The Infrastructure Problem That Doesn’t Look Like One
High-speed internet access is infrastructure as fundamental to economic participation in 2026 as road access was in the 20th century — it connects people to employment opportunities, healthcare information, educational resources, government services, and financial tools that are increasingly available only online. The digital divide between those with reliable broadband access and those without has direct, documented effects on economic outcomes, educational attainment, and access to healthcare services.
Despite significant investment and decades of policy attention, the broadband gap persists. The US Federal Communications Commission’s own maps have been repeatedly criticised for overstating coverage, and the communities most affected — rural areas, low-income urban neighbourhoods, and tribal lands — are underserved not because the problem isn’t understood but because the economic and political structure of broadband infrastructure creates incentives that don’t align with universal service.
Why the Market Doesn’t Solve This on Its Own
Broadband infrastructure is expensive to deploy and disproportionately expensive per household to deploy in low-density areas. A cable or fibre provider deploying infrastructure in a dense urban area amortises the infrastructure cost over many customers within a small geographic area; the same deployment in a rural area serves far fewer households per mile of cable laid. The economics of private broadband deployment produce profitable service in dense areas and insufficient service in sparse ones.
The incumbent provider problem compounds this: in most US markets, a single cable company has a practical monopoly (or duopoly with a local phone company) on wired broadband. Without competition, prices remain high and service expansion into underserved areas happens only when subsidised. The same ISP that pleads insufficient economics to justify rural deployment in the absence of subsidy is often the company lobbying against competitive broadband policies that might bring competition into their profitable urban markets.
Federal Investment and Its Limitations
The Infrastructure Investment and Jobs Act (2021) allocated $65 billion for broadband expansion, including $42.5 billion in BEAD grants to states for high-speed broadband deployment in underserved areas. This is the largest federal broadband investment in US history. The implementation has been slower than the original timeline anticipated: state BEAD plans required federal approval before deployment could begin, and the complex requirement structure (preference for fibre over fixed wireless, requirements for affordable pricing plans) created implementation delays that pushed actual deployment years after initial expectations.
International comparisons are instructive: South Korea, Japan, and several European countries have achieved near-universal high-speed internet access through combinations of strong regulatory frameworks, meaningful competition policy, and public investment that the US has historically been reluctant to implement. The policy choices are clear; the political economy of broadband — the lobbying power of incumbent providers who benefit from the status quo — is the primary obstacle to implementing them.
Starlink and Fixed Wireless: The Alternative Path
Starlink and similar low-earth-orbit satellite broadband services (Amazon’s Kuiper, OneWeb) have changed the rural broadband picture by providing genuine broadband-class speeds in locations where terrestrial infrastructure deployment is economically difficult. The coverage model is different from terrestrial broadband: satellites cover geographic areas without needing physical infrastructure at every served location, making the economics of rural coverage more tractable.
The limitations of satellite broadband as the universal solution: the per-user equipment cost ($599 for Starlink hardware as of mid-2025) and monthly service cost ($120/month) are prohibitive for the lowest-income users who are most underserved. Network congestion as subscriber counts grow in specific areas has caused performance degradation in some markets. And the technology is owned by a small number of private companies rather than being a regulated public utility — creating long-term dependency on private entities whose service terms may change.
What Progress Looks Like and What Remains
The broadband access picture in 2026 is meaningfully better than in 2016: more households have broadband access, speeds have increased across tiers, and low-earth-orbit satellite service has extended coverage to areas that previously had no viable options. The BEAD programme’s eventual full implementation will further extend fibre access to rural communities. The Affordable Connectivity Programme (which subsidised broadband access for low-income households before Congressional funding lapsed in 2024) demonstrated that take-up of affordable broadband by low-income households is high when the cost barrier is removed.
What remains: the urban digital divide (affordable access in low-income urban areas, where infrastructure exists but cost is prohibitive), the persistent rural coverage gap in the most sparsely populated areas that satellite serves with cost limitations, and the ‘digital divide within digital divide’ of households with access but without devices or digital literacy skills to benefit from access. Solving broadband access is a prerequisite, not a complete solution, for digital equity.
