No skin in the game

An experiment in publicly-owned fiber to the home in Utah was on the brink of bankruptcy in April. The project was oversold and underfunded, and found itself at an impasse where it had to go back to the taxpayers for a bailout or liquidate. They built it, but nobody came. A big part of the … Continue reading “No skin in the game”

An experiment in publicly-owned fiber to the home in Utah was on the brink of bankruptcy in April. The project was oversold and underfunded, and found itself at an impasse where it had to go back to the taxpayers for a bailout or liquidate. They built it, but nobody came. A big part of the problem, apparently, is that the project was saddled with a structural separation ideal that forced the public infrastructure to act as a wholesaler with third parties providing retail services. See The case for UTOPIA and iProvo: Double down or cut bait?

From the beginning, UTOPIA and iProvo either chose, or were saddled with, a business model that has proved least successful in fiber rollouts, analysts say.

In 2001, the state Legislature passed the Utah Municipal Cable Television and Public Telecommunications Services Act, which allows cities to construct telecommunication infrastructure but not become the retail service provider for those systems. Instead, they have to use a wholesale model in which they build the digital pipe and then lease the lines to retail service providers such as Mstar.

That leads to underselling of the system and friction between the municipality, which needs to see a return on its multi-million dollar investment, and the service providers, which haven’t risked as much, says Michael Render, president of RVA, a market research company that focuses on private and public fiber systems.

“They don’t have skin in the game,” he said. “The more difficult ones have been the wholesale systems such as iProvo and UTOPIA.”

Projects like this are similar to publicly-financed sports arenas. They’re great if you happen to be a fan of the sport, but not so great if you’re simply a hapless taxpayer footing the bill for somebody else’s entertainment. Not to mention the mismanagement that goes hand-in-glove with free money. In order for muni networking to be successful, it apparently needs to be run as a hard-core vertically-integrated monopoly, and that’s pretty distasteful.

The resolution for iProvo was a sale to a private company:

Leaders of Provo City and Broadweave Networks were harder to find Monday (June 30th) than cheap gas.

They were holed up in city offices hammering out the details of the $40.6 million deal that privatizes the fiber optic network, in turn taking the money-losing venture off the city’s hands. The deadline was Monday and by now, Broadweave owns the system — we think.

“I’m holding my breath hoping that it gets done,” said Councilman George Stewart, who was also awaiting word Monday. There was supposed to be a small ceremony, but nothing had been made public, even to council members, by 5 p.m. Monday. The council approved the sale in June.

The sale was concluded in July.

UPDATE: UTOPIA lumbers along, with only one city, Payson, voting to leave the consortium. Take-up rates have been much lower than anticipated, due to good competitive options, but supporters continue to have high hopes for its ultimate success.

For an up-beat view, see the Free UTOPIA wiki and its links. One bright spot is a citizen advisory board, U-CAN, which brings fresh ideas to the civil servants.

Technorati Tags: , ,

18 thoughts on “No skin in the game”

  1. UTOPIA wants their member towns to increase the bond to 33 years and more than half a $BILLION of salestaxes. That means local citizens, including poor people, have to subsidize people who can afford a ~$2,500 “membership fee” and a fat monthly subscriber fee.

    Oh yeah, it’s absolutely wonderful if you’re one of the few percent of citizens using the service. It’s not so wonderful if you’re subsidizing it and the city wants you to pay even more money.

  2. Government should build roads; it should not compete with private delivery services that operate on those roads. Structural separation is the right thing to do and should not be blamed for the failure of this network. It failed because fiber is expensive and broadband uptake in places like Utah is slow.

  3. Brett,

    Everywhere it’s been tried it has failed. Vermont BT is the only good example but their service is very expensive compared to Verizon. Here’s some price and speed comparisons.
    8M/8M: $72/month
    • up to 8Mbps symmetrical speed
    • for the ultimate internet user

    Now compare this with Verizon
    Faster Plus
    Up to 20 Mbps download & upload speeds!
    with 1-year agreement.

  4. I guess there was competition from the preexisting network, e.g. customers could choose between chip, already amortized copper and more expensive, to be amortize fibre.
    Maybe if they’d no choice, everybody would have use fibre and prices would have come down dramatically being shard amongst all users.
    I don’t know in Utah, but in Italy average cost per home passed is considered to be in the 1400USD range; that would bring a 10 years monthly cost in the range of 10-12 USD/mo so you could expect a price of 25USD/mo.
    There’s no doubt that if you compete between something which have a capex cost (new infrastructure) and something which does not have a capex cost (old copper) the latter will prevail.
    This is why I say that if we think broad-broadband (fibre) to be something useful, it is a matter of public policy and the only ones who can build it are those already running the copper network and it must be a replacement.
    This is supported by an economic analysis of WIK, the former german analysis arm of the ministry of communication (in a number of states in Europe)

  5. You’ve missed some important footnotes in the history of UTOPIA. Lindon and Brigham City both offered to pay Qwest to build a fiber network similar to UTOPIA, yet Qwest denied both cities’ requests. Qwest will complain about the cost to deploy fiber, but when the city offers a subsidy to do so and you deny it, don’t be surprised when they try and find alternate avenues to obtain that service.

    Prior to UTOPIA, Brigham City had no high-speed options aside from very overpriced T-1 lines; now they’ve spurred Comcast and Qwest into offering services in most of the city. Both companies have also offered reduced pricing in UTOPIA cities. It’s not just UTOPIA cities that benefit from the deployments as the city gets to hold the incumbents’ feet to the fire.

    Certainly there are many areas where improvement must be made. Comcast rules the roost on VOD and local programming options such as high school sports. Public awareness is low overall and many don’t even know that UTOPIA exists. Those who know that UTOPIA exists don’t know where service is orderable and their website hasn’t seen significant updates for years. I think the complete sweep of management and replacement with industry professionals (such as Todd Marriott and Chris Hogan) along with the addition of at least 8 new service providers will rectify this.

    UTOPIA’s situation is not as bad as most media outlets would have you believe. They currently meet operating expenses and once they move the NOC in-house and dissolve the relationship with DynamicCity/Packetfront, they stand to save 15-20% of their costs. In the next two years or so, they need to move from a 17% take rate to a 40% take rate with the current mix in order to meet operating expenses. Given that iProvo hit about 35% without a telcom executive at the helm (a much-needed change at UTOPIA), I don’t think that to be impossible.

    I’d say wait and see what happens over the next two years before passing any kind of ultimate judgment.

  6. Jesse, prior to UTOPIA, a lot of cities and towns outside of Utah didn’t have broadband. That’s because back in those days, broadband penetration was a lot lower and it was still being deployed. Today, we’re still working on getting broadband deployed. You can’t assume that without UTOPIA, those towns wouldn’t have any other broadband providers.

    As for Payson, voting not to back the new bond weakens the survivability of the system. It’s clear they’re in financial trouble though it’s possible they can turn things around. But I think we can all agree that Muni fiber has proven to be very problematic.

  7. I doubt that. Woods Cross still has extremely low availability for DSL and cable modems, yet it’s hardly a farm town. The first deployments didn’t occur until this summer after the city council had threatened to join UTOPIA on several occasions. Salt Lake had DSL available in only 60% of the city when they considered UTOPIA and that was in 2004! There are also many areas of Provo and Orem, both significantly urbanized cities, where you can’t get a wired broadband connection, yet Qwest is pushing FTTN and Comcast keeps on talking DOCSIS 3.0. Their official policy seems to be broadband for the haves, dial-up for the have-nots. After spending $200B under the Telco Act of ’96 for ubiquitous broadband and having nothing to show for it, what remedy would you propose if not muni broadband? The feds and states both flubbed their shots at getting it right, let’s give the cities and counties a chance.

    Payson was small potatoes and comprised maybe 5% of the total bond amount. It would have been nice to have them on-board, but it was by no means necessary. Muni fiber has run into a lot of problems, but so does any first-generation effort. They knew incumbents would fight, but not how hard (or dirty) they would. It was also a mistake to not put telco execs in charge of getting things on-track. Expectations for take rates were grossly inflated. Now they’re wiser for their experience.

    One of the reasons it’s proven problematic is that a false expectation is created that the projects need to sustain both operating expenses and the entirety of the debt service. Provo learned this the hard way by not charging city departments appropriately for their own use of the network and not assessing part of the construction costs back to those departments. In short, they expected subscribers to essentially subsidize city uses! Instead, cities need to treat these projects the same as they would any RDA project, a long-term investment that might not see a direct and immediate payback but is worth the money nonetheless.

  8. You think iProvo did things wrong but you think it’s right for tax payers to subsidize those who can afford fiber to the home in the form of a half billion dollar sales tax?

    And what is wrong with DOCSIS 3.0 and FTTN? Sure it’s not as good as FTTH technology, but is it the end of the world? FTTN can offer 50 Mbps dedicated bandwidth to each home and DOCSIS 3.0 with sufficient channels and sufficient node splits can offer competitive bandwidth. Sure it’s not as fast as fiber in potential throughput or in latency but today it can be competitive.

  9. I think that if a city wants to make the decision to engage in that kind of project, let it. There are plenty of other cities to move to that aren’t doing it if it’s that big of a problem to you. That’s what the competitive marketplace of local government is all about.

    I would also point out that you’ve mixed up your numbers a good bit. Provo bonded for $39.5M to build iProvo with a total bond cost of ~$63M over the life of the bond or $90 per household per year. UTOPIA bonded for ~$200M with a total bond cost of ~$505M over the life of the bond or $85 per household per year. The city of Orem is already saving about 600 households worth of bond from their use of UTOPIA and has seen many businesses locate there because of the network. Same story in Murray and Midvale.

    What’s wrong with DOCSIS 3.0 is that the coax last mile gets saturated… quickly. Currently, you’re splitting about 800Mbps of peak bandwidth between about 500 users on the node. It doesn’t take much to saturate it and start to degrade the connection quality for all users, especially at speeds of 50Mbps+. Some cable companies (Cox comes to mind) are trying to get down to 125 per node, but that still means that it only takes 16 users at 50Mbps to saturate it. While cable companies are looking at freeing up bandwidth with SDV and DTA adapters, they’re filling that space back up with high-bandwidth HD channels and VOD options. They’re squeezing blood from turnips.

    FTTN has similar issues: copper simply can’t scale. 50Mbps is nothing by today’s standards when many other countries are pushing 1Gbps and beyond. It’s also worth noting that DSL has extremely limited range and you’ll usually only hit peak speed at about a mile from the CO. If you start going further than that, the speed gets halved or worse.

    The point is that when DOCSIS 3.0 and FTTN hit their stride in a few years, they’ll be able to match what fiber is already doing today. How is that competitive to be years behind another technology? At least Verizon has a clue and is pushing fiber deployments, albeit with PON instead of point-to-point.

    Both phone and cable providers also charge far more than what their bandwidth costs. A gigabyte of transfer rarely costs more than a few pennies on the open market, yet we’re seeing cases where providers can be charging a dollar or more for that bandwidth. Everyone is entitled to make a profit with value-added services, but is a markup of 30 times over cost acceptable especially when both dominant providers are doing it? That sounds a like like market manipulation and price fixing to me.

    The muni fiber debate goes far beyond simple bandwidth. It’s about creating a more competitive marketplace with more ISPs and more transparent network operating procedures.

  10. Interesting remarks, Jesse, they help to illuminate the debate around government networking (GN). Your central claim seems to be this: “The point is that when DOCSIS 3.0 and FTTN hit their stride in a few years, they’ll be able to match what fiber is already doing today.” So if we go to a GN today, we’ll have more raw bandwidth at each connection point than the capitalist networks (CN) can provide for some years (or months, perhaps.)

    The other side of this is that we’ll also have to start paying for the potential bandwidth provided by the GN long before we have any need for it, and in fact long before most people have any desire for it. As people demand more from their CN. won’t they be forced to pull their fibers closer to the home? They are all running hybrid copper/fiber today in any case.

    So what’s your hurry?

  11. That central point is true of the technology. The problem is that once DOCSIS 3.0 and FTTN cap out in 5 years, they have to be scrapped for something else with more bandwidth to keep up. That’s wasteful and very expensive. Given how long we’ve been stuck with ADSL and DOCSIS 1.1, I’d bet that ADSL2+ and DOCSIS 3.0 are going to be around through the end of next decade. This also fails to address areas where there are very few broadband options, if any. Here’s a challenge: call up your local incumbents and ask what it would cost to get a gigabit connection. I’ll bet some of my internal organs that they don’t even offer the product and have no plans to do so. Fiber networks, meanwhile, do it right now.

    The bigger problem is that the demand for big bandwidth is here right now, but telcos and cablecos are ignoring it and have been for quite some time. UTOPIA snagged 17% of available homes without doing much in the way of marketing at all. It would likely be higher with better public awareness. iProvo got up to a 35% take rate. Now how can a company say (with a straight face) that it responds to consumer demand when anywhere from 20-35% of the market obviously wants something better?

    The problem is that we don’t have a free market in telecommunications. Telcos and cablecos are the gatekeepers and work very hard to stamp out competition. Qwest is using their FTTN/ADSL2+ upgrades to purposefully degrade ADSL connections and compel customers on other ISPs to switch. And since the ADSL2+ stuff uses some fiber (albeit in the same way that Pop-Tarts use fruit), they don’t have to share the lines anymore. That’s the result of a free market, that a company paid billions of taxpayer dollars to provide a vital service can assert monopoly status? We must have differing definitions of what a free market is.

    UTOPIA’s entire mission is to build a vendor-neutral next-generation platform over which private companies can provide telecommunications services. Given that Comcast has never allowed competing service providers and Qwest is actively clubbing them to death, it seems that this is the most free market solution of them all.

  12. Sure it’s not a problem for you because you’re benefitting from other tax payer’s dollars. It’s not so cool when you’re the majority of people not benefiting from it but you have to subsidize it.

    When you look at Vermont as a “successful” example, municipal fiber costs a lot more than commercial fiber operators like Verizon FiOS. The whole argument of from the pro Municipal fiber side is that this somehow makes the system more competitive and cheaper. The problem is that just the reverse has proven to be true.

    DOCSIS 3.0 and FTTN make a whole lot of sense today because that’s the amount of bandwidth that the market is demanding today. Technology always drops in price and it’s foolish to try to get 5 years ahead of the game because you pay through the nose. It’s the same reason it makes little sense to buy a $10K Intel Skulltrail computer because that kind of computing fire power will be available for $2K in less than 2 years. Verizon has successfully deployed fiber at an affordable price because of their massive scale and institutional efficiency. But having a city do it at the cost of hundreds of millions of dollars when the market for the most part don’t want it has proven to be foolish.

  13. George has it exactly right. The muni fiber experiments in Utah and Vermont simply demonstrate what happens when a group of zealots are allowed to spend other people’s money. The longer you wait to deploy FTTH, the lower the cost. But the longer you wait, the less the utility to the customer. So you have find the crossover point where the utility is sufficiently high that a significant number of customers are willing to bear the cost. Utah isn’t there yet, so you have a small number of customers using a service at a minimal fraction of capacity after paying the price for very great capacity.

    This may be fun and futuristic, but it’s an irresponsible way to spend taxpayer dollars.

  14. Just for the record, I don’t currently live in a UTOPIA service area.

    Verizon has cut costs by using GPON instead of point-to-point active Ethernet, an architecture that locks them into a single equipment vendor and has many of the same limitations as HFC. PON can do, at most, 2.4Gbps per ring whereas HFC does around 4Gbps right now. Verizon has managed to use the network more efficiently than cable providers since they got to design it from scratch which is why they can currently beat them in the market. Give them a few years, however, and they’ll run into the same node-splitting issues that Comcast, Cox, Time Warner, etc have already had to deal with. It’s cheaper now, but a lot more expensive over the long haul. You can only claim Verizon’s fiber deployments make more sense if you ignore the details of what exactly they are doing. (It should also be noted that Verizon is using the FIOS upgrades to push competitors out of the market the same as Qwest is doing with FTTN.)

    If we want to talk about irresponsible use of taxpayer dollars, let’s talk the Telco Act of ’96. It gave the telcom provider industry incumbents over $200B in tax breaks and federally-mandated fees to build a ubiquitous 45Mbps FTTP network by 2006. We were also promised more competition and lower prices. Instead, we have the opposite scenario. Prices have increased as much as 80%, there are fewer ISPs today than in 1997 and Verizon seems to be the only one trying to make good on the fiber promise, albeit very late and at a much higher price point. What happened? Federal and state regulators forgot to hold telcos to their end of the deal and seem to still be unwilling to do so.

    We determined over a decade ago that we needed next-generation services and the market has failed to provide it, even when offered fat sacks full of cash money to make it so. We have two options: beg the incumbents like Oliver Twist (“please sir, can I have some more?”) or have our local community take the torch. At least local government is trying to do something. What is private industry doing? About as close to nothing as you can get. What are pro-incumbent zealots like you proposing? Nothing but carping about what you don’t like. How about come up with a competing proposal that accomplishes the same aim instead of being an apologist for a group that has demonstrably abused their duopoly status to screw consumers and feed from the trough of federal dollars?

    (Pardon my lack of civil tone, but you’ve pressed my buttons on this one.)

  15. You’re doing a fine job of proving my point, Jesse. There’s no point in stringing point-to-point fiber or Cat 6 to each home until you can establish some current or near-future utility that such an arrangement can facilitate that’s so compelling it justifies the extra cost vs. a low population fiber ring. FiOS only serves 32 homes from that 2.4 Gb/s budget, so it’s good for 75 Mb/s minimum with a spectacular burst rate. And that’s just per one wavelength, and the cable can obviously support hundreds, if not thousands. This is an arrangement with a lot more headroom than baseband Ethernet, so not to worry.

  16. That 75 Gbps MINIMUM budget is only for data but more realistically it’s probably double that for minimum and it’s probably good for a gigabit most of the time with statistical multiplexing. Verizon can always go to 10-GPON if its necessary but they’re not even close to that point now and they have way more last-mile bandwidth than they need today. The idea that Mr. Harris would criticize Verizon’s FiOS implementation when it’s way faster and cheaper than any muni-fiber implementation is laughable.

    Furthermore, Verizon runs no additional compression on its digital HD feeds on other wavelengths and they can run hundreds of these HD channels or more independent of their BPON or GPON data connection.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.