Broadband Bytes for 2014-04-11

Broadband Bytes for 2014-04-04

Broadband Bytes for 2014-03-28

Broadband Bytes for 2014-03-21

The end of the road? It’s Macquarie or bust for UTOPIA

utopia-logoPart of the visit with UTOPIA was figuring out how they’re moving forward and assess the financial health of the operation. Unfortunately, it looks like despite hitting a number of financial targets, they’ve fallen short financially. Right now, they’re still running a deficit on operations (I hear around $150-175K-ish a month) and it’s not closing as quickly as they need to meet the home stretch of the UIA plan. Right now, break even is about three years out as they concentrate on business customers. Given that subsidizing operations isn’t an option, that puts them in an operational pickle.

They were pretty candid with me when I asked questions. The switch from focusing on homes to focusing on businesses came at the explicit request of the cities. Business accounts cost about the same to get hooked up as residential, but they generate about four times the revenue. Even small businesses are worth about two homes a pop. With limited resources, they have to focus on those specific areas to at least try and cover operating expenditures.

The biggest barrier to most businesses is the brand name. Despite offering great speeds at great prices and excellent SLAs, their PR kind of bites. (Shocking, right?) The limited deployment means it’s much harder to serve multi-location businesses, though service providers have been creative in using UTOPIA where it can be had. Small businesses are challenging because many of them want the reliability of UTOPIA over cable and DSL, but there’s not a cheap enough product offering for them.

There’s some bright news, though. They’re signing up about 60-70 new accounts per month without doing much marketing or direct sales. There’s been a marked increase in high-dollar products like gigabit and even 10Gbps circuits. Most of the cities (including Layton, West Valley City, Murray, Orem, Brigham City, and Tremonton) are using the network extensively for city operations and reducing their own internal costs. If you’re in a completed footprint, you can pay to get hooked up. If you already have the wire on the side of the house, they don’t even have to roll a truck to do it. And operational costs are much closer to being covered than they ever have been.

And yes, the silver lining is pretty thin right now. Despite all of the good things UTOPIA has done (building to ~95K homes with ~11.5K taking service, offering cheap gigabit, 17 service providers, etc), money matters. I support UTOPIA, but the stark reality is that without any money, it’ll die on the vine. It’s here because the early problems, including an illogical build plan, the Qwest right-of-way lawsuit, and constant legislative meddling, tripped things up right out of the gate

I’ll be blunt: either the Macquarie deal happens or we could be waiting decades for fiber to get to more homes. UTOPIA is pretty fortunate to be negotiating with a company that’s bringing a lot of mutual benefit including the cities retaining ownership of the network. If it doesn’t pan out, though, I think we’re hosed.

Google Fiber will leave the duopoly intact and only change the players

Google_fiber_logoThere’s a lot of talk about Google Fiber bringing competition to the marketplace, but I think it’s a lot of smoke and mirrors. In reality, Google is very good at leveraging its brand to cover up many plays it’s pulling straight out of the incumbent playbook. In the end, they’re acting a lot more like Comcast than I think many are comfortable with confronting, much less admitting. I’ll make a bold prediction: Google will be the new Comcast within 5 years. I’ll make the case as to why.

The biggest problem I see with Google Fiber is their practice of redlining and cherry-picking. Their pattern so far has been to break up neighborhoods into very small “fiberhoods” and only build if there is sufficient demand. Given that they ask for either a 2-year contract at $70/mo or $300 to install the service, it’s easy to see how low-income neighborhoods are unlikely to reap the benefits of a new vertical monopoly in town. This is the kind of practice that Comcast and CenturyLink has been dying to get into: upgrading only the most lucrative areas and letting the low-margin subsidized lines languish.

Let’s not also forget that you have a very limited time to sign up or be left behind. Google has made it very clear in their FAQs that it has no plans to reopen to new subscribers once the signup period has closed. In a rental-heavy market like Provo, this could exclude a large proportion of the user base from ever getting service. There’s also no mention on if someone can reactivate a terminated line to get service again.

So why is Google doing this? Based on the company’s history, I think it’s all about costs. Google is famous for designing hardware to meet very specific needs, a process that leads them to be extremely efficient. It’s not much of a stretch to think this same efficiency is being applied to building Google Fiber. (Warning: speculation ahead.) My theory is that they’re building the network exactly to capacity in an effort to reduce costs and maximize ROI. If you don’t have to plan for potential future additions or build the network where demand won’t meet certain profit goals, you can slash your cost per subscriber to under $1K. Assuming they make about $35 per subscriber per month in profit (which is consistent with numbers I’ve heard from UTOPIA service providers), that works out to a payback of under 2.5 years. With a 5-7 year contract, it’s not hard to see how Google is going to make money hand over fist.

This makes it all the more curious, then, as to why they need all sorts of concessions from municipal governments to make it happen. While their official checklist for the latest round of cities claims that no subsidies are expected, it’s hard to see how really going to be the case. Kansas City greased the wheels with millions in tax dollars whereas Provo literally gave away the network to get Google’s attention. This is much like CenturyLink’s hypocrisy in decrying municipal systems as unfair competition while available themselves again and again and again of available tax dollars. Google may say that they don’t want subsidies, but the unspoken understanding is that without significant municipal concessions, they’re probably going to pass you over.

With all of these behaviors that remind us of the many ways in which Google is behaving like an incumbent carrier, it doesn’t take much to connect the dots. CenturyLink is probably going to let residential wireline rot on the vine as it pursues high-margin business services. Comcast will quickly hit the end of its upgrade capacity and focus instead on entrenching its vertical monopoly between content production and distribution, replacing CenturyLink as the “cheap” provider. Google is then free to fill the void left by Comcast as the “fast” provider, and we’re right back in the non-competitive state we had before, just with different names on the door.

To be quite frank, I don’t think Google has the capacity to be a good ISP. Google has a history of very technocratic decisions, depending heavily on the technical specs, prowess of their products, and brand name to compensate for their lack of customer service and direct marketing. This is uncharted territory for them, especially since they haven’t proven to be very adroit at dealing with entrenched companies whose lobbyists have very deep government connections. While I’m willing to be proven wrong, I don’t think they’re really prepared to survive in the regulation- and politics-heavy world of telecom, especially when the margins are relatively low. Once the reality of operating a utility settles in, you have to wonder if Google is going to treat their fiber products like they did their WiFi network in Mountain View.

When cities are considering Google’s proposal, they need to look at it with clear vision. There’s a limited amount of skin they have to put in the game, but they’re also not getting the same level of benefits that they could be. Overall, Google is offering to rearrange the deck chairs, not right the telecom ship. I hope that more cities will be wise to it.

Broadband Bytes for 2014-03-14

Confirmed: Macquarie tips its hand on details of the upcoming proposal

macquarie_logo_2638After hearing some of the rumors about the Macquarie deal, I’ve been watching skeptically to see how much of it is wishful thinking and rumor-mongering. I’m now seeing that almost all of it is completely true. Macquarie had a representative answering questions at the most recent Brigham City city council meeting (skip to 43:30) and he confirmed a lot of what I had previously reported.

Macquarie wants a 30-year contract with the cities to expand and operate the network. There would be no transfer of ownership of the network and the cities would retain the title to all of the assets. Their purpose in doing this is to secure low-risk (in their words, “boring, stable”) investments with a moderate rate of return for pension funds and life insurance pools. Their view is that because they don’t have the typically myopic view of most of the telecom market that requires a very fast, very high return on investment, they can approach differently and see the long-term effects. Macquarie is committed to building out every home (165K of them by their count) to a “service ready” state. I assume that means including all portals and needed cabling so that getting service is as easy as a phone call.

In exchange for the contract, Macquarie would assess a per-address fee to the cities. While they recommend a utility fee with waivers for the financially indigent, the cities are given full latitude to determine how this fee would be collected. They can’t quote specific numbers yet since they’ve just started to receive proposals in response to their RFQ, but I’ve heard numbers between $15 and $25 per month. Macquarie would commit to providing the connection to every address and service providers would offer a free “basic” level of service comparable to high-end DSL or low-end cable. The service providers would not be charged a fee for access to these customers and would only incur the customer service costs. Word is that they view this arrangement favorably since this gives them a way to market 100Mbps and 1Gbps services to those customers.

Cities are actually going to get a pretty sweet deal on those upgraded customers too. Macquarie wants to make sure the cities keep almost all of the funds paid by the upgrading customers. These funds will help pay off the debt service and could be used to reduce the utility fees. I personally also like that the cost of UTOPIA will become a transparent thing. Brigham City is planning to leverage this universal buildout to switch to smart meters that will pay for themselves within 2 years and greatly reduce the operating costs of the city electrical utility.

Macquarie is doing a good job at keeping their ear to the ground locally. They were on top of Sen. Valentine’s attempts to amend SB190 from the floor and worked vigorously behind the scenes on both that and the original ill-conceived bill. They’re keenly aware of the perception of UTOPIA being forced on city residents and want to focus on showcasing the benefits of the network. They also took a moment to slam some of the woefully uninformed comments from the previous meeting by pointing out that they’ve been doing business in the US for two decades and have over 5,000 employees here.

The devil is always in the details, but so far this looks like a very solid proposal that’s win-win. The cities get world-class infrastructure and money to pay the bonds, the citizens get at least a free baseline of Internet service (with cheap upgrade options), and the rest of the state gets the potential to get gigabit everywhere else too. Macquarie also gets their toehold on what they believe to be a great long-term investment for low-risk portfolios, potentially spurring other companies into an overbuilding gold rush. I have yet to see anything giving me pause.

PS As a bonus, note that Ruth Jensen spends most of the council meeting continuing to concern troll on both smart meters and the free tier of service. She also comically states that DSL and cable are “good enough”, parrotting the standby line of the “you’ll take it and like it” incumbents. Jensen also goes so far as to insinuate that people actually LOVE their existing options, apparently unaware of how poorly they have been performing in customer satisfaction rankings for well over a decade. Her near-automatic gainsay reminds me way too much of the Monty Python argument clinic sketch.

Victory Again: HB60 withers without so much as a hearing

"HB60 is dead, Jim." (from Wikipedia)

“HB60 is dead, Jim.”
(from Wikipedia)

Without any official action, HB60 died in the same fashion as SB190 and was sent back to rules to rot out the rest of the session without so much as a committee hearing. Between these two bills, municipal broadband advocates in Utah have racked up some big wins when we’re used to nonsense punitive laws sailing through without any opposition. What changed this year was being on top of these bills and swiftly letting legislators know how we feel about them. It might not be a bad idea to write Rep. Curt Webb to express appreciation that he backed down on HB60 once we spoke up. I have a gut feeling that he was had.

And no, I’m not going to take credit for anything. Each of you who took the time to write legislators and share this information as widely as possible across social networks (Facebook, Twitter, Google+, Reddit, etc.) had at least as much of a hand in this success. I think we’ve proved that we’re a force to be reckoned with. We just have to show up.

While I think we’ve reached a point where the war is now going in our favor, it’s not over. I’m sure Comcast and CenturyLink will be more than happy to use their hatchetmen at the Utah Taxpayers Association yet again to try and throw up roadblocks next year. Sen. Valentine is not the type to go quietly into the night, and SB190 (or something like it) is probably going to be discussed in interim committees. I have little doubt that other restrictive measures will come up too. Once I find out about them, you’ll know too.

Legislation Proposal: Assignable tax credits for building gigabit broadband

After seeing SJR18, it makes it obvious that some kind of broadband bill is likely to come to fruition next session. While the current session isn’t even over, legislators can start submitting new bills on May 13 for the 2015 session. If we get in early, we could have a shot at influencing the debate for good. A local broadband expert clued me in on something he’s been working on at the federal level, and I think it would work at the state level: an assignable tax credit for building gigabit broadband.

Tax credits are nothing new when it comes to incentivizing broadband construction, but they often come with such weak conditions that they amount to nothing more than discounts on a system that’s barely a marginal improvement over what we already have. Without strong conditions, we end up with situations like the $300B+ scandal that is the Telecommunications Act of 1996. We can’t afford to flush tax money down the drain like that.

Here’s how we fix it. The first condition should be that tax credits will only be available for a service that delivers 1Gbps or better symmetric connections. This ensures that we aren’t dropping money on last-gen ADSL2+ or DOCSIS 3.0 systems. There should be an additional tax credit if the retail price is less than the average selling price for a current broadband connection. Based on data from SpeedTest.net (average download speed of 26.42Mbps at a cost of $3.62/Mbps), this would require gigabit service to be sold for under $95.64 per month including below-the-line fees. Finally, an additional tax credit should be offered if the infrastructure offers the choice of two or more retail service providers. This would encourage real competition rather than simply shuffling the deck on who the incumbent providers are.

This creates a very level playing field for all providers. If Comcast is serious about using the upcoming DOCSIS 3.1 standard to deliver gigabit service, they could get the tax credit. If CenturyLink is serious about building gigabit FTTH, they could get the tax credit. If Google is serious about expanding to Salt Lake City, they could get the tax credit. If Macquarie is serious about expanding UTOPIA across the entire state… well, you get the idea. Setting serious standards is what makes these tax credits actually worth something.

So what does this “assignable” bit mean and why does it matter? It means that the tax credit is applicable to the individuals served by the network, but the entity actually building the network would be eligible to claim it on your behalf and pass it on to you. This means consumers don’t have to front the money and get reimbursed, but rather companies with the money to invest will do it for you. That makes it much more likely that the infrastructure will actually get built.

Right now, I need to find a legislator willing to sponsor this kind of bill. You can help me find one. Write to your members of the House and Senate summarizing this proposal and copy me on it. We may have a shot at turning the tide for good in 2015.