Fiber to the Home — For $10,000

20 Comments

Want fiber to the home? If you’re not in FiOS territory, it’s going to cost you. I’ve been following the various research projects over at Google (s GOOG) and up in Canada debating the benefits and business models of building fiber to the home, and wondered how much it would cost. Today I sat down with a table of Time Warner Cable Business Class salesmen to hear that it wasn’t possible. Finally, one told me it might happen, but I would have to pay the cost to build out the fiber to the home — about $5,000 to $10,000, depending on how far out from the core fiber line I lived.

Okay, that is some serious infrastrucutre cost right there, something many of these “homes with tails” studies gloss over.  I have no idea how accurate that price is since it’s not something people want to talk about, but Verizon FiOS estimates range from $850 per home to $4,000. So we’re not far off on the low end and perhaps, since it’s a one-time effort rather than a neighborhood-wide deployment, the high end might be reasonable.

Regardless, that’s about what I would pay to be hooked up to the city’s water system or sewer system. I am of the the belief that broadband is more like water or electricity in that everyone should have access to it, but as a homeowner looking at a $5,000 (or $10,000) cost for something that doesn’t even begin to approach that type of value, I think homes with tails will be a hard sell. Until there’s a killer application that requires that kind of infrastructure investment — and HD streaming for all content might count — I can’t justify the cost. So, the bandwidth gap between copper and fiber to the home seems insurmountable for now — but maybe, when LTE arrives in a few years, wireless networks can bridge that divide.

20 Comments

thegeniusfiles

I live in rural Wisconsin. Ironically, there is a fiber cable running right through my front yard, yet I must pay CenturyTel to run a phone line across and under the road to get DSL because the fiber cable belongs to a company that the Feds don’t allow to serve my location. You might be surprised at just how much fiber is already out there… but government regulations may prevent your access to it. The whole argument about infrastructure costs is often wildly inflated.

Brett Glass

Who says that broadband does not have the same value as water, sewer, or electricity? Also, $10K is cheap for the amount of work required to bury fiber. That’s why wireless is so much more practical and cost-effective. Our company can put high speed wireless on your home for $200 or less.

John Thacker

And remember anyway that water isn’t government-provided in many rural areas, even some suburban areas. Wells and neighborhood wells are quite common. Electricity and especially the phone network are better comparisons to Internet service, due to the need to hook up to a network.

John Thacker

I am of the the belief that broadband is more like water or electricity in that everyone should have access to it, but as a homeowner looking at a $5,000 (or $10,000) cost for something that doesn’t even begin to approach that type of value, I think homes with tails will be a hard sell.

And you complain in that linked post about the 1% of US homes that can’t get broadband at all. Don’t you think that for those homes it would cost even more than $10,000 to run fiber to?

The government does subsidize running phone lines to remote rural locations. But that’s why there are people in very remote locations where the government subsidizes their local phone service to the tune of $5,000 to $10,000– per year, not a one time fee. If someone wants to live in the middle of nowhere, with no one else within miles, fine, but I don’t see why their cost has to be subsidized.

Dave Burstein

The fiber tail idea is inexpensive iff the whole neighborhood is done at the same time, not house by house. The Time Warner salesmen are way off base with the $5-10K figure. I’d guess that they are estimating digging streets and/or running fiber exclusively to you but not your neighbors at the same time. If you’re the only home in a great distance, that could be right, but few of us are like Gordon Moore with miles in some directions to his neighbors.

Even then, if the fiber was put in automatically with the water or electric when the house was built (a good idea and now becoming typical,) it is actually cheaper than copper. “Fiber is cheaper than udon” is the Japanese saying. The fiber itself is cheap; the cost is labor, which is needed for new copper as well. So fiber is the right choice for any new build, which is becoming standard.

The fiber long tail work makes far more sense if you think of it as built “neighborhood at a time,” as Verizon generally does. That brings the cost to something more like $1,000/home.Built into your phone charge is the copper currently to your home at $10-20/month. Fiber comes in similar. Over 60 months, that’s $17/home. Over 120, much less. It’s realistic to expect fiber to have a very long lifetime, with only the end electronics needing to be occasionally changed out.

The likely cost of fiber to 95-98% of U.S. homes is somewhere between $500 and $2,000 each, with something on the lower end most probable. Verizon’s 2008 number for running fiber is $700 per home passed and dropping. Iliad/Free.fr’s figure is €1,500, which is based on 25-40% penetration, I believe. (I have to check with Xavier.)

The $4,000 figure for Verizon is from Craig Moffett, who triples Verizon’s cost per home because he assumes only 1 in three will take the service. The BSG figures make a similar assumption. Moffett’s figures include the cost of the set top and the home wiring, etc (Verizon puts at $650. Those installs are taking a long time, still. ) There are many other but much smaller issues with Moffett’s data, but the key thing is it’s looking at a different problem than you are.

sm

Ok, I’ve got FTTH last month for a price that might shock you – Euro 20, I pay Euro 30 monthly, and soon they will be selling settop boxes for TV.

Fiber Fan

Based on the number of truck rolls and excavation Verizon did in our neighborhood, I’d guess the $4K figure is gonna be pretty close.

But do i want that cost bundled into my new house? No way.
Give me a DSL over copper, please.

I don’t know what you guys think is so necessary to go over, say, 256kbps, but I haven’t seen it, and it ain’t worth $10K, or $4K.

Verizon was given a monopoly on FTTH; they will pay off their investment soon enough. Fios costs at least $50 a month, and typical users will be paying $100-150. Cable will not be able to offer true 20MBPS uploads.

Stacey Higginbotham

Tom, if I could get U-verse I would try it, but alas, I’m not in an area that’s close enough to the the CO, or populated enough depending on who I ask at AT&T. It’s not a great demographic for broadband is my guess.

Garry, of course it occurred to me. I’m hoping someone can shed some more light on possible pricing beyond the Verizon numbers.

Allen, I should say it doesn’t approach that type of value today. I have no doubt it will, but when contemplating my household budget it’s not something I can defend, even amortized out over a period of years. I hate to fall in with the telco thinking, but I can’t be a visionary AND pay my bills at this point even if it does make me sad.

@Tim and Todd, I’m thinking some kind of fiber to the node deployment used as backhaul for LTE. But that may not end up being the panacea I hope for.

Allen

“$5,000 (or $10,000) cost for something that doesn’t even begin to approach that type of value”

Interesting. That’s what the carriers are saying.

Garry King

Did it occur to you that the TW guys were pulling your chain?? At least where I live in Texas, the biggest cost is in the permitting and labor with material being only a small fraction of the CPE connection cost. The actual cost of fiber isn’t drastically more than than that tired old Coax the TW guys want you to be happy with.

Tom Schwitzgebel

Thus the beauty of AT&T’s U-Verse product. They bring fiber to a node near your neighborhood and then use the POTS twisted pairs to bring it into your house. I have the service and love it.

Cheers!
Tom

Heather Kennedy

How is this not the same 1985 argument about cable, i.e. we didn’t get cable til I was in 5th grade and survived on ONE channel until then. The people 4-5 miles up the road had to wait even longer. I wish I could show you the map/FIOS service in my area. Sometimes it’s just bureaucracy that keeps FIOS out. Or Comcast.

Jesse Kopelman

Sorry Stacey, for some reason I hallucinated Om in the byline.

Jesse Kopelman

@Om

The real problem is that fiber deployment costs tend to be highest where home prices tend to be lowest. For most places where the average single family is > $300k, even $5k for fiber is probably too high, but out where $100k buys you two bedrooms and more you could easily be looking at at least $10k for the fiber (not mention greater distances to the headend, making for more expensive PON equipment). So what is a home builder to think? In places where the cost of adding in the fiber is easy to hide in the bottom line, the carriers are probably going to build out themselves. In places where the fiber connection is a significant percentage of the overall price, no carriers are coming unless they are forced.

Tim

LTE also isn’t going to solve the problem. Todd’s right, backhaul for LTE is a real challenge, mostly because of the radio vendors making it an issue due to inflexible architecture and a refusal to innovate at rapid pace to maintain the revenue stream. Greenfield WiMAX deployments will bring us part of the way, and in five or ten years we may see a huge level of improvement in bandwidth and services available over the mobile network, but by that time I think FTTH will be a national priority. There are plenty of killer apps out now that will drive this for us – the question is who will capitalize on them and how the revenue gets distributed.

Tim

I’ve never bought these kinds of arguments against FTTH deployments. And I don’t necessarily buy the cost, either, without hearing it broken down into its components. I think when we see more neighborhood deployments the cost will come down significantly for your users farther off the grid, as the infrastructure will already built out.

The reliability of these connections tends to be higher than your average cable connection with the exception of the backhoe factor, so we can expect operational and support costs to be lower though probably not capital expenditures. This translates into more money for the operator, particularly if we really start seeing a service delivery ecosystem in the US, which I don’t think is as far away as we make it out to be.

Your average DirecTV customer might pay $90/month for service – this is $5,000 over five years. Cable Internet right now around $15-$40/month – around $2,000 or so for five years. Seeing nationwide fiber deployments for the delivery of these services suddenly becomes pretty attractive. People throw out these numbers as if they’re really hugely expensive, but I think in your average suburban areas, particularly ones that have to have a lot of new infrastructure run anyway, it’s not that dramatic an increase as we’d expect.

What it will really take for FTTH deployment in the US, and I think this is obvious to everyone, is municipalities and states making it a priority. A great number of them are doing the right thing, and demanding that the “utilities” (Verizon, for instance) provide the service to all residents rather than only a partial subset of “luxury” residents. Your utility doesn’t want to do that unless there’s some kind of agreement from the municipality which it doesn’t want to give. The municipality would find an easier time of this if the builders in the area shouldered part of the cost as part of the home price, or there were better tax initiatives in certain areas. But this isn’t likely to happen with the current state of economic affairs, where a lot of municipalities are going bankrupt.

Anyway, the cost is not the problem. That problem is solvable. The problem is prioritization.

Todd

Oh Contraire, LTE has a huge backhaul problem so expect a lot of femto deployments in the first iteration, all dependent upon that wireline broadband link.

Comments are closed.