To finalize our series on reflections from Seattle and Gigabit Squared, I discuss open access networks and how the requirement that a network directly pay all its costs effectively dooms it in the U.S. Read part one here and part two here. I started this series because I felt that the Gigabit Squared failure in Seattle revealed some important truths that can be glossed over in our rush to expand access to fast, affordable, and reliable Internet connections.
The benefits of public-private-partnerships in these networks have often been overstated while the risks and challenges have been understated. We have seen them work and believe communities should continue to seek them where appropriate, but they should not be rushed into because they are less controversial than other solutions.
Sometimes we have to stop and remember that we will live for decades with the choices we make now. It was true when communities starting building their own electrical networks and is still true today. I hope the series has provided some context of how challenging it can be without removing all hope that we can stop Comcast, AT&T, and others from monopolizing our access to the Internet.
In this final piece, I want to turn to a different form of partnership - the open access network. I think it follows naturally as many in Seattle and other large cities would be more likely to invest in publicly owned fiber networks if they did not have to offer services - that being the most competitive, entreprenuerial, and difficult aspect of modern fiber networks.
The desire to focus on long term investments rather than rapidly evolving services is a natural reaction given the historic role of local governments in long term infrastructure investments. Fiber certainly fits in that description and as many have noted, the comparison to roads is apt. An open access fiber network allows many businesses to reach end users just as roads allow Fedex, UPS, and even the Post Office, to compete on a level playing field.
In an open access approach, the local government would build the network out to connect all residents and businesses but not directly deliver services. Instead, multiple independent Internet Service Providers (ISPs) would compete on the network for business, ideally specializing in different niches - some providing great video game optimizations and others focusing on meeting small business needs.
Unfortunately, there are reasons we have not seen this approach gain widespread traction. The model is more difficult than is readily apparent.
A large part of the difficulty comes from incumbent providers that refuse to use the fiber network. The cable and the telephone companies claim that they don't want to abandon their assets, but that is not the main reason they have refused to participate in these networks. The big cable and telephone companies know that they have terrible reputations and would be slaughtered in a competitive market - so they put great effort in ensuring that they face as little competition as possible. Allowing the open access market to develop would all but ensure mighty Comcast would have to compete against local providers that offer much better customer service, lower prices, and more.
From an economic perspective, an ideal open access network would be one physical fiber network on which all ISPs compete. With a take rate over 80 percent, the revenue would likely be sufficient to pay the costs of building the network, operating costs, AND the ISP costs. But because the cable and telephone companies have fought against open access, subscribers are often split among three different physical infrastructures (cable, copper telephone lines, and the fiber network), generating too little revenue to pay the costs of building the fiber network.
If a major metro area does a feasibility study to build a citywide open access network fiber, it will find that the network will almost certainly not pay for itself using a conventional private sector accounting system. The interest on the debt required to build the network accrues faster than revenue. Of course, the roads and bridges don't pay for themselves via user fees either, but we still invest in them.
In a recent podcast, we discuss how over the first five years, a network can save more in aggregate for the community that it costs to build. But those benefits acrue individually to households. Thus far, very few communities have used this approach - to raise monthly taxes by $3 to save $10 on household telecom bills, for instance. Leverett is a rare example of this approach.
That does leave another option - building an open access network incrementally, as Danville has done in Virginia and Palm Coast FiberNet in Florida, among others. This is a viable option for just about any community but comes with the difficult reality that connecting everyone could take decades. And there are still other gotchas.
Some communities that wanted to build an open access network have found it can be challenging to find service providers that will operate on the network. Sometimes a local ISP can step up, as in Danville and in other cases, but not always. Until a network has thousands of potential subscribers, ISPs may not be interested in offering services. But incremental approaches will often start with just tens or hundreds of subscribers.
We have written elsewhere of how important it is to have at least one strong, trusted provider on the network. An important lesson from Provo, among other places, is the difficulty in recovering once a network has a bad reputation. A bad provider can ruin the name of a perfectly good network, especially as most people will not know whether to attribute any problems to the physical network or ISP.
All of that said, open access offers a tremendous promise. Networks like UTOPIA and Chelan PUD (Washington) have been unable to pay the capital cost of building the network solely from revenues but offer some of the fastest speeds in the nation at a fraction of the price we pay elsewhere. I recall the testimony of a local business to the Utah Legislature who basically said, "Yeah, my taxes went up a little -- but my monthly telecom bill went down a lot."
Nonetheless, nearly every municipal fiber network has been built and financed with the expectation that it would pay for itself - generally breaking even years after the high upfront investments have been made. Each community should be free to choose what expectations it has in building the network it needs to ensure a vibrant economy and high quality of life for everyone. Our role has been to help them understand that choice and push back on those who want to take it away. I hope this series helps in that effort.
Many of the municipal fiber networks that now directly provide services started with a hope of working with a local partner or building an open access network. As they considered their options, they found they effectively had to choose between doing nothing and venturing into a very challenging business.
There are few easy answers for communities stuck with subpar Internet access, or even for those that regard "par" as unacceptable. When Lafayette Mayor Joey Durel was presented with the idea of a municipal fiber network shortly after taking office, he was skeptical. But he ultimately decided they should examine it - saying "shame on us" if they didn't at least see what they could do. Maybe they would hit a brick wall... or maybe they would build one of the most impressive broadband networks in the country. That was good advice.
No one solution works for every community. Thus our guiding philosophy: communities should be free to choose for themselves the solution they prefer.
Construction photo courtesy of Chattanooga Electric Power Board