I’ve been continuing to follow the federal plans to launch a massive $1 trillion program to rebuild infrastructure, with an eye on possible funding for broadband.
The latest White House proposal includes $25 billion for broadband, spread over ten years. That’s obviously not enough to solve our broadband problem everywhere, but it certainly could put a big dent in it.
The first thing needed to understand the issue is a good estimate of the size of needed investment to build fiber. Back in 2013 an article in Forbes estimated the cost to build fiber everywhere at $140 billion, with the cost to get to most populated areas at half that. Just recently I saw a few news articles citing an estimated cost of $85 billion to bring ‘broadband’ to rural America. But I can’t track down who’s making that estimate or if ‘broadband’ means fiber everywhere or some mix of broadband technologies. But it’s obvious that the cost is going to be greater than $25 billion.
But there is another aspect of the White House proposal that we need to keep an eye on. They envision the government kicking in $200 billion with the rest of the $1 trillion coming from the private sector. Even if the entire $200 billion is in the form of outright grants, that means that a project funded under the federal plan would be getting a 20% grant and would have to somehow finance the other 80% of a project.
I’ve created a lot of rural business plans over the past few years and I can tell that a 20% grant is not going to work in financing rural fiber. That amount of grant might be sufficient when talking about rural county seats and other pockets of somewhat dense population. But all of the studies I’ve done show that it will require grants of 40% to 80% to finance building in rural America.
I also worry that part of the federal funding might also include loans, like was done a few years ago with the Stimulus broadband awards. Some of those projects got a mix of outright grants as well as long-term loans from the RUS. If the federal contribution is not all grants then its usefulness in rural America will be even more diminished.
This is not to say that the federal program might not offer different levels of grants and not stick to the overall 20% for everything. But if 20% grants are all that is offered there are not going to be many takers. A grant of that magnitude probably might bring fiber to suburbs and mid-sized towns, but not to rural America.
If the grants were set to 50% of the cost of a project it would stimulate a lot of rural fiber construction. We’ve seen this in action in programs like the DEED grants in Minnesota that have been funding $20M to $30M per year in the form of matching grants. In that state the various LECs, from the smallest up through CenturyLink, are using grant money to bring broadband to unserved rural customers. But Minnesota is unusual and is one of a handful of states where there are numerous telcos willing to branch out to serve the areas around them if these kinds of grants are available.
One of the biggest hurdles I see for building rural broadband is the availability of private capital. Even with a 50% grant the operators of these new networks will need to finance the other 50% of the projects. There is certainly cash available for this and having a federal infrastructure program might attract more lenders. But there is a natural lending limit on all telcos, big and small.
A large percentage of the smaller telcos I know have already been borrowing money to build fiber within their own operating areas, and those companies are not going to be able to borrow much more money even if it is available. Even the big companies have constraints. CenturyLink currently has an annual capital budget of about $3 billion per year and that is largely going towards building fiber in their urban markets. It’s hard to see them taking any real interest in building rural fiber if they have to borrow to do so. Many of the mid-sized telcos like Frontier and Windstream are already heavily leveraged and would have a hard time borrowing much. And Verizon and AT&T have made it very clear that they no longer want to be in the rural wire business. I’m not sure those companies would take on these networks if the federal government paid for all of it.
So having a federal broadband infrastructure program sounds great. But when you look a little closer at how it might work it starts to look troublesome. There are certainly a number of companies that would step up to build rural broadband if the grants are large enough to make the numbers work. But I’m not sure that there is any combination of companies that are able or willing to tackle all of the areas without broadband. It could end up being a program where there is more funding than takers.
We work with Rural Service providers every day. I agree that it will take more than most companies could, or are willing to, borrow to build out many of the remaining areas with FTTH or even wireless, especially 5G. When it costs tens of thousands to build to one customer, it is hard to make the business case even with a 100% grant. The on-going pole attachment costs, along with maintenance costs, will not be paid for by monthly recurring revenues in many cases.
Many telephone companies upgraded their networks to FTTH using borrowed funds with the “promise” of continuing USF funds. The reductions in existing funding in order to provide additional funding for unserved and underserved locations has reduced the ability of those companies to borrow funds for further expansion.
With 5G you would need FTTH to every house in rural Montana just to feed the 5G sites. 4G would work in some cases, but you cannot ever afford to build the same network in an area where the network is measured in miles per subscriber instead of subscribers per mile.
CAF II funding will require that you build to a location where you can serve the location within 10 days. If the person lives just across a railroad tract, you will not get a permit in 10 days. So, do you build across the tracts just so you can get funding for that site, even though they may never take service? How about when you need to do pole attachments to reach the location? Do you build thousands (or more) of feet of strand and drop to get to the transformer pole for a single location, just to get funding? If you do not build to get funding, what happens when they do apply for service and it will cost thousands to build the “drop”?
I am very concerned that in trying to serve 100% of the locations we will miss opportunities to serve the majority of people without broadband. I see comments that we need to be able to provide the exact same level of service in rural areas or nothing at all. Based on that logic, we would need to build 4-lane highways to every house in the U.S. since that is what is available in some urban areas.
We need a lot more discussion, but ultimately it will take a lot of funding to reach everyone with high-speed internet access.