"TV white space" spectrum for backhaul - Yipes!

An RCR Wireless article reports:

Sprint Nextel Corp. and T-Mobile USA Inc. urged the Federal Communications Commission to embrace a plan to allocate vacant TV channels -- known as white spaces -- on a fixed-licensed basis for wireless backhaul services.

And quotes Sprint & T-Mobile from their FCC filing (italics are mine):

"Because backhaul comprises a significant cost for wireless carriers, and incumbent local exchange carriers’ special-access charges are exorbitant, Sprint Nextel and T-Mobile must find more affordable alternatives to the ILECS’s special-access offerings.  Despite this need, the amount of spectrum in the lower bands that is realistically available for the provision of wireless backhaul services has declined dramatically over the years. As wireless carriers expand the development of their 3G and 4G wireless networks, the need for reliable and cost effective backhaul will increase.”

Doesn't anyone see how ludicrous this is?

Cell sites are fixed locations.  The vast majority are in urban or suburban neighborhoods.  Their bandwidth requirements will continue to grow, indefinitely.  In any rational world, one would purchase dark fiber to most of these sites.  In the irrational real world that is the US today, one could at least raise hell about the existence of exorbitantly priced special-access services and the lack of dark fiber.

The root problem is we have given Verizon and AT&T privileged access to the public right-of-way, without requiring them to sell dark fiber, or lease dark fiber, or offer anything less than high level service bundles at "exorbitant" prices.

What would it cost for dark fiber to 95% of these sites, if it were part of a community-wide dark fiber build like that being planned in Singapore and already substantially built in many parts of Stockholm?  Construction costs for point-to-point fiber are 20% or so above those of passive optical networks (PON), but the cost of the dark fiber alone (no electronics) is perhaps 80% of the total construction cost, so it's all in the same ball park.  Verizon has directly and indirectly reported various costs for their FiOS construction (FiOS is a PON network), but all current estimates are below $1000 per home passed.

Even at $5K or $10K, the payback for a mobile operator would be measured in months.

Why isn't anyone talking about the root problem?  We've given service providers privileged, i.e. monopoly, access to our public right-of-way.  That might have made sense in 1900, but today we should only give such access to infrastructure providers. 

The only natural monopolies are the right-of-way, poles and conduits and, perhaps, the dark fiber. 

Let's have multiple service providers competing to offer services over dark fiber infrastructure, put wireless backhaul on fiber, and save the "whitespace" for innovative new services.

Comments

[Todd Spraggins]

Not that I agree, but the LECs claim that compelling them to open these new data networks up put them at an unfair disadvantage as they try to recoup capital expenditures on the new plant. IMHO, the problem starts at the regulators who in greed charge excessive fees for the rights to the monopoly that only a socially punitive business model (read cross-subsidy) can compensate the LECs investment.
I think in earlier articles, you rightly point out that there are investment time horizons that are mismatched, especially on the regulatory front. Wall Street further exasperates the problem by expecting 18-24 month returns on investments like FiOS. We need the books to match the company (business model) to match the regulation.
However, from a pure technology perspective and in total violation of the policy rant above, the easiest way to encourage ICT growth is sharing at the active ethernet layer. I have yet to figure out how to reconcile the two beliefs

[Brough Turner]

Todd, to continue my rant... :-)

My point is we need to be discussing basic premises! i.e the extent to which current regulatory and financial models no longer align with today's reality. The only real bottleneck is the right-of-way. Given the relative rates of change in the technologies of 21st century networks, it makes no sense grant monopoly use of the right-of-way for more than dark fiber. See:
http://blogs.nmscommunications.com/communications/2007/05/useful_life_of_.html
http://blogs.nmss.com/communications/2006/05/broadband_acces.html

I don't agree that sharing active Ethernet is the best approach. Perhaps it's convenient, if you are only considering an average consumer desiring to access the Internet. But what about business users that want dark fiber for their own purposes? If you can light your own fiber, there's an enormous additional flexibility over getting stuck with one particular provider's current selection of (rapidly evolving) Ethernet technology.

Fighting about how to compensate incumbents, within the confines of historic policy, will never get us there. We need to a clean sheet discussion. Even if that clean sheet leads to an answer that is not politically possible, such a discussion will suggest how much one should fight for specific proposals like network neutrality, active Ethernet, etc.

rant, rant, rant... :-)

[Shai Berger]

Or we just get femtocells into everyone's home and a big chunk of those backhaul costs just get transfered to the local ISPs, which have more reasonable cost structures. :-)

[Nik]

Then the ISPs sad little infrastructures will collapse and we will turn off our femtocells in frustration - or is that why you put in the :-)

[Nik]

What does it mean to share the right of way? I can see that the telephone poles seem to get shared OK between the electric company, the cable company and the phone company. And I guess another player could get up on the poles without too much friction. But what other "assets" need to be shared to run dark fiber? Holes in the street? Anything actually challenging, about which Verizon could whine that they couldn't let others see/touch?

[Brough Turner]

The biggest problem with telephone poles is vertical spacing. In most sharing arrangement, the phone copany, cable company and other "low-voltage" applications are spaced 12" or 18" apart. If a new user wants to run cables on existing poles, some of the poles will not be tall enough to accommodate the new cable and thus will have to be replaced with taller poles. Assuming the power company owns the pole, the new entrant must pay the power company to replace their shorter poles with new poles tall enough to hold everyone's cables.

As far as buried plant goes, there is typically enough room under the street for multiple conduit banks (power company, phone company, cable company, plus water, sewer, gas, steam, ...), but new construction of buried conduit is relatively more expensive than overhead construction on poles. So, in many locations people are fighting about access to existing conduits. Most recently this has come up in Paris.
http://www.fiberevolution.com/2007/07/consultation-on.html

[Shai Berger]

Nik: If all our torrent'ing and skype'ing isn't bringing down the ISPs, I don't think some extra 16kbps audio traffic will be the problem.

[brough]

Shai, While femotocells may help coverage for some residences, the bulk of wireless traffic still goes through operator cell sites. In mid-2005 there were 143,000 such towers in the US. Steel-in-the-Air is a trustworthy resource for such figures:
http://www.steelintheair.com/Blog/2005/08/total-cell-towers-and-cell-sites-by.html

My best estimate is ~170K-175K cell sites in the US today. Each of these cell sites has at least a fractional T1 line, but usually one or more T1s. With mobile data, the clear migration is towards multiple T1s per cell site and that's where the operators get killed by the ILECs' exorbitant special access charges.

[Shai Berger]

Yes, I'm just being tongue-in-cheek about femtocells. It'll be a long time - if ever - before they make a significant dent in the big picture. But from the perspective of using resources most efficiently, it's compelling. I saw a stat that 40% of cell phone calls are made either at home or at the office. So, if we found a way to route those calls using the bandwidth that already exists at those places (either through femtocells or wifi-phones), that takes a lot of load off the cell towers.

[Michael Talbert]

While the WSC wants to use the white space for wireless broadband, Cellcos Sprint and T-Mobile want whites spaces for wireless backhaul, and are willing to pay for a fixed license. In todays economy, will the FCC be pressured by Congress to get some money for this spectrum?

http://voip-facts.net/voip-blog/

Seems like it would be a waste to use these valuable waves for a cellco's backhaul.

[Ajay]

Wow, great to find a blog that mirrors my own thoughts on fiber build-out and has some good data on costs and technologies. So what's stopping Sprint and T-Mobile from building a fiber network of their own and sharing it with each other and others? What specific regulations currently stop them from doing so? It sounds like they can always just pay to get the fiber strung out but they're not doing so solely out of their own stupidity. If there's relatively open access to lay fiber, I'm not sure we need to worry about the regulatory aspect of this at all.

[Brough Turner]

Ajay, the issue for anyone wanting to string fiber from one place to another is gaining access to the right-of-way. By comparison, it's relatively easy to negotiate space for a cell site. There are local zoning restrictions to be dealt with, but there are many different land and building owners that you can negotiate with. Rights of way are more difficult. They are mostly owned by the local municipalities. Access is a political issue which may come with very steep fees to each city or town along the way, or a political process that prevents anything getting done. Also note, there are existing companies, with regulated monopoly rights to use city rights-of-way for specific purposes (power, telephone, cable TV), who may or may not see a new comer as a threat.

Many of the long distance (inter-city) fiber networks have been built on railroad, power company and pipeline rights-of-way just to minimize the cost of dealing with local political processes. Such rights-of-way may be fine between NYC and Chicago, but they're not available for access to most cell sites.

[Ajay]

Thanks for the explanation, Mr. Turner, I have not seen this right of way issue discussed in detail elsewhere, not that I've looked that hard. :) I think the fundamental issue here is publicizing to voters how these municipalities are choking growth with their short-sighted policies, a classic example of cutting off one's nose to spite one's face. Ideally, the right of way would be handled in the way you describe, with anyone able to pull dark fiber as long as they're willing to pay for municipal upkeep of poles and conduit.

For FTTH, I wonder if a way around this is to have one company implement a shared solution themselves. A company could string fiber with active ethernet everywhere and guarantee that they will allow anyone who wants to use their network open access to the fiber and their COs at transparent prices. It would be difficult for the municipalities to refuse them if they're increasing local network capacity and keeping the network open. I'd imagine the company could bring a fair amount of political pressure on the locality if they blocked their getting the right of way to build such an open network by publicizing this red tape to the community. Also, if they asserted that they had no intent to be video providers but would stick to data access, leaving video to be sold by 3rd parties like iTunes over the data link, they could avoid having to buy video franchise fees. Given how cheap all the equipment has become, I would imagine that such an open network would be a great opportunity for any entrepreneurial company out there interested in the telco market. What do you think the problems would be with such an approach currently? Is it just securing the right of way to begin with? The capital costs of starting with one market and then building out to others for such an open FTTH network don't seem too high to me.

[brough]

Mr. Turner - ouch! "Brough" will do.

There are a number of places where the focus is on dark fiber in the rights-of-way. Sometimes this is a company with a franchise to install and offer dark fiber for sale or lease and sometimes it is a municipal organization that installs the dark fiber. Prominent examples are Sweden (Stokab AB for example), Singapore (see http://blogs.nmscommunications.com/communications/2007/12/singapores-next.html )and various European cities.

Costs vary widely depending upon housing density.

One thing that does appear certain is, if a private company is doing the installation, they can always make more money by bundling higher level services with the dark fiber, and therefore they will try to do that. The right-of-way is a "natural monopoly" (technical economics term - see Wikipedia). If you allow people to bundle other services with a "natural monopoly" then you are effectively giving them monopoly rights to offer those other services, even though those other services would not be a natural monopoly if dark fiber was on offer.

[Ajay]

OK, Brough it is. As a libertarian, I'm uncomfortable with aspects of the two ideas that you put forward. First, I'm skeptical that the government can adequately build and maintain a dark fiber build-out and then appropriately upgrade the equipment according to the timescales you've helpfully detailed. Given that the fiber policy in this country has been a mess so far, I think the evidence argues against a government solution, unless they limit themselves to simply putting up poles and conduit (which, as you note, have lifetimes that match normal government regulatory lifespans). However, the situation is much worse than that. A google search for "fiber right of way" finds plenty of links about the railway class-action lawsuits (brought by adjacent land owners to the rails) that have dragged on for decades against all the ILECS. You noted previously that the ILECs tried to avoid local politics by stringing fiber along the rails but because the government never gave the railroad companies the right of way for fiber, or to anyone else for that matter, the telcos will probably end up paying out billions before all is said and done. This sort of uncertain regulatory atmosphere speaks to the incompetence of those in government.

Secondly, I would dispute the use of the term "natural monopoly," one that has traditionally been used by those seeking to nationalize industries. While there is no doubt that there are scarce resources like rights of way or wireless spectrum that have to be carefully allocated with some level of community control, the wording of "natural monopoly" implies that there is ONE firm that must be in charge. The truth is that there is room for allocating scarce resources to multiple competing firms that then compete to maximize value from those resources. Implementing something like this is a delicate negotiation between markets and government that I cannot see the government getting right.

Rather, I'm suggesting that there is an opportunity for enlightened entrepreneurs to build an open platform that plays within the current, archaic rules. Surely there is an opportunity for CLECS to build their own regional fiber networks by negotiating for rights of way and legally guaranteeing to keep their networks and COs open from the get-go. Yes, I understand they have historically desired to keep the networks closed, bundle services on top, and keep all the revenue to themselves but these strike me as idiotic moves on their part. They almost always do not have the ability to provide bundled services at a level of quality as a free market would. Essentially, I would make the comparison between the Apple approach of a closed OS in the 80s vs the Microsoft open model that killed them. I believe the telcos are still stuck in the Apple mode of thinking only they can make great software and that there is an opportunity for enlightened entrepreneurs to build open regional fiber networks and show that it can be done better by open networks. Given how low the capital costs have fallen for these networks, I don't see why somebody doesn't do it.

I think we agree on much of this; I'm just trying to emphasize that I think entrepreneurs could figure this out even under the current regulations and are much more likely to do so than government coming around and doing it right. Unfortunately, at this moment, nobody is doing it right, meaning according to the lifespans and fundamental principles that you have already specified.

[Brough Turner]

Ajay, Thanks.

I don't like the term "natural monopoly" either. Usually it's just an excuse for someone to obtain or maintain an statutory monopoly. On the other hand, there is a distinct bottleneck in the right-of-way in front of my house whether that is owned by the state, the municipality or a neighborhood association, or is a deeded arrangements between adjacent property owners. I written another blog post that touches on this here:
http://blogs.nmss.com/communications/2008/03/broadband-acces.html

But I don't have a magic answer... :-)