April 2, 2010 – It should be affordable; every American should have access, and its ubiquity is critical for the nation’s future. Health care? No, broadband, and very high-speed broadband at that, according to two recent proposals from the FCC and Google.
Amid claims of untapped applications and business potential and a poor showing of the U.S. against other countries on the broadband penetration curve, those campaigning for aggressive bandwidth expansion have gained much media attention. But market realities, including costs as well as demand factors, leave others wondering what the urgency is all about.
While there’s little or no evidence of a great consumer clamor for ultra high-speed broadband, Google has clearly touched a nerve at the community level with its recently announced plans to build 1 gigabit-per-second FTTH networks “in a small number of trial locations across the country” and offer service “at a competitive price to at least 50,000, and potentially up to 500,000 people.” Google’s RFI attracted responses from 600 communities and more than 190,000 people by the March 26 deadline.
The FCC’s National Broadband Plan, meanwhile, advocates regulations that would give at least 100 million households “affordable” access to at least 100 megabits per second downstream and at least 50 mbps upstream. Schools, hospitals, government buildings and other so-called “anchor” facilities in every American community would have access to 1 gbps.
Both plans are ambitious, but Google’s has generated the most buzz – and stunts, including Topeka renaming itself after the company and Sarasota’s mayor jumping in a shark tank, all to draw attention to their bids. Some in telecom wonder aloud whether Google’s proposal is itself a PR stunt.
“That was like throwing a big rock in the national broadband pond,” says Mike Paxton, principal analyst at In-Stat’s Digital Entertainment Group. “A lot of reaction. A lot of movement. A lot of people being very enthusiastic. A lot of people being very skeptical.
“There are so many details that are lacking that the further away we get from that announcement, the more skeptical the service providers and technology vendors seem to be.”
“Affordable” and “Competitive Price”
The FCC’s proposal uses “affordable” 45 times, while Google says its service would be offered at a “competitive price” – all vague, relative terms. Google and the FCC didn’t respond to multiple interview requests, making it difficult to determine whether there are formulas behind those terms or whether they’re just well-intentioned placeholders.
“I don’t think the commission has in mind some specific amount,” says Harold Furchtgott-Roth, a former FCC commissioner.
More likely, some say, the FCC would follow the 1996 Telecom Act, which has several provisions addressing affordability, as well as comparability between rural and urban areas.
The comparability provision “basically states that consumers in rural areas should have access to the same services and prices as consumers in urban areas,” says Dan Mitchell, vice president of legal and industry affairs at the National Telecommunications Cooperative Association (NTCA), a trade association representing more than 580 small and rural telephone cooperatives and other companies. “The act doesn’t define what comparable means in terms of pricing, but the FCC is required by statute to apply the comparability principle” for services and prices.
For example, the FCC could look at the average speed available to urban consumers during peak busy hours and the average price of those services to calculate what appears to be the affordable threshold for consumers in that area. The FCC then could do the same survey in rural areas and compare the two sets of figures.
“If we had an average speed of 5 mbps at $42.50 in urban and rural areas, that would enable the FCC to say, ‘We’ve met our affordability and comparability requirements within the act,'” Mitchell says.
The fastest U.S. residential broadband services on offer in the U.S. come from two medium-size cable companies – Suddenlink’s 107 mbps, which costs $120 outside of a bundle and Mediacom’s 105 mbps service, priced at $150. It’s anyone’s guess how much a 1 gbps service would retail for, but it probably would be significantly more unless the operator can sell enough services over that pipe to price access as a loss leader.
That strategy almost certainly will be put to the test, because, based on ScreenPlays’ interviews with some of the communities bidding for Google, many potential customers are unwilling to pay more than $100 for 1 gbps.
“As both a consumer and a business, I feel pricing comparable to Verizon FiOS or Comcast would be reasonable for both the physical connection into the home and Internet access,” says John Wojcik, who’s involved with Jersey City, N.J.’s efforts to attract Google. “I would hope that pricing could be structured so that residents would pay a cost proportional to their income so that it will be deployed universally.”
$1,000 per GPON Sub?
Of course, the cost of building and operating a super-fast network – whether it’s 100 mbps, 1 gbps or somewhere in between – is another major influence on the retail price. Those costs vary by technology.
For example, scratch-building a Gigabit Passive Optical Network (GPON) costs about $1,000 per subscriber, while DSL implemented on an existing copper-pair network costs about $300, according to Tellabs. When wiring a neighborhood for FiOS, Verizon currently has a target of about $750 per home.
“Actually connecting a home to the network when a consumer orders the service is about $600,” says John Schommer, Verizon’s director of consumer broadband product development. “Much of our cost is associated with labor surrounding actual placement of the fiber.”
Some analysts say Tellabs’ $1,000 estimate is in line with what they’re hearing from operators and other vendors.
“The assumptions behind this GPON price focus on an aerial deployment in an urban environment,” says Ken Berniklau, a Tellabs staff product manager. “We’ve seen competitive pressure pushing down capital prices, while technology adaptation proliferates this market.
In addition, he says, “deployment efficiencies are driving down operational expenses. We’ve seen offers for discounted fees in exchange for consumers trenching their own property for fiber drops. These unique and creative models certainly reduce the price to connect subscribers while increasing time to market with new services.”
For operators serving rural communities, there are also concerns about how the FCC’s National Broadband Plan will affect sources of funding for super-fast networks. For example, the plan proposes transitioning Universal Service Fund mechanisms from voice to broadband over several years. Some operators are concerned about a drop-off in funding during that transition.
“Unless that’s done very carefully, we could have a circumstance where [there’s a] lack of funding to maintain our networks and pay our debt,” says the NTCA’s Mitchell. “A lot of these networks are built on credit from the Rural Telephone Finance Cooperative and Rural Utilities Service based on 20-year loans. If we can’t pay that debt back timely and sufficiently, you can go out of business.”
Cart Before the Horse?
Today, only about 11 percent of consumers say they’re willing to pay more than $40-$45 per month for faster service, according to In-Stat’s periodic surveys. The low take rates for today’s 50 mbps and 100 mbps fiber and DOCSIS 3.0 services confirm those findings.
“Time Warner Cable in New York City has something like 400,000 to 500,000 cable modem subscribers, and their DOCSIS 3.0 50 mbps service has been available for maybe six months,” says In-Stat’s Paxton. “They’ve had 2,000 people sign up.”
Operators that currently offer 50 mbps or faster services say there’s currently no killer app driving upgrades.
“Some customers are high users of peer-to-peer services, [and] some are early adopters of over-the-top video,” says Verizon’s Schommer. “Many simply use the higher bandwidth for work at home.”
Why aren’t more consumers upgrading? It’s not an academic question. Instead, it goes to the heart of the main assumption behind the FCC and Google proposals that many, if not most, people want super-fast broadband but can’t get it. The paucity of 50 mbps and 100 mbps subscribers indicates there are price barriers to adoption which an aggressive built-out strategy might not be able to overcome.
The outlook is further muddied by the fact that for some consumers the choice of broadband provider depends more on the other components in a triple- or quadruple-play bundle.
“Verizon FiOS Internet is almost always sold in bundles; therefore pricing is more complex,” Schommer says. “Some customers view HD content in a bundle as their decision anchor, while others link choice to broadband speed and still others base it on voice service integration.
“It’s therefore difficult to assess the demand for higher speeds. However, Verizon doesn’t currently see customers clamoring to pay higher rates for ultra-high speeds.”
Besides price, the other factor holding back adoption is that the vast majority of consumers aren’t using apps and services – such as HD movie downloads – that require more than the 5-10 mbps they currently have. So the next question is whether they will be by the time the FCC and Google proposals become reality.
“This probably is a product that’s not ready for the residential world quite yet,” says Tom Larsen, vice president of legal and public affairs at Mediacom, which offers 50 mbps and 105 mbps consumer and business services in several markets. “Our take rates are going to be higher, at least for the first few years, in the business sector. By and large, the demand is coming from the small business owner. I just don’t see the average household needing this much bandwidth today.”
Others liken the situation to the initial migration from dial-up to broadband, which seemed like expensive overkill until apps and services such as VoIP, Hulu and YouTube made it justifiable.
“People would say the same thing: ‘I’m paying $9 for Internet. Why would I pay $60?'” says Rich Swier, Jr., co-founder of The HuB Sarasota, a tech and creative incubator that worked with its city on a Google proposal. “It is chicken and egg. Many people don’t understand the value because there’s no value to be had yet.
“I think what Google is starting to do is seed the market and get people excited, [do] some test markets and then deploy a slew of applications that would demonstrate the power of having it.”
Other companies are seeding the market, too. One example is the growing selection of TVs with built-in broadband capabilities that make it easier to watch over-the-top (OTT) video.
As the installed base of those sets grows, so does the likelihood that more OTT content providers will increase their selection and video quality. A household with one or more TVs tuned to OTT video could strain its broadband connection to the point that upgrading to a 100 mbps or 1 gbps seems justifiable, especially if the OTT selection is good enough that they can drop the video portion of their bundle and apply that savings toward faster broadband.
“At some point beyond five years, I expect the major consolidators of sports programming – whether it’s the NFL, NBA, MLB or the NCAA – will cut out the middle man and say, ‘Our programming is for sale on the Web,'” says former commissioner Furchtgott-Roth. “And if you want the highest quality video, you’re probably going to need at least 20 mbps to enjoy it.”
Network technology comes into play here, too, in terms of how quickly and cost-effectively an operator can upgrade to meet the bandwidth demands of emerging apps – or to counter a rival’s courtship of that new market. For example, if an MSO has already deployed DOCSIS 3.0 and digital terminal adapters, then it’s a matter of bonding together enough channels. For a telco with an extensive fiber plant, it’s a matter of swapping out electronics to shovel more bits up and down the glass.
None of those upgrades is cheap, but they’re incremental compared to, say, replacing copper with fiber.
“The question becomes, do you deliver speed and consistency prior to the demanding applications, or build applications that drive the need for more bandwidth?” says Ken McMahon, vice president and general manager for CenturyLink’s North Missouri business unit.
In that respect, operators are in the same position as their customers: When and how much they’re willing to spend on upgrades depends on whether they see value in the extra speed. (For a look at why Mediacom thinks now is the right time, see “Mediacom HSD Hits 105 Mbps in Aggressive DOCSIS 3.0 Push” in the February issue of ScreenPlays.)
Boost Upstream First?
Some business users – including those who are helping their communities court Google – say they’re generally satisfied with today’s downstream speeds. Take Sarasota, where businesses can get 50 mbps for about $250 from Comcast.
“It’s very reasonable for what we use it for,” Swier says.
Swier and his associates are more concerned about upstream, which struggles to handle, for example, 5 gigabyte video files that his firm sends to clients.
“Upstream is actually the most important thing,” Swier says. “50 mbps [downstream] is pretty good. We don’t feel the hurt. But when we start uploading something, we’re banging our head against the wall.”
Swier’s experience suggests that among businesses, the killer app for 100 mbps or 1 gbps networks might be in the upstream rather than the downstream. The same might apply to the residential market, especially as consumers upload more and more video.
“Verizon has found that, with the rise of community-content sharing applications, more customers are focusing on the upstream capability and upstream speeds,” Schommer says. “As such, Verizon’s mid- and upper-bundle tiers feature symmetric speeds (25/25 and 35/35), which are not being matched by any other consumer service providers in the U.S. market today.”
A Google-Verizon Partnership?
Some of the communities courting Google believe that it would consider partnering with one or more local service providers instead of building its own FTTH network.
“I think they’re going to be very aggressive working with telco providers,” Swier says. “If I were Verizon right now, I’d be the first one at their door saying, ‘How can we help you do this?’ because they’re going to sell a lot more broadband using Google than they would with their existing strategy.”
In one scenario, Google could resell Verizon’s FiOS, which currently delivers only 50 mbps in Sarasota and elsewhere but can support far more.
“Late last year, using a standard FiOS fiber optic connection to a real customer, we fed a 10 gbps feed over the network with no problems,” says Jim Smith, Verizon’s director of media relations. “The PON we built makes is so that all you need to do is adjust the hardware on each end and you can [increase] the bandwidth. We could make it happen over the FiOS network without much trouble, should a market for it develop.”
But would Verizon consider partnering with Google? Smith didn’t answer that question, nor did a March 30 Wall Street Journal op-ed co-written by the two companies’ CEOs. If they’re receptive to it, one possibility is for Google to make a big investment in Verizon – or another operator – just as Microsoft invested $1 billion in Comcast in 1997 to fund the MSO’s broadband expansion. The strategy then and now: More broadband means a bigger market for services running over that pipe.
Which comes back to the pricing issue. If Google believes that 1 gbps expands the market for its advertising, video and other services, then it might be able to make a business case for offering broadband as a loss leader, pricing around the same level as rivals’ 50 mbps services.
“I really don’t believe they think there’s profit in the bandwidth,” Swier says. “I think they need the bandwidth to make their advertising network run smoother.”
For both Google and any partner, profit also hinges on the cost of the upgrades necessary to hit 1 gbps. The same goes for an operator that wants to upgrade to compete with Google.
“If the market demands speeds above our current capabilities, Verizon can easily upgrade the network electronics – the hardware on either end of the fiber – with a somewhat minimal investment,” Schommer says.
Google’s publicly available information doesn’t even hint at whether it would consider partnering rather than building. Some analysts believe that smaller operators might be more receptive to a Google partnership, assuming that they can roll out a 1 gbps network in time to meet Google’s schedule.
“I think they’d have to go with a CLEC or some type of smaller service provider,” says In-Stat’s Paxton, who recently returned from a FiOS conference. “I don’t get the feeling that they’d even be willing to partner with them, especially because Google and the incumbents are so far apart on things like Net Neutrality. I would be very surprised if the partner turned out to be one of the major incumbents.”
Much Ado about Something
The bigger question is, is Google serious about offering 1 gbps services? It’s easy to be skeptical, based on past experience.
“Google has made a lot of noise in a lot of different market segments outside of their paid search bailiwick, and historically they haven’t come through on a lot of those, [such as] the spectrum auction and municipal broadband,” Paxton says. “I’m not 100 percent convinced that this is going to happen at all.”
Even if that turns out to be the case, some potential suitors argue that Google still has helped make next-gen broadband a high-profile issue at both the local and national levels.
“If Google hadn’t done what they did, and I took a sign down the road saying, ‘Hey everybody, we need a Gig to the home,’ they’d think I was crazy, and nobody would get behind me and march,” says HuB’s Swier. “This was an opportunity for us to capture that momentum and make broadband one of the top priorities in Sarasota.”