Terrence McGarty is founder of the Telmarc Group, an investment and technology development company that focuses on telecommunications, software, and medical management systems. His WTF 2004 presentation addressed the development, design, and deployment of local broadband networks and why traditional telcos won’t be able to compete. What follows is a partial transcript of his talk.
We go out an actually build these things. Let me start with a story that got us into this. I was a group president at Warner 20 years ago. I spent years at Nynex in the cellular group. And I spent the last seven years in Europe. We built and operate the largest fiber network in central Europe. If you want duck, sausage, and sauer kraut, thats a great diet. I’m happy to be back here.
When we moved our operations to Prague and I was speaking at MIT, a woman came up to me and said we hate our cable company and we want to build our own fiber network. At first I said she was crazy, but we did a feasibility study town by town in New England. Localism is an important factor. After about 26 of these feasibility studies and analyses of these towns, towns fell into three categories: towns that were economically viable, towns that were not economically viable — largely the upper 2% of the income bracket — and the politically unacceptable category.
It’s politically incorrect for two reasons. One, a town remembers when they put in the trolley line, a very bad memory. And two, in one case, we were threatened to be sued by Comcast. We’re not in the business of litigation. What I’m going to talk about today is what we learned going out selling to towns and what we’re doing now.
We started out looking at these as opportunities to build municipal broadband networks using municipal bonds as financing. We stumbled into the department of agriculture’s bond financing, and they said welcome to a government agency. We applied for the money last July, and the net result was that we got the money. Whoops, now we need to build the network. We started in Hanover, New Hampshire, because the town manager was interested, because of Dartmouth, the hospital — and because it was an atrractive market to start in.
How do we make this work? What we’d really created was a public-private partnership. We went out with 80% federal financing, 20% private capital, and gave towns equity participation. Then they sit at the board. We took the adversarial approach the cable companies had done and made it participatory. The other thing we said was what are we really selling? Are we selling video, broadband Internet, and telephony? Or is there something else here that’s being provided? We’re providing a virtual local area network, a network that would connect the different towns and cities, and a portal to a very wide area network. We’re providing localness in an open network.
We started an open video system, which terrified the cable guys. Openness is good. Move the intelligence to the edge of the network. Have a public-private partnership. Now we have a lot of local support. We have local committees doing our marketing for us even though we’re a carriers carrier.
There are about 250 municipal networks active today. About a week ago the Supreme Court ruled that towns are not entities. Under the 1996 act, any entity can become a telecommunications carrier. Scalia was the only one who voted against that. If you take a reading of that ruling, it basically says it’s a states rights ruling and that the states have the right to control any entities in their states and rule that they’re not entities. There are also local power companies. There are approximately 600 entities — that bad word — who have started delivering fiber to households in the United States.
We’re focusing on about 150,000 households, putting fiber to the user. 80% of the households in the town are economically viable for fiber only. For the other 20%, we go wireless. This is New Hampshire. Our strategy is to go through the counties and install a network. This is the Keene cluster, a collection of towns sharing a common infrastructure. 80% of the people drive on 20% of the road. I don’t have to build out 70 miles if 80% of the people live on 14% of the road.
What does it cost? The fiber is roughly $25,000 a mile. That’s the biggest problem. You try to cover as few miles to get to as many households as possible. I have a database of thousands of pictures of telephone poles in New England. The problem is delay. When you apply to Verizon for pole attachments, there’s a six-month waiting period. They won’t deny it, they’ll delay it. If you’ve got a clock running with money, you’re in deep, deep trouble. The electronics are dropping. The interface in the home is down to the $200 level. The drop from the street to the home is about $200. As you increase the penetration, the costs drop and you always have a positive cash flow. We go in and use the local ISPs. We have a deal with AOL, but today we don’t actually know if anyone is home at AOL. So we go to the local ISPs. We don’t outsource. We insource. The local customer is the ISP’s customer. Then the ISP is our customer. That works very effectively and efficiently.
Will the incumbent local exchange carriers ever be able to build broadband? No. Do they have the money? No. Is it in their interest to do this? Not really. Where would they do it? The large cities. Will the architecture allow them to do it effectively and preserve their monopoly control? No. Openness is the cornerstone of a fiber network. We create new applications and new innovations by being an open carrier. Inherent to the carriers are hierarchical organizations.
One of the questions that keeps coming is that if you suddenly deploy an open broadband network, what does that do? It creates a new electronic marketing and distribution channel. That can be disruptive to existing relationships. I can buy video directly from the studio. That’s the way it can be, and that’s the way the world’s going to go.