Municipality provides broadband infrastructure; ISPs provide content?

October 20, 2010 4:04 PM

Smart people: I'm trying to understand the issues relating to municipal broadband. It seems to me -- a complete layperson, understand -- that the proper approach here is obvious: The municipality should build out the fiber infrastructure, and leave the provision of services open to private ISPs. What is the current thinking on this issue?

Background:

Here in Anchorage, telecom consumers are basically hostages of a duopoly comprising one cable company and one DSL provider. There is no real market pressure on these companies to offer products consumers want -- unbundled high-speed internet, for example -- at a reasonable price. So they don't.

I was pondering this, and it seems like the proper way to address this is obvious: A municipal utility should provide the physical infrastructure (the way analogous utilities provide roads and water pipes) while private ISPs offer the actual services.

To throw one model out there, the city could address the "last mile" problem by requiring that all new construction include the laying of a fiber connection.* Upgrades and expansions could be covered with a small tax either on ISPs or directly on subscribed homeowners.

I'm not naive here: I understand that there would still be a significant public cost to all this. But it seems to me that cost would be small in comparison to the costs (financial and otherwise) of being under the thumb of the monopoly as it now stands. And this is an issue that I think the public, which is widely critical of the current telecom situation, could pretty easily grasp.

So what have I missed? What is the reason this seemingly logical model has not been adopted? Or has it? What are the pro's and cons?**

* This doesn't seem that unreasonable. After all, these homes all clearly somehow got wired for cable to begin with.

** This Municipal Broadband Wikipedia page seems to think such a thing would be a good idea (see last para under "Pros of Municipal Broadband"), but the page is pretty clearly biased.

posted by Alaska Jack to technology (10 answers total)

It costs a tremendous amount of money to connect homes with fiber.

The well-publicized number is Verizon investing $18 billion to provide for 18 million customers with their FIOS service -- or approximately $1000 per home passed. This is in line with many other real-world fiber builds, as initial costs generally run $1000-$1500 per home passed to get the service up and running.

Having new construction run fiber from the lot to the street would only address a tiny fraction of that cost -- it doesn't take into account existing homes, the fiber backhaul, terminating equipment, and all of the appropriate signalling equipment -- and could also potentially create problems with compatibility if the housing developers don't follow strict guidelines.

For big cities, that's a billion-dollar project. For mid-sized communities, that's still hundreds of millions. In today's economic climate, when schools, police and fire departments don't have the funding they need, earmarking $100m+ to do the initial work building fiber to the home is just plain difficult.

Add in lobbyists from the duopoly doing their best to prevent municipal network builds from happening, and you've got something very difficult for a local government to take on.

Municipal fiber is a lot like high-speed rail. Everyone would love it, but the startup costs are tremendous, and the model hasn't been proven in this country -- even if it clearly works well overseas. So the politicians won't touch it. But having it happen really is just a matter of money and will.

I dream of seeing it happen, too.
posted by eschatfische at 4:19 PM on October 20, 2010

This is kind of, according to my very limited understanding, what is happening with the National Broadband Network in Australia.
posted by Wantok at 4:33 PM on October 20, 2010

It used to be this way when dial-up was the default residential connection. ISPs would rent or lease significant bandwidth from a larger company, and dole out connections at the local level. The difference was (I think) that the main lines were privately owned, so there was still a commercial monopoly beneath everything.

Infrastructure costs a lot of money, and to build it all up front means having a big pot of change, or getting a loan from a bank, getting a bond approved, or maybe getting pay-in from future ISPs. That's why major road improvements or water line upgrades often mean waiting for a major project to pay for such municipal upgrades, or increases in water fees (for pipes).

And then, the Government is in the business of maintaining another utility line. Power lines are (generally) not government-controlled, ditto telephone lines. In many places, there are private water purveyors. Every new function to be covered by government means another government agency, or a new role for an existing agency.

But if providing low-cost internet access to more people is a municipal goal, then such improvements might take place. I think the more common method is municipal WiFi due to decreased infrastructure costs, but 1) there are still costs, and 2) competing commercial ISPs, often interstate/international companies, get upset at "government interference" or other hoopla. Even places where muni wifi was set up, it cost too much to maintain. Still, the "challenging" FCC Broadband Plan might force some changes in the future.
posted by filthy light thief at 4:35 PM on October 20, 2010

Oops, sorry, just realized I forgot to include the link to the Wikipedia page on muni broadband.

Link
posted by Alaska Jack at 4:36 PM on October 20, 2010

Unfortunately, there is no "win" for government in that scenario. For some voters, it won't be good enough. For others, it will be an obscene amount of money to waste on them computer machines. Why should they take the risk, when the reward is maybe breaking even? When they are getting pure profit by restricting competition and allowing only one franchise per area.

I agree that they should do it, or at least set up a government owned corp like the post office to do it, but it will never happen.
posted by gjc at 6:25 PM on October 20, 2010

There are some small Canadian municipalities that have done something like this (see page two of this article).

You might argue that the closest analogous utility to broadband is telephone service (as long as the networks are neutral....), which in North American and many other places is run at the federal/national level rather than the municipal level. The Australian national network project mentioned above (which runs at a pretty high cost per household) and the plan laid out in Digital Britain basically attempt to eventually establish broadband as nationally accessible utilities. In the British plan, access for remote areas that would not be served by market competition is supposed to be funded through a levy on phone lines (i.e. a tax). In some countries (i.e. Korea, which is the world leader in broadband access), government regulation opens up infrastructure built by big telecom companies to new entrants/competitors.

In the U.S., there isn't really a plan as concrete/comprehensive, and obviously the geography is a greater challenge. However, the FCC does have a pretty ambitious national broadband plan. However, this has kind of put it in conflict the big telecom companies in a, broadly put, "big government interference" vs. "glorious American free market" thing. Although needless to say there is an economic benefit (and national interest) in establishing national broadband networks - at the very least, no one wants to end up "last" in the rankings.

Anyway, there's a LOT of policy discussion about this issue out there.
posted by scribbler at 7:11 PM on October 20, 2010

Yeah, as Wantok suggests, you might like to look at the current debate surrounding Australia's NBN. It's national / federal rather than on a municipal basis, but the drivers and issues seem largely the same.

Alongside that, there's also recent talk of a municipal-scale network installation project in Brisbane, Australia. Although it's (a) another of the regularly-occurring suggestions of fibre via sewer, and (b) probably more the local council's way of leaning on the NBN to move Brisbane up the priority list (notwithstanding the fact parts of Brisbane are scheduled for Phase 2 NBN trial rollout next year).

One issue that keeps cropping up with smaller-scale implementations like this is that too often one of the big telcos comes along and throws a bucket of money at them for exclusive or restrictive service provision rights (see TransACT in the ACT, the Springfield development just outside of Brisbane, etc.). That's a good argument for a national-scale plan independent of the existing infrastructure owners, rather than a municipal-scale plan which will generally require co-operation with one or more existing telcos in order to fund/provide the infrastructure, as it's somewhat less likely to become a closed-access network (at least in the short term…).
posted by Pinback at 7:38 PM on October 20, 2010

You might be interested in what happened in Monticello, Minnesota. The town wanted broadband, local telco wouldn't install it, so town held referendum to do a municipal installation, project got underway, local telco sued and stopped the project and then started running their own broadband. As far as I know the town's efforts are still held up in the courts. Here's the telco side of the story. The telco's actions really were unconscionable. It still makes me mad reading about that.

So from a practical viewpoint, were Anchorage voters to decide to roll out a municipal broadband service, GCI (and whoever else) would likely sue to to block the installation, vastly adding to the costs and delaying the project for years (or indefinitely).

I'm totally sympathetic, btw. One of my parents' neighbors had a teen housesit for her, he downloads/torrents gobs of stuff, she comes back to a multi-thousand dollar bill from GCI.
posted by 6550 at 8:42 PM on October 20, 2010

There are a number of communities that do this. It can result in very high bandwidths and tremendous cost savings after the high costs have been amortized. If you think of government as a community that gets together to do things that benefit the majority of the people, then this is an excellent example of government at its best. It benefits homeowners, businesses, and schools. It brings in businesses and increases home values.

But there are a large number of problems.
As has been mentioned, it is very expensive in the short term. In the long term (3-5years), it becomes quite a bargain because the cost of running such a system are much lower than what your typical internet provider charges (I've heard costs as low as $1 a month for high-bandwith connections). Most local governments these days are hesitant to make such a large short-term expenditure in the hopes of long term savings and benefits to the community.

There are enormous political obstacles. Basically the phone and cable companies don't want to give up that future, long-term profit, and will spend millions to spread the fear and uncertainty and 'socialism' arguments to make sure that voters vote it down. They'll go so far as to get laws passed preventing it from happening.
posted by eye of newt at 8:33 PM on October 21, 2010

If you are serious about this, I suggest reading back issues of Broadband Properties.
posted by eye of newt at 8:45 PM on October 21, 2010