Towns that built it themselves
Ammon Owns the Strand and Lets Providers Compete on It
The Idaho city built the conduit, lit the fibre, and then got out of the retail business — leaving competition to do what private markets would not.

Ammon separates the plant from the service: the city owns the strand, the provider is a choice.
Photo: Maui Media / Pexels
One Network, Several Providers
Ammon, Idaho is a small city of roughly 17,000 people on the eastern Snake River Plain, adjacent to Idaho Falls. It is not where a broadband experiment would be expected to begin. But since 2016, Ammon has operated a municipal fibre network structured on a principle that distinguishes it from nearly every other publicly built system in the country: the city owns the physical infrastructure, and it does not sell internet service.
The model works in layers. The city of Ammon owns the conduit buried under streets and the fibre strands pulled through it. That physical plant is available to any qualifying internet service provider that wants to reach customers inside Ammon. Providers lease capacity on the network and compete directly against one another for subscribers — on price, service quality, or both. Residents pick a provider the way they might choose a utility plan, with the underlying infrastructure held in common beneath them all.

Aerial strand is cheaper than trenching, and the pole it hangs from usually belongs to somebody else.
Photo: Дмитрий Рощупкин / Pexels

Selling retail service means staffing a room like this one continuously, whoever owns the strand.
Photo: Fernando Narvaez / Pexels
This is the open-access network model in its clearest municipal form, and Ammon is among the most cited examples of it operating at the city level in the United States.
What Makes It Different
The distinction matters because most municipal broadband networks are vertically integrated: the city builds the fibre, operates it, and sells service directly to residents. Chattanooga's EPB and Wilson's Greenlight both follow that structure. Those networks have delivered genuine results — competitive speeds, lower prices, improved service reliability — but they require the public entity to function as a retail internet provider in perpetuity, absorbing operational overhead, customer service costs, and the commercial risk of subscriber churn.
Ammon's design separates infrastructure ownership from retail service entirely. The city finances and maintains the physical plant. Providers handle customer acquisition, billing, and the service relationship. The city collects lease revenue from those providers rather than subscription fees from households. If a provider performs poorly, residents can switch to another on the same fibre — without the city renegotiating a franchise or the resident waiting for a second conduit to be dug.
That structural portability is the argument for open access. A vertically integrated municipal network can undercut a private incumbent on price but still leaves residents with a single provider. Ammon's model means the discipline of competition operates on publicly owned infrastructure rather than being foreclosed by it.
Ammon finances residential connections through a self-funding mechanism: property owners who opt in pay an assessment to connect, which covers the capital cost of bringing fibre to their premises. The city does not subsidise connections from general revenue, and participation is voluntary. Once connected, a household selects from whichever providers are operating on the network. That pay-to-connect structure keeps the programme financially self-contained, though it also means households that cannot afford the upfront cost of connection face a barrier that a fully subsidised build would not create.
The Incumbent Problem It Sidesteps
Where a private cable or telephone company already operates, a municipality building a competing vertically integrated network faces the full weight of an incumbent's political opposition, predatory pricing, and — in roughly twenty states — a preemption statute that may block the project before ground is broken. Idaho has no such statute, which gave Ammon room to act. But the open-access model also presents a softer target politically, because the city is not entering the retail market as a competitor. It is building infrastructure and inviting providers — potentially including private ones — to use it.
Whether private incumbents participate in open-access networks in practice is another question. In Ammon, the providers operating on the network have generally been smaller ISPs rather than the regional cable or telephone carriers that dominate most markets. The incumbent's incentive to join a system that introduces direct competition to their existing customers is limited, and market structure appears to remain a constraint on open-access adoption even where the legal environment is permissive.
Ammon's model does not solve every problem — it reaches households that opt in and can pay the connection assessment, not universally — but it demonstrates that a city can own the infrastructure layer of a broadband market without also running the service layer, and that competition on publicly built fibre is operationally feasible at the municipal scale.