FCC has found that a utility may not impose the full cost on a new attacher that did not solely cause the replacement. Instead, the pole owner must bear costs attributable to preexisting violations or a replacement it already had an obligation to make.[72] This approach prevents utilities from using a new attachment request to shift the cost of overdue maintenance onto the requesting provider.

The FCC has also adopted one-touch make-ready (OTMR) rules for simpler work. OTMR allows one qualified contractor to move existing equipment in a single pass rather than waiting for a succession of incumbent contractors.[73] The process can reduce delays that might otherwise consume much of a year.[74]

E.              How Permitting Burdens Compound

Each permitting layer can impede deployment on its own. Together, they impose a much larger burden. Independent authorities and landowners control separate veto points, yet none has an incentive to account for the project’s total cost. The provider therefore faces more than the price a single, coordinated decision-maker might charge. It must bear the cumulative demands of many actors, each seeking value for its own approval.

Each layer also adds more than a fee. It introduces a delay of uncertain length. Because broadband projects require large, sunk investments in assets exposed to obsolescence, those delays raise financing costs, shorten the period available to recover capital, and increase the project’s risk premium.

These burdens fall unevenly. Each additional cost or delay raises the hurdle rate, so the marginal project fails first. That project is disproportionately likely to serve a rural or underserved community where expected private returns are thin but the broader social benefits are substantial. Permitting barriers thus inflict their greatest harm where deployment may create the most social value. That mismatch supplies the central case for reform, to which the next section turns.

V.             A Reform Agenda for Faster, More Predictable Deployment

Effective reform must address more than posted fees. Uncertain delays, fragmented approvals, and the risk of regulatory reversal often do more damage because providers must price those risks into every project. These burdens fall hardest on marginal builds, especially networks serving rural and underserved areas.

Since mid-2025, the FCC’s “Build America” agenda has targeted many of these barriers. The agency has opened proceedings on wireless siting and wireline rights-of-way, revised its pole-attachment rules, and proposed changes to its environmental and historic-preservation reviews.[75] Further reforms should streamline federal reviews, constrain state and local fees and timelines, modernize cable franchising, expand pole-access protections, resolve pole-replacement disputes, coordinate fiber installation with roadwork, and codify durable rules that investors can rely on. Together, these changes would reduce the cost and uncertainty of building the communications networks that AI and other next-generation applications require.

A.             Streamlining Federal Reviews

Federal review regimes impose their greatest costs through delay, uncertainty, and litigation rather than permit fees. Reform should therefore limit full review to projects likely to cause significant effects, accelerate the reviews that remain, and give approved projects greater finality.

Categorical exclusions offer the most efficient tool for routine deployments. When experience shows that a defined class of projects does not ordinarily cause significant environmental effects, an exclusion can avoid repetitive case-by-case assessments while narrowing both direct costs and the grounds for litigation.[76] Programmatic agreements can serve the same function under Section 106. Rather than reconsidering recurring issues site by site, agencies, preservation officials, and tribes can establish one framework governing an entire category of deployments.[77]

Recent Supreme Court decisions have narrowed the governing federal obligations. In Seven County Infrastructure Coalition v. Eagle County, the Court reinforced agency discretion under NEPA and limited demands to analyze attenuated effects tied to separate upstream or downstream projects.[78] In Sackett v. EPA, the Court narrowed federal wetlands jurisdiction under the Clean Water Act.[79] These decisions provide meaningful limits, but their practical effect still depends on how agencies implement them and how courts apply them in later cases.

Congress and agencies should translate those boundaries into clear, durable rules. Agencies should adopt categorical exclusions for well-defined classes of communications facilities that reliably produce no significant effects. Firm review deadlines should govern the projects that remain, with automatic approval when an agency fails to act on time.

Reform must also respect the limits courts have already imposed. In United Keetoowah Band of Cherokee Indians v. FCC, the D.C. Circuit vacated the FCC’s attempt to exempt small wireless facilities from NEPA and Section 106 review.[80] The court found that the FCC had not adequately justified its public-interest determination or addressed potential effects on historic and tribal resources. The decision shows that streamlining measures need a well-developed record that squarely addresses environmental, historic-preservation, and tribal interests.

The same principles apply to projects crossing federal land. Agencies should use master right-of-way agreements, standardized applications, cost-based rather than value-based charges, and firm decision deadlines. The BLM’s 2024 communications-use rule offers a useful model for making federal approvals faster and more predictable without abandoning substantive review.

B.             Cost-Based Fees and Self-Enforcing Deadlines

For facilities outside the FCC’s small wireless framework, state and local reform should focus on two issues: fees and review deadlines. In both cases, the rule matters only if providers can enforce it effectively.

Fees for construction permits and access to public rights-of-way should recover the government’s actual costs. They should not capture the value of the provider’s investment or exploit the government’s monopoly control over the corridor. Value-based and gross-revenue fees operate as taxes on deployment. They fall hardest on thin-margin rural projects that policy should encourage.

Presumptively reasonable benchmarks can give providers and localities a clear reference point. The FCC’s small-cell safe harbors—$500 for certain applications and $270 per site each year—illustrate the approach. A locality could impose a higher fee, but it would bear the burden of showing why its actual costs justify the increase. Governments should not be able to evade fee caps by demanding in-kind contributions instead of money.

Deadlines require equally strong enforcement. A shot clock that merely allows a provider to sue after time expires preserves much of the delay and uncertainty it aims to eliminate. The provider must still account for the expense, duration, and risk of litigation.

A self-executing “deemed granted” remedy works differently. Under Section 6409(a), an eligible application receives automatic approval when the locality fails to act within the prescribed period. The provider therefore knows it will receive an answer by a date certain, regardless of the locality’s diligence. Extending this model to ordinary siting, right-of-way, and construction permits would provide the strongest available protection against permitting delay.

The American Broadband Deployment Act of 2025 and the BROADBAND Leadership Act reflect this approach by pairing deadlines across multiple permitting layers with cost-based fee limits.[81] The FCC is pursuing similar goals under its Section 253 and Spectrum Act authority, including a proposed “rocket docket”—an expedited process for resolving preemption disputes.[82]

Reform should still preserve local authority over legitimate concerns such as public safety, traffic management, and aesthetics. The aim should be to discipline fees and timing while allowing reasonable, evenhandedly applied substantive conditions.

C.             Reforming Local Franchising

Title VI’s local cable-franchising regime predates modern competition among fiber providers building their own networks. Today, it can operate as both a toll and a source of delay for video-capable broadband entrants. Moreover, the 5% gross-revenue franchise-fee cap cannot serve as a meaningful ceiling if localities may extract additional value through in-kind demands.

The FCC’s 2019 order, which counts many in-kind obligations toward the cap, was a sound response. Congress should codify that approach so it does not depend on changing FCC majorities. Governments should not be able to evade a monetary cap by demanding services, facilities, or other contributions instead.

The broader reform is to subject franchising to the same timing rules as other permits. The American Broadband Deployment Act of 2025, for example, would establish a 120-day deadline so localities cannot use a process negotiated with an incumbent to delay a second provider.[83] States that have replaced separate local regimes with a single statewide video franchise offer a model that other jurisdictions could expand.

Franchising can also obstruct providers that do not offer cable service. Some localities have conditioned ordinary right-of-way permits on a noncable provider first obtaining a cable franchise. This practice erases the distinction between permits and franchises and imports a slower, more demanding process into routine right-of-way access.

Gateway Fiber encountered this problem in Minnesota. After the company sought permits to build fiber in several cities, Maple Grove said it would defer action until Gateway completed the cable-franchising process. Gateway is not a cable operator, yet the city effectively refused to issue the right-of-way permits without a cable-franchise agreement.[84]

Gateway petitioned the FCC in June 2025 for a declaratory ruling preempting the requirement under Section 253(d). The company argued that Maple Grove’s refusal to act “has the effect of prohibiting” telecommunications service in violation of Section 253(a). According to Gateway, the city’s position prevents the company from operating its existing network and blocks expansion into new areas.[85]

The Wireline Competition Bureau placed the petition on public notice in July 2026 and situated the dispute within the FCC’s broader Section 253 wireline-deployment inquiry.[86] Whatever the Commission decides, the episode exposes a broader vulnerability. The line between a permit and a franchise can turn on disputed questions about a provider’s regulatory classification and a locality’s authority under state law. A locality can exploit that uncertainty to repurpose a legacy franchising process as a gatekeeping tool against a new fiber entrant—the kind of last-mile chokepoint that Section 253 is meant to police.[87]

D.             Closing the Pole-Attachment Gap

Because access to existing poles often represents one of the largest costs of a wireline build, policymakers should extend Section 224’s guarantee of just, reasonable, and timely access more broadly. Recent FCC orders have improved the federal regime. OTMR allows one approved crew to complete simple work rather than waiting for multiple attachers to act in sequence. The FCC’s July 2025 order also established access timelines for large and bulk applications, required parties to meet and confer, expanded self-help when utilities miss deadlines, and accelerated approval of qualified contractors.[88]

These reforms reach only part of the problem. Section 224 exempts poles owned by municipalities and cooperatively organized utilities, placing them outside the FCC’s jurisdiction.[89] Municipal utilities and rural electric cooperatives own a substantial share of the distribution poles in rural America—the same unserved and underserved areas targeted by federal broadband programs and where deployment economics are already most difficult.[90]

Providers seeking access to municipal or cooperative poles therefore face rates and timelines unconstrained by federal rules. In many states, no comparable state protections apply. Congress should extend Section 224—or an equivalent set of federally enforceable deadlines and cost-allocation rules—to these poles. Doing so would narrow the gap between the places where connectivity is most needed and those where access rules make deployment feasible.

Any extension should still allow pole owners to recover legitimate costs. Municipal and cooperative utilities are nonprofit entities owned by citizens or members, so ratepayers ultimately bear the cost of below-cost attachment rates. Reform should prevent monopoly pricing and unreasonable delay without forcing those ratepayers to subsidize broadband attachers.

E.              Enforcing Pole-Replacement Cost Sharing

Broadband providers and pole owners increasingly dispute who should pay when a pole must be replaced to accommodate a new attachment. The FCC’s December 2023 declaratory ruling clarified that an attachment does not “solely” necessitate replacement when the utility has already red-tagged the pole (marked it for replacement) or when a change in the utility’s construction standards requires the upgrade. In those circumstances, the new attacher should not bear the full cost.[91]

The central challenge is now enforcement. The rule helps providers only if they can invoke it quickly and at reasonable cost. Otherwise, the utility’s initial allocation governs while the dispute proceeds, and the FCC’s cost-sharing principle may have little practical effect.

Three complementary reforms could close that gap. First, the FCC should resolve these disputes through an expedited complaint process or “rocket docket” governed by firm deadlines.[92] Where authorized, fee shifting could further reduce the cost of asserting a valid claim. Faster and less expensive adjudication would make the proper allocation—not the utility’s opening demand—the operative result.

Second, Congress should codify the cost-allocation principle. Codification would protect it from reconsideration petitions and changes in FCC leadership, giving providers a more stable basis for investments expected to last 20 years.

Third, targeted public reimbursement funds could help resolve cases in which genuine cost sharing is warranted. Limited support for qualifying pole replacements could break deployment impasses without requiring the parties to litigate responsibility pole by pole.

F.              Coordinating Broadband with Roadwork

Excavation accounts for much of the cost of underground fiber, making coordination with planned roadwork one of the highest-return deployment reforms. A 2021 Federal Highway Administration rule requires state transportation departments receiving federal-aid highway funds to designate a broadband utility coordinator and establish processes for registration, notification, and coordination. A January 2026 Government Accountability Office review found that most states had made progress implementing those requirements, although their effects on deployment remain difficult to measure.[93]

Weak “dig-once” policies merely notify providers of planned roadwork and give them an opportunity to install conduit. That approach helps only providers whose construction plans happen to align with a particular project. Middle-mile and backhaul routes, which connect local networks to the broader internet, frequently will not align.

Stronger policies require road sponsors to install spare conduit whenever a federally funded project opens the ground. The sponsoring agency can then own and lease that conduit to broadband providers. This approach preserves far more of the roadwork’s future value by allowing providers to install fiber later without excavating again. Proponents estimate that installing conduit during planned construction adds only a few percentage points to the road project’s cost.

G.            Making Reform Durable

As outlined above, many of the most consequential recent reforms rest on agency orders or judicial interpretations. But both can change. Investors will hesitate to commit capital for 10 years based on a rule that a new FCC majority, a reconsideration petition, or a later court decision could undo in five.

That uncertainty limits the value of agency streamlining. Even a sound reform will attract less investment if providers must account for the risk that it will disappear before they recover their costs. Congress should therefore codify the durable core of these reforms.

Codification would do more than settle individual policy disputes. By reducing uncertainty across every project, it would lower the return investors require to commit capital. That reduction would matter most for marginal projects, especially in rural and other hard-to-reach areas where modest changes in cost or risk can determine whether a network gets built.

VI.          Conclusion

Debate over the infrastructure needed for next-generation applications often focuses on data centers and electricity. Yet AI, AR, and VR can reach users only through communications networks built across public land, local rights-of-way, railroad crossings, and utility poles—one approval at a time.

Broadband investment already carries high fixed costs, uncertain demand, obsolescence risk, and long recovery periods. Fragmented permitting compounds those risks. Each additional gatekeeper can impose fees, delay construction, or block a route, while no single authority bears the project’s cumulative cost. Providers must price that uncertainty into every build, even when approvals ultimately arrive on time. The resulting burden falls hardest on rural and underserved projects whose social value may be high but whose expected private returns are thin.

Reform should preserve legitimate environmental, historic-preservation, public-safety, and land-management interests while making approvals faster, more predictable, and more proportionate. Firm shot clocks, self-executing deemed-granted remedies, cost-based fee caps, streamlined federal reviews, modernized franchising, broader pole-access protections, and rapid resolution of pole-replacement disputes would directly reduce deployment barriers. OTMR and dig-once policies can lower construction costs by reducing repeated work. Codifying the durable core of these reforms in federal legislation would further reduce the risk that future regulators will reverse them before investors recover their capital.

The jurisdictions that succeed in the next technology race will need more than computing power and electricity. They will need rules that allow the networks connecting those resources to people and businesses to get built.

[1] Kristian Stout, Infrastructure Is Destiny: The Geography of the Next Technology Race (Int’l Ctr. for L. & Econ. May 26, 2026), https://laweconcenter.org/resources/infrastructure-is-destiny-the-geography-of-the-next-technology-race; Jeffrey Westling, The Invisible Infrastructure: Spectrum Policy for the AI Era (Int’l Ctr. for L. & Econ. July 24, 2026), https://laweconcenter.org/resources/the-invisible-infrastructure-spectrum-policy-for-the-ai-era.

[2] See Intelligent Networks: Powering Artificial Intelligence and Transforming Communications: Hearing Before the Subcomm. on Telecomm. & Media of the S. Comm. on Com., Sci., & Transp., 119th Cong. (2026) (statement of Jonathan Spalter, President & CEO, USTelecom—The Broadband Ass’n), https://www.commerce.senate.gov/wp-content/uploads/meetings/379d9950-dc84-b1bf-2c6a-ee1ebb42fbe2/Spalter_Written_Testimony_7eddc348-33d3-4f43-b974-3023b1205597.pdf.

[3] Jeffrey Westling, 2020 Broadband Scorecard Report (R St. Inst. Feb. 10, 2021), https://www.rstreet.org/research/2020-broadband-scorecard-report.

[4] See, e.g., Comments of the International Center for Law & Economics, The State of Competition in the Communications Marketplace, GN Docket No. 26-78, at 11 (2026), https://laweconcenter.org/resources/icle-comments-to-the-fcc-on-the-state-of-competition-in-the-communications-marketplace [hereinafter ICLE Competition Comments]; Stout, supra note 1, at 18; Jeffrey Westling, The Last Mile Is a Paper Trail: Why Broadband Gets Stuck, Truth on the Mkt. (Apr. 22, 2026), https://truthonthemarket.com/2026/04/22/the-last-mile-is-a-paper-trail-why-broadband-gets-stuck.

[5] See Ben Sperry & Kristian Stout, Issue Brief: Pole Attachments and Broadband Build-Out, at 1 (Int’l Ctr. for L. & Econ. 2021), https://laweconcenter.org/resources/issue-brief-pole-attachments-and-broadband-build-out; Kristian Stout & Ben Sperry, Guiding Principles & Legislative Checklist for Broadband Subsidies (Int’l Ctr. for L. & Econ. 2022), https://laweconcenter.org/resources/guiding-principles-legislative-checklist-for-broadband-subsidies; Jeffrey Westling, The Cost of Holding Up Broadband, Truth on the Mkt. (July 14, 2026), https://truthonthemarket.com/2026/07/14/the-cost-of-holding-up-broadband.

[6] Int’l Energy Agency, Key Questions on Energy and AI (2026), https://www.iea.org/reports/key-questions-on-energy-and-ai; Int’l Energy Agency, Energy and AI (2025), https://www.iea.org/reports/energy-and-ai.

[7] Stout, supra note 1, at 13; Westling, The Invisible Infrastructure, supra note 1, at 17; Jeffrey Westling & Kristian Stout, The AI Boom Still Runs on Old-Fashioned Airwaves, Truth on the Mkt. (July 29, 2026), https://truthonthemarket.com/2026/07/29/the-ai-boom-still-runs-on-old-fashioned-airwaves.

[8] ICLE Competition Comments, supra note 4, at 2; Westling, The Last Mile Is a Paper Trail, supra note 4.

[9] See Comments of the International Center for Law & Economics, Implementing the Infrastructure Investment and Jobs Act: Prevention and Elimination of Digital Discrimination, GN Docket No. 22-69, at 6 (May 16, 2022), https://laweconcenter.org/resources/icle-comments-to-the-fcc-on-prevention-and-elimination-of-digital-discrimination [hereinafter Digital Discrimination Comments]; Geoffrey A. Manne, Kristian Stout & Ben Sperry, A Dynamic Analysis of Broadband Competition: What Concentration Numbers Fail to Capture, at 30 (Int’l Ctr. for L. & Econ. 2021), https://laweconcenter.org/resources/a-dynamic-analysis-of-broadband-competition-what-concentration-numbers-fail-to-capture; ICLE Competition Comments, supra note 4, at 8.

[10] ICLE Competition Comments, supra note 4, at 8; see also Westling, The Last Mile Is a Paper Trail, supra note 4.

[11] ICLE Competition Comments, supra note 4, at 8; Joe Kane, Ben Sperry & Kristian Stout, Brief of ICLE and ITIF to the Eighth Circuit in Minnesota Telecom Alliance v. FCC, at 21 (Int’l Ctr. for L. & Econ. 2024), https://laweconcenter.org/resources/brief-of-icle-and-itif-to-8th-circuit-in-minnesota-telecom-alliance-v-fcc.

[12] ICLE Competition Comments, supra note 4, at 8.

[13] Manne et al., supra note 9, at 10.

[14] See Digital Discrimination Comments, supra note 9, at 4–5; Eric Fruits, Ben Sperry & Kristian Stout, The Competitive Effects of the Proposed Charter/Cox Transaction, at 4–6 (Int’l Ctr. for L. & Econ. 2025), https://laweconcenter.org/resources/the-competitive-effects-of-the-proposed-charter-cox-transaction.

[15] ee Jeffrey Westling & Kristian Stout, ICLE Reply Comments to FCC on Eliminating Barriers to Wireless Deployments, at 12 (Int’l Ctr. for L. & Econ. 2026), https://laweconcenter.org/resources/icle-reply-comments-to-fcc-on-eliminating-barriers-to-wireless-deployments; Eric Fruits & Geoffrey A. Manne, Quack Attack: De Facto Rate Regulation in Telecommunications, at 16 (Int’l Ctr. for L. & Econ. 2023), https://laweconcenter.org/resources/quack-attack-de-facto-rate-regulation-in-telecommunications.

[16] See generally ICLE Competition Comments, supra note 4; see also Comments of the International Center for Law & Economics and Scholars of Law & Economics, Protecting the Privacy of Customers of Broadband and Other Telecommunications Services, WC Docket No. 16-106 (May 27, 2016), https://laweconcenter.org/resources/comments-icle-scholars-law-economics-matter-protecting-privacy-customers-broadband-telecommunications-services.

[17] ICLE Competition Comments, supra note 4, at 10; Westling, The Last Mile Is a Paper Trail, supra note 4.

[18] Westling, The Last Mile Is a Paper Trail, supra note 4.

[19] Chandler Dempsey et al., The Economic Impact of the Rural Digital Opportunity Fund, at 61 (Fla. State Univ. Dep’t of Econ. 2021), https://cosspp.fsu.edu/economics/wp-content/uploads/sites/10/2021/10/TheEconomicImpactoftheRuralDigitalOpportunityFund.pdf.

[20] Eric Fruits et al., Dynamic Competition in Broadband Markets: A 2024 Update, at 42 (Int’l Ctr. for L. & Econ. 2024), https://laweconcenter.org/wp-content/uploads/2024/06/Broadband-Competition-2024-Update.pdf.

[21] R.H. Coase, The Problem of Social Cost, 3 J.L. & Econ. 1 (1960).

[22] Westling, The Last Mile Is a Paper Trail, supra note 4.

[23] Id.

[24] Id.; see also Drew Clark, Railroad Crossing Killing Rural Broadband Deployment, Broadband Breakfast (Nov. 18, 2025), https://broadbandbreakfast.com/railroad-crossing-killing-rural-broadband-deployment-says-virginia-telecom-official.

[25] Ben Sperry et al., ICLE and CEI Amicus to the 6th Circuit on Safeguarding and Securing the Open Internet Order, at 4 (Int’l Ctr. for L. & Econ. 2024), https://laweconcenter.org/resources/icle-and-cei-amicus-to-the-6th-circuit-on-safeguarding-and-securing-the-open-internet-order.

[26] ICLE Competition Comments, supra note 4, at 8.

[27] Inquiry Concerning Deployment of Advanced Telecommunications Capability to All Americans in a Reasonable and Timely Fashion, 2024 Section 706 Report, GN Docket No. 22-270, FCC 24-27 (rel. Mar. 18, 2024), https://docs.fcc.gov/public/attachments/DOC-401205A1.pdf.

[28] Lazar Radic, Geoffrey A. Manne & Dirk Auer, Regulate for What? A Closer Look at the Rationale and Goals of Digital Competition Regulations, 22 Berkeley Bus. L.J. 73 (2025), https://laweconcenter.org/resources/regulate-for-what-a-closer-look-at-the-rationale-and-goals-of-digital-competition-regulations.

[29] Stout, supra note 1, at 18–19.

[30] Id.

[31] Robert D. Cooter & Michael D. Gilbert, Public Law and Economics 42 (Oxford Univ. Press 2022), https://laweconcenter.org/resources/public-law-and-economics (illustrating the unanimity holdout problem through a road crossing multiple parcels, where any owner can block the entire project); see also Comments of the International Center for Law & Economics, Implementation of Section 621(a)(1) of the Cable Communications Policy Act of 1984, as Amended by the Cable Television Consumer Protection and Competition Act of 1992, MB Docket No. 05-311, at 5 (July 18, 2019), https://laweconcenter.org/resources/icle-comments-on-implementation-of-section-621a1-of-the-cable-communications-policy-act-of-1984 (applying Klein-style hold-up theory to show how gatekeepers can extract rents after providers make relationship-specific sunk investments and explaining why the 1984 Act capped franchise fees).

[32] 47 U.S.C. § 332(c)(7)(B)(ii) (requiring state and local governments to act on siting requests “within a reasonable period of time”); id. § 332(c)(7)(B)(iii) (requiring written denials “supported by substantial evidence contained in a written record”).

[33] See Kristen Hite, Cong. Rsch. Serv., IF11932, National Environmental Policy Act: Judicial Review and Remedies (2024), https://www.congress.gov/crs-product/IF11932.

[34] Nikki Chiappa et al., Understanding NEPA Litigation (Breakthrough Inst. 2025), https://thebreakthrough.org/issues/energy/understanding-nepa-litigation.

[35] Seven County Infrastructure Coalition v. Eagle County, 605 U.S. 168 (2025); Kristian Stout, Testimony Before the House Judiciary Subcommittee on the Administrative State, Regulatory Reform, and Antitrust: The 30,000-Foot View: Competition and Regulation in the U.S. Airline Industry 12 (June 24, 2026), https://laweconcenter.org/resources/testimony-of-kristian-stout-to-the-house-judiciary-antitrust-subcommittee-on-competition-in-the-airline-industry.

[36] National Environmental Policy Act of 1969 § 102(2)(C), 42 U.S.C. § 4332(2)(C) (requiring federal agencies to prepare detailed environmental-impact statements for recommendations or reports on “major Federal actions significantly affecting the quality of the human environment”).

[37] Westling, The Last Mile Is a Paper Trail, supra note 4.

[38] Stout, The 30,000-Foot View, supra note 35, at 12.

[39] Seven County, 605 U.S. at 168.

[40] National Historic Preservation Act § 106, 54 U.S.C. § 306108.

[41] Westling, The Last Mile Is a Paper Trail, supra note 4; Geoffrey A. Manne, Gus Hurwitz & Kristian Stout, Putting Politics over Policy at the FCC, Truth on the Mkt. (Oct. 4, 2018), https://truthonthemarket.com/2018/10/04/putting-politics-over-policy-at-the-fcc.

[42] 54 U.S.C. § 302706(b) (requiring agencies to consult with Indian tribes that attach religious and cultural significance to affected historic properties).

[43] Statement of Commissioner Michael O’Rielly, Accelerating Wireless Broadband Deployment by Removing Barriers to Infrastructure Investment, Second Report and Order, WT Docket No. 17-79, FCC 18-30 (2018), https://docs.fcc.gov/public/attachments/FCC-18-30A4.pdf.

[44] Clean Water Act § 404, 33 U.S.C. § 1344 (authorizing Army Corps of Engineers permits for discharges of dredged or fill material into “waters of the United States,” including wetlands); 33 C.F.R. pts. 320–332; see U.S. Env’t Prot. Agency, Section 404 of the Clean Water Act, https://www.epa.gov/cwa-404; Rivers and Harbors Act of 1899 § 10, 33 U.S.C. § 403; U.S. Army Corps of Eng’rs, Section 10/Section 404 Permitting Reference Guide, https://www.spk.usace.army.mil/Portals/12/documents/regulatory/Reg_workshop/2015-11-06/RPW-Sec10-404.pdf.

[45] Clean Water Act § 404(e), 33 U.S.C. § 1344(e) (authorizing the Corps’ general and nationwide permit program); 33 C.F.R. pt. 330 (implementing the program).

[46] Endangered Species Act § 7(a)(2), 16 U.S.C. § 1536(a)(2) (requiring federal agencies, in consultation with the secretary, to ensure that their actions neither jeopardize listed species nor destroy or adversely modify designated critical habitat); 50 C.F.R. pt. 402 (implementing Section 7); U.S. Fish & Wildlife Serv., Section 7 Consultation, https://www.fws.gov/services/consultation-and-technical-assistance/esa-section-7-consultation (explaining that the Fish and Wildlife Service handles terrestrial and freshwater species, while NOAA’s National Marine Fisheries Service handles most marine and anadromous species).

[47] Federal Land Policy and Management Act of 1976 tit. V, 43 U.S.C. §§ 1761–1771 (authorizing the Bureau of Land Management to grant rights-of-way over public lands for communications sites and facilities, among other uses).

[48] Update of the Communications Uses Program, Cost Recovery Fee Schedules, and Section 512 of FLPMA for Rights-of-Way, 89 Fed. Reg. 25,916 (Apr. 12, 2024) (effective May 13, 2024), https://www.federalregister.gov/documents/2024/04/12/2024-06997/update-of-the-communications-uses-program-cost-recovery-fee-schedules-and-section-512-of-flpma-for.

[49] 23 C.F.R. § 645.211 (requiring each state transportation department to adopt an approved utility-accommodation policy governing utilities’ use and occupancy of highway rights-of-way); Fed. Highway Admin., Guidance on Longitudinal Telecommunications Installations on Limited Access Highway Right-of-Way, https://www.fhwa.dot.gov/utilities/001222.cfm (explaining that states allow fiber-optic installation along freeway rights-of-way under utility-accommodation policies, subject to reasonable permitting conditions governing time, place, and manner).

[50] 47 U.S.C. § 254(e) (limiting federal universal-service support to eligible telecommunications carriers designated under § 214(e) and requiring recipients to use that support only for its intended facilities and services); id. § 214(e) (authorizing state commissions—and, in specified circumstances, the FCC—to designate eligible telecommunications carriers).

[51] See Mass. Exec. Off. of Energy & Env’t Affs., MEPA Regulatory Review (2021), https://www.mass.gov/doc/mepa-regulatory-review-presentation/download (reporting that “[a]bout 16 states have ‘little NEPA’ programs”).

[52] Cal. Pub. Res. Code §§ 21000–21189.70 (2026).

[53] See Dana A. Scherer, Cong. Rsch. Serv., R46077, Potential Effect of FCC Rules on State and Local Video Franchising Authorities (2020), https://www.congress.gov/crs-product/R46077.

[54] Cable Communications Policy Act of 1984, Pub. L. No. 98-549, 98 Stat. 2779 (codified as amended at 47 U.S.C. §§ 521–573).

[55] Implementation of Section 621(a)(1) of the Cable Communications Policy Act of 1984, as Amended by the Cable Television Consumer Protection and Competition Act of 1992, Third Report and Order, MB Docket No. 05-311, FCC 19-80 (2019), https://docs.fcc.gov/public/attachments/FCC-19-80A1.pdf.

[56] See American Broadband Deployment Act of 2025, H.R. 2289, 119th Cong. (2025), https://www.congress.gov/bill/119th-congress/house-bill/2289.

[57] Title VI requires cable operators to obtain franchises before providing cable service. 47 U.S.C. § 541(b)(1). Noncable telecommunications and broadband providers instead access local rights-of-way under Section 253, which preserves local authority only to “manage the public rights-of-way” and require “fair and reasonable compensation.” Id. § 253(c). Some local codes therefore distinguish cable-television franchises from the permits and use authorizations that telecommunications providers obtain for rights-of-way access. See, e.g., Wash. Rev. Code § 35.99.040.

[58] 47 U.S.C. § 253(a), (c).

[59] 47 U.S.C. § 332(c)(7)(A), (B)(i)(II).

[60] Fed. Highway Admin., U.S. Dep’t of Transp., Pavement Utility Cuts: Controlling Pavement Cuts by Implementing Policy (2018), https://www.fhwa.dot.gov/utilities/utilitycuts/man03.cfm.

[61] Accelerating Wireless Broadband Deployment by Removing Barriers to Infrastructure Investment, Declaratory Ruling and Third Report and Order, WC Docket No. 17-84, ¶¶ 149–150, FCC 18-111 (2018), https://docs.fcc.gov/public/attachments/FCC-18-111A1.pdf.

[62] Petition for Declaratory Ruling to Clarify Provisions of Section 332(c)(7)(B) to Ensure Timely Siting Review and to Preempt Under Section 253 State and Local Ordinances That Classify All Wireless Siting Proposals as Requiring a Variance, Declaratory Ruling, WT Docket No. 08-165, ¶ 32, FCC 09-99 (2009), https://docs.fcc.gov/public/attachments/FCC-09-99A1.pdf.

[63] Accelerating Wireline Broadband Deployment by Removing Barriers to Infrastructure Investment et al., Declaratory Ruling and Third Report and Order, WT Docket No. 17-79 & WC Docket No. 17-84, ¶¶ 40–46, 104–110, FCC 18-133 (2018), https://docs.fcc.gov/public/attachments/FCC-18-133A1.pdf.

[64] 47 U.S.C. § 1455(a) (providing that state and local governments “may not deny, and shall approve” eligible requests to modify existing wireless towers or base stations when the modification does not substantially change their physical dimensions); Acceleration of Broadband Deployment by Improving Wireless Facilities Siting Policies et al., Report and Order, WT Docket No. 13-238 et al., ¶¶ 205–215, FCC 14-153 (2014), https://docs.fcc.gov/public/attachments/FCC-14-153A1.pdf (adopting a 60-day shot clock and a self-executing “deemed granted” remedy when a locality fails to act within 60 days after receiving the applicant’s notice).

[65] Accelerating Wireline Broadband Deployment, supra note 63, ¶¶ 84–90 (governing how incompleteness determinations affect the shot clocks).

[66] 47 U.S.C. § 224(b)(1).

[67] 47 U.S.C. § 224(a)(1).

[68] 47 C.F.R. § 1.1402(o).

[69] Westling, The Last Mile Is a Paper Trail, supra note 4.

[70] Id.

[71] Accelerating Wireline Broadband Deployment by Removing Barriers to Infrastructure Investment, Declaratory Ruling, WC Docket No. 17-84, ¶¶ 8–22, DA 21-78 (WCB 2021), https://docs.fcc.gov/public/attachments/DA-21-78A1.pdf (finding that utilities often charged new attachers the full cost of replacing poles already near the end of their useful lives or otherwise due for replacement and clarifying that new attachers need not pay to correct preexisting violations or replace “red-tagged” poles that utilities must replace).

[72] Id. ¶¶ 8–26.

[73] Accelerating Wireline Broadband Deployment, supra note 61, ¶¶ 33–86.

[74] For recent enforcement, see Fed. Commc’ns Comm’n, FCC Issues Precedent-Setting Order Under Expedited Review Process for Pole Attachment Complaints (Feb. 5, 2026), https://docs.fcc.gov/public/attachments/DOC-418491A1.pdf.

[75] Brendan Carr, Chairman, Fed. Commc’ns Comm’n, A Build Agenda for America (July 2, 2025), https://docs.fcc.gov/public/attachments/DOC-412663A1.pdf; see also Fed. Commc’ns Comm’n, Chairman Carr’s Build America Agenda, https://www.fcc.gov/build-america.

[76] 40 C.F.R. § 1501.4; 42 U.S.C. § 4336(a)(2).

[77] 36 C.F.R. § 800.14(b); see also 54 U.S.C. § 306108.

[78] Seven County, 605 U.S. at 168.

[79] Sackett v. EPA, 598 U.S. 651 (2023) (holding that the Clean Water Act covers only wetlands with a continuous surface connection to bodies that constitute “waters of the United States” in their own right, making the wetlands indistinguishable from those waters).

[80] United Keetoowah Band of Cherokee Indians v. FCC, 933 F.3d 728 (D.C. Cir. 2019) (vacating the FCC’s 2018 order insofar as it excluded small wireless-facility deployments from NEPA and National Historic Preservation Act Section 106 review because the FCC failed to justify that deregulation served the public interest or adequately address potential effects on historic and tribal resources).

[81] American Broadband Deployment Act of 2025, supra note 56; BROADBAND Leadership Act, H.R. 278, 119th Cong. (2025), https://www.congress.gov/bill/119th-congress/house-bill/278/text.

[82] Build America: Eliminating Barriers to Wireless Deployments, Notice of Proposed Rulemaking, WT Docket No. 25-276 (Sept. 2025), https://docs.fcc.gov/public/attachments/DOC-414414A1.pdf.

[83] American Broadband Deployment Act of 2025, supra note 56.

[84] Petition of Gateway Infrastructure, LLC d/b/a Gateway Fiber for Preemption and Declaratory Ruling Pursuant to Section 253(d) of the Communications Act of 1934, WC Docket No. 26-166 (filed June 17, 2025), https://www.fcc.gov/ecfs/search/search-filings/filing/26109834869 [hereinafter Gateway Fiber Petition]; see Wireline Competition Bureau Seeks Comment on Petitions for Preemption and Declaratory Ruling, Public Notice, DA 26-689 (WCB July 8, 2026), https://docs.fcc.gov/public/attachments/DA-26-689A1.pdf.

[85] Gateway Fiber Petition, supra note 84; see 47 U.S.C. § 253(a), (d).

[86] Wireline Competition Bureau Seeks Comment on Petitions for Preemption and Declaratory Ruling, supra note 84.

[87] Jeffrey Westling, Fiber Freeze: How Maple Grove Made a Cable Franchise the Price of Broadband, Truth on the Mkt. (Aug. 20, 2026), https://truthonthemarket.com/2026/08/20/fiber-freeze-how-maple-grove-made-a-cable-franchise-the-price-of-broadband.

[88] Accelerating Wireline Broadband Deployment by Removing Barriers to Infrastructure Investment, Fifth Report and Order, Fourth Further Notice of Proposed Rulemaking, and Orders on Reconsideration, WC Docket No. 17-84, FCC 25-38 (July 24, 2025), https://docs.fcc.gov/public/attachments/FCC-25-38A1.pdf.

[89] Ben Sperry, Geoffrey A. Manne & Kristian Stout, The Role of Antitrust and Pole-Attachment Oversight in TVA Broadband Deployment, 29 J. Tech. L. & Pol’y 73 (2025), https://scholarship.law.ufl.edu/cgi/viewcontent.cgi?article=1243&context=jtlp.

[90] Ben Sperry, Broadband Deployment, Pole Attachments, & the Competition Economics of Rural-Electric Co-ops, Truth on the Mkt. (Aug. 16, 2023), https://truthonthemarket.com/2023/08/16/broadband-deployment-pole-attachments-the-competition-economics-of-rural-electric-co-ops.

[91] Accelerating Wireline Broadband Deployment by Removing Barriers to Infrastructure Investment, Fourth Report and Order, Declaratory Ruling, and Third Further Notice of Proposed Rulemaking, WC Docket No. 17-84, FCC 23-109 (rel. Dec. 15, 2023), https://docs.fcc.gov/public/attachments/FCC-23-109A1.pdf; see also Westling, The Cost of Holding Up Broadband, supra note 5.

[92] Westling, The Cost of Holding Up Broadband, supra note 5.

[93] Broadband Infrastructure Deployment, 86 Fed. Reg. 68,553 (Dec. 3, 2021), https://www.federalregister.gov/documents/2021/12/03/2021-26231/broadband-infrastructure-deployment; see also 23 C.F.R. § 645.307(a); MOBILE NOW Act § 607, Pub. L. No. 115-141, div. P, tit. VI, 132 Stat. 1097, 1104 (2018); U.S. Gov’t Accountability Off., GAO-26-107734, Broadband Infrastructure: States’ Reported Progress Implementing Requirements to Facilitate Deployment Along Federal-Aid Highways (2026), https://www.gao.gov/products/gao-26-107734.