I work with a lot of ISPs that own rural fiber. Some rural network owners have been successful in providing fiber to cell sites near their networks over the last decade. A few sell directly to a cellular carrier, but most of these connections are sold to an intermediate carrier that bundles together cellular connections across a large geographic area.

This has been good business. Depending upon the bandwidth being purchased for the cell site, the monthly fee for providing bandwidth to a cell tower has ranged from $500 to several thousand dollars per month. That’s good revenue for a rural ISP that mostly serves residential customers and that doesn’t typically have a lot of higher-margin business customers.

I’m hearing about requests from cellular companies or intermediate carriers to increase bandwidth at cell sites. I’ve had four separate discussions about this from different parts of the country in the last month. This makes a lot of sense. All of the statistics say that bandwidth usage on cell phones is growing at an even faster pace than bandwidth usage for households.

There is another phenomenon driving the recent need for more bandwidth at cell sites. T-Mobile and Verizon have added almost 7 million FWA cellular broadband customers in just the last few years and have gained over a 6% market share of all national broadband connections. AT&T and smaller cellular carriers are also launching the FWA product.

FWA broadband is a major upgrade from the cellular hotspots that have been sold in rural America over the last decade. Hotspots hit the news everywhere when school systems and employers tried to use hotspots to provide home broadband during the pandemic, only to find that the bandwidth available was not adequate to meet the demand. FWA cellular is using the new spectrum that the cellular carriers have labeled as 5G, and this has meant a significant increase in speeds—with T-Mobile and Verizon providing speeds between 100 Mbps and 300 Mbps within a few miles of an upgraded cell tower.

These new requests for higher bandwidth create a few issues for local ISPs. Some local networks will have to upgrade electronics to be able to deliver a 10-gigabit connection to a cell site. At the same time, there has been a continuous downward pressure on backhaul bandwidth pricing, and it’s likely that the cellular carriers are going to want multi-gigabit connections at a similar, or even lower cost than what they might have been paying for a gigabit or less today.

The real dilemma is that an ISP that increases bandwidth for an FWA site is enabling a last-mile competitor. ISPs have not worried about competition from hotspots because the speeds are relatively slow, and hotspots have tiny data caps that make them too costly for a household to use in any meaningful way. But FWA cellular is both cheap (in the range of $60 per month) and provides unlimited usage. There is always a significant percentage of customers in any market for whom price is more important than bandwidth—there is always a market for low prices.

If an ISP provides increased bandwidth to a cell tower, the carrier at that tower might use that bandwidth to capture 10% or more of the customers around the tower. It was one thing to support cell towers when they were used for rural cellphone coverage. But it’s a new equation to be asked to provide faster bandwidth to an ISP that will use the bandwidth to win over local customers.

I’ve been having some interesting conversations with ISPs about this new dilemma. Is the extra revenue from selling bigger bandwidth to cell sites high enough to replace what might be lost from last-mile retail sales? ISPs have never been required to enable competitors to compete against them unless their network was partially funded by a grant. Last-mile ISPs have never thought of cellular carriers as competitors, but suddenly they are.

Part of the answer to the question comes from looking at the likelihood that somebody else will bring the backhaul if an ISP declines to increase the bandwidth. At some cell sites, it might be worthwhile for somebody else to build new fiber or microwave routes to supplement the cell site bandwidth. We are also possibly going to see satellites able to fulfill rural backhaul. The business plan for OneWeb has changed from retail broadband to wholesale backhaul. There are several other satellite companies designing satellite constellations aimed directly at providing cell site backhaul.

I don’t have any easy answer for this, and every local situation is a little bit different. Perhaps FWA competition is inevitable, and maximizing sales to cell sites is a way to offset the coming revenue losses. Most rural fiber network owners have not had to worry much about competition. Every network owner, including those providing fiber broadband, has always seen some customers choose other ISPs. But rural ISPs have never feared losing significant customers. Should a network owner embrace the increased bandwidth sale to towers, or should they drag their feet for a while to push off the time for competition? At what point Do they risk losing backhaul sales to cell sites—since the worst result would be losing both the wholesale and retail revenues?

One of the old truisms of rural broadband is there is almost no real competition once somebody builds a fast rural network. But the advent of FWA wireless and faster fixed wireless radios means that rural markets will have competition, at least in some portion of a network.