Tag Archives: capital cost

Gigabit Seattle raising FTTH attention but not cash

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

Adding lift to a trial balloon.

The Gigabit Seattle team is trying to tap into Google Fiber’s buzz by releasing a fiber-to-the-home pricing plan that sounds a lot like what’s on offer in Kansas City, albeit for a few dollars more and with a little less freebie time. Otherwise, there’s been precious little in the way of specific information about the project since it was announced six months ago.

What I wrote then is true today: Gigabit Seattle’s financial vehicle is still a concept car. Zero private sector investors or lenders have been announced, and actual public sector contributions are minuscule.

Service is supposed to begin somewhere in Seattle “in early 2014”. The roadmap outlined in December had the project starting out in a dozen demonstration neighborhoods. No particular construction timetable has been set, even though engineering work was supposed to be well along by now. The latest announcement said that the project team will let residents know next month how they can sign up. Previously, they said that they’ll prioritise neighborhoods on the basis of pre-commitments, again similar to Google Fiber, with a 15% take rate being mentioned as a threshold for moving ahead in a given area.

It’s also unclear exactly who will be building, owning and operating Gigabit Seattle. The company behind it – Gigabit Squared – now describes itself as a “a digital economic development corporation specializing in the planning, implementation and rollout of IT-enabled infrastructure in core markets”. Not a telecoms company, in other words.

Although Gigabit Squared’s CEO says it will own its own projects, it doesn’t have any track record or significant, visible assets yet. Judging by the few financial details discussed so far, it doesn’t have a firm grasp on how much it costs to build an urban FTTH system and the operational telecoms experience of its principals appears slim.

Gigabit Seattle might be able to evoke Google’s business model in a press release, but it’s still a long way from raising the money to pay for it.

Metro broadband: without the political cards, you’re not playing with a full deck

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

Political value: the need for speed at the San Leandro public library.

There’s an argument to the effect that the prices charged for broadband service by telcos and cable companies in urban areas are higher than necessary to provide that service and make a reasonable profit.

It’s not crazy talk. You can make a case that more densely populated areas have lower per household costs – opex and capex – and that more affluent areas have higher profit margins. There are counter arguments too, not least of which is that telecoms network costs should be spread across all users. Personally, I favor the whole system approach – the more people reached, the more valuable the network – but the marginal cost approach has valid uses.

However, it doesn’t follow that an independent competitor in a metropolitan area will be able to charge less for equivalent service or the same for better service. The telecoms business has huge economies of scale: fixed operating costs are high relative to variable costs and large purchases by big companies bring hefty discounts. Particularly for television programming. A local competitor operates at a significant cost disadvantage.

A significant fraction (30%? 40%?) of households passed have to be willing to pay more ($50 per month more is a good placeholder) to either incentivize an incumbent to bring in fiber or support the operating cost and capital requirements of an independent system. The market research I’ve seen says that’s not happening.

People may value significantly better broadband services highly in many senses of the word, but not economically. At least not to the extent that an independent, privately financed metro scale FTTH overbuild in a competitive market is economically sustainable. Not yet.

Something else has to be on the table for an independent FTTH overbuild to work. Construction and operating subsidies, (significantly) below market rate financing, publicly owned assets are examples. In other words, you’re adding political value to whatever economic value is present in order to make a business case.

Whether the political value exists is a legitimate topic for debate, and some communities or state and federal policy makers might conclude that it is. The California Advanced Services Fund (CASF) is one example of policy-driven broadband investment. Leveraging a public owned electric utility, such as in Chattanooga (FTTH) or Palo Alto (dark fiber), is another. So is partnering up public assets and private investment, as in San Leandro. And there are more. And there are counter-examples too.

Claims made by some that ordinary metro FTTH overbuilds are self sustaining investments with no risk to taxpayers are at best distractions. For now, it is as much a political question as an economic one. Debate should be encouraged.

The problem with FTTH is there’s no problem

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

It’s not about finding a mass market solution. It’s about finding a sufficiently acute mass market problem.

The struggle to develop a general fiber-to-the-home (FTTH) or premises (FTTP) business model for city-wide deployments doesn’t result from a market failure. Quite the contrary. It’s evidence that the laws of supply and demand are in full effect.


Demand, meet supply.

People generally get the broadband service someone else – a business or government agency mostly – is willing to give them for the price they’re willing to pay. FTTH market research tracks closely with actual results. If you ask consumers if they’d like faster broadband, they say yes (who wouldn’t?). But when you test price points, they’re generally pleased with what they’re paying now and don’t perceive enough additional value from higher speeds to motivate them to pay more.

From the point of view of a city or other prospective overbuilder, it’s a competitive market. AT&T, Comcast and the rest do a fair job most days meeting most customer expectations. They leverage that complacency to fiercely defend their turf. Successfully, for the most part.

Cities are good at filling broadband infrastructure gaps where immediate economic demand exists, either directly or by bringing a private partner to the table. Lit San Leandro, Palo Alto’s dark fiber and Mountain View’s WiFi system are good examples. But those are specific solutions in largely unique business circumstances that also suit the particular political character of each city.

There won’t be a market-driven case for FTTH until a sizable fraction of the residents and small business owners in a community have a problem that 1. they’re willing to pay an extra, say, $50 a month to fix, and 2. can’t be solved to their satisfaction by existing technology and service providers.

Adding institutional IT budgets to the kitty is not as helpful as some FTTH backers, such as Gigabit Squared, think. An organization with an IT budget hefty enough to make a difference is really looking for wholesale service. Big IT systems need big pipes and budget accordingly. That’s helpful, maybe decisive, for funding a middle mile project, and there are examples where it’s done the trick.

You need a significant fraction of the available homes and businesses ready to spend more now, to tip the balance for an FTTH business case. Until the economic demand (i.e. marginal willingness to pay) develops, the Gigabit Squared model will only work if it leverages political demand: grants, direct tax money, cross-subsidies from other municipal utilities or other public support, in healthy quantities.

Gigabit Seattle’s financial vehicle is still a concept car

FacebookTwitterGoogle+PinterestLinkedInRedditEmail
Car of the Future as conceived by Studebaker's Director of Styling, Raymond Loewy, in the August 1950 issue of Science and Mechanics. Loewy wrote about the new styling for tomorrow's rocket age population. Via Wikimedia Commons.
Thanks for the down payment. Just need to find someone to co-sign the loan.

“Gigabit Squared is providing the capital, although details of the financing model aren’t clear,” wrote Stacey Higginbotham in a story for GigaOM following Gigabit Squared’s announcement last May that it had formed a partnership with Gig.U and was bringing $200 million to the table to fund fiber networks in as many as six cities.

The financing model was equally unclear last week when the City of Seattle and the University of Washington blessed a plan by Gigabit Squared to build a demonstration fiber-to-the-premises network in 12 Seattle neighborhoods. No cash is committed or, according to the City of Seattle, even contemplated.

Higginbotham aside, most of what’s been written about the $200 million fund is uncritical and assumes it’s a done deal. Not so. Gigabit Squared’s public statements are nuanced, to put it gently.

In the 23 May 2012 announcement, president Mark Ansboury didn’t say he had the cash in hand. “We intend to make available $200 million in investment capital,” he said. His words speak to plans, not accomplishments.

The next day, Ansboury expanded on his funding strategy in an interview with industry blogger and consultant Craig Settles.

“Our initial commitment of $200 million is based on the combination of some equity and leveraged financing. Each of our deals will be different,” Ansboury said. “So how much equity versus how much financing we’re going to do are going to be really dependent on the mix of what a community brings to the table: how much in kind, how much support and the things we need to do.”

Translation: we don’t have the money yet, but we think we can find it if the locals put enough on the table.

“It was the idea that a community has underutilized assets,” Ansboury explained. “That a community has a certain pent up service demand, that the community has the capability to aggregate capacity and demonstrate the need and value for broadband. In doing that, then you can create the financial vehicles. You don’t care if its public, private, grant…you can create the vehicle that justifies the value proposition for bringing that kind of capital to the table to help build out out the network.”

Translation: give us your dark fiber and city, county, school district and university IT budgets, wheedle some pork out of the feds and the state and have residents sign pledges (with maybe, say, a $100 deposit) to pay for installation and subscribe to service. We’ll get back to you.

His financial model assumes that if community demand can be demonstrated and big users, particularly government and educational organizations, commit future budget dollars, plus whatever broadband assets and grant money they can find, then that’ll be a sufficient guarantee for private investment and bonds, bank loans or vendor financing.

That puts the Seattle announcement in a clearer context. “The City, the University and Gigabit Squared have signed a Memorandum of Understanding and a Letter of Intent that allows Gigabit Squared to begin raising the capital needed,” the joint press release read.

There’s the demonstration of demand. Now it’s time to show that the financial vehicle has wheels.

Seattle passes the fiber (50 mega) buck

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

The unveiling of Gigabit Seattle yesterday is just the first step on a long road to building a fiber to the premises (FTTP) service for residents. The City of Seattle and the University of Washington have endorsed a plan by a consulting firm – Gigabit Squared – to “begin raising the capital needed” for a demonstration project.

Gigabit Seattle coverage

It’s not small change. The 200 miles of fiber needed to reach 50,000 homes and businesses in 12 neighborhoods will cost something like $50 million to install and light up. In round numbers, the Seattle demo looks remarkably similar to plans for building an FTTP network in Palo Alto: similar mileage, existing city-owned dark fiber network, urban terrain, prevailing wage rules, environmental standards and university-leaning demographics. Depending on the assumptions made, construction costs would be around the $40 to $60 million range.

I did an extensive analysis of the costs, potential revenue and overall FTTP business case for the City of Palo Alto earlier this year. Specifically, I looked at whether or not it could be built and operated solely on the basis of subscriber revenue, including up front charges. The short answer is no. The long answer is hell no.

On the other hand, if you build it with money that doesn’t need to be paid back for a couple of generations, then it’s possible. Not certain, though. Depending on the assumptions, such a network might generate enough revenue to pay operating costs. Or might not.

Either way, the City of Seattle won’t be picking up the tab. “The City’s only costs are for existing staff,” says the FAQ on the City’s website. “There is no additional City money going into this project, and there is no risk to the taxpayer.”

In fact, the City of Seattle is expecting to be paid for the dark fiber it’ll be contributing. It’s up to Gigabit Squared to find the money. And as Esme Vos points out, “they are an engineering and consulting firm, not a traditional ISP” with a track record to show investors and cash flow to smooth out the bumps.

So far, the only source mentioned is a $200 million kitty that gigabit Squared says it has raised in partnership with Gig.U, a consortium of U.S. universities. Gig.U is led by former FCC staffer Blair Levin, who headed up development of the National Broadband Plan. That money is intended to be split amongst at least six projects, of which Seattle is the second announced (first was Chicago).

Even though details on the cash are vague, Gigabit Seattle has surprisingly firm plans. Initial engineering work is scheduled to begin in the next two or three months, with project completion by the end of 2014.

That’s for the demo project, which will only reach 12 Seattle neighborhoods out of more than 100. According to the city’s FAQ, Gigabit Seattle has set a benchmark of a 15% take rate. Once 15% of the potential subscribers in the first 12 neighborhoods sign up for service, the network will be rolled out to the rest of the city in phases. That’s not an impossible figure to hit. Palo Alto’s research shows there’s a fair chance of getting to 15% even with a $100 per month price tag.

But first they need to find the cash to build it, and it won’t be easy if they have to show a plausible timeframe for an investment grade return on investment.

Policies, partnerships and common goals attract broadband investment to communities

FacebookTwitterGoogle+PinterestLinkedInRedditEmail

Capital expense, operating expense and revenue are the basic parameters of a business plan. With broadband-specific incentives that improve those metrics – even marginally – local governments and economic development agencies can attract private broadband investment into underserved areas.

Public policies can be tailored to significantly reduce construction costs. Uniform, broadband-friendly right of way and permit procedures eliminate a huge source of uncertainty for business planners. The more certain they are of their estimates, the more likely they are to invest.

demand study
   In the long run, it might not seem like much,
   but even a little guaranteed anchor revenue
   can make a huge upfront difference
Offering public facilities, for example vertical assets or space for nodes, on a co-investment basis and pre-installing empty conduit whenever roads are built or trenches are opened will also lower the hurdle for network builds. Of course, standard economic development tools such as sales tax concessions, community development funds and local seed capital work for broadband too.

Reducing the capital cost in a given locality improves its competitive position versus other regions by broadening the pool of potential service providers and increasing their return on investment. It also makes it easier for projects to qualify for assistance from the likes of the California Advanced Services Fund and the federal Rural Utilities Service.

Reducing capital costs isn’t always the answer, though. There are tradeoffs between capital and operating expenses. For example, it’s cheaper to hang fiber on poles than bury it, but the ongoing costs are higher. Capitalizing leases for node locations and vertical assets reduce operating expenses while raising capital costs.

Another way to reduce operating costs is for local agencies to partner with service providers on items like bulk Internet access and maintenance. One big wholesale bandwidth purchase will usually be cheaper than two medium size contracts. Local agencies might be able to set up agreements for joint pole maintenance or trenching. There’s a long list of possibilities worth discussing with prospective broadband system operators.

Documenting demand and leveraging public sector IT and telecoms budgets will brighten revenue prospects. The cost of an investment-grade demand study ranges into the low six figures for a local or regional-scale project. A service provider will spend that money on localities it already finds attractive, leaving local organizations to fund research for the area they represent.

A local agency can be an anchor tenant for a new broadband system, particularly when it can suggest ways of configuring a network so that key points are included. The agency should be able to reduce its own operating costs, while at the same time providing an early, guaranteed revenue stream to the service provider.

Given the tradeoffs between operating and capital expenses, the fixed cost of running a broadband system can be relatively low. The greatest value of an upfront contract to a system operator is its reliability, not necessarily the dollar amount involved.

It’s surprising how even small incentives – such as slightly lower costs, upfront contracts or small loans – can grab the attention of potential broadband operators and tip the balance in favor of a given locality. Sometimes, it’s just a matter of everyone speaking the same language.