UpHere Logo

WEB EXTRA: Nunavut’s internet is slow, expensive and unreliable

WEB EXTRA: Nunavut’s internet is slow, expensive and unreliable

SSi Micro’s Jeff Philipp thinks a monopoly could help fix these problems
By Herb Mathisen
Jul 27
2016

“What we have in Nunavut is market failure. Private sector providers are competing for subsidies and not for customers.” That was Oana Spinu, executive director of the Nunavut Broadband Development Corporation, summarizing the situation in the territory at a CRTC hearing this spring, held to look at the growing gap between northern and southern internet service in Canada.

She wasn’t wrong. But there’s a reason why subsidies play such a huge role in Northern telecoms. Last year, the feds awarded SSi Micro $35 million to improve broadband internet in Nunavut over the next three years. On top of that, SSi Micro, through its Qiniq subsidiary, promised to invest $40 million of its own money: $25 million in infrastructure, and $15 million on buying capacity. Nunavut depends solely on satellite for internet, so most of the money Qiniq gets from the feds will go directly to Telesat, which owns internet-providing satellites in space.

Most of the Nunavut’s communities are so small that competition can’t exist, and subsidies are necessary to keep internet providers—and the internet—running in these places. But as bandwidth usage and speeds increase, so does the amount of money that goes to Telesat. And with Nunavut consumers already paying some of the most expensive internet bills in the world, it’s hard to imagine users being able to pony up any more for the essential service.

So what gives? Well, Jeff Philipp has an idea. Right now, Qiniq has redundant infrastructure (that’s 120-foot towers, dual satellite dishes looking at two satellites, back-up generators) in Nunavut’s ten largest communities. He wants to bring this into the 15 smallest communities—many with populations less than 1,000 people—to bring that same relative redundancy and capacity that the larger centres experience.

But there’s a problem with that plan: if a competitor decides to enter the fray in the money-making centres like Iqaluit, Cambridge Bay and Rankin Inlet, and starts taking away his customers, Qiniq can no longer afford to provide a service to all communities. The only way it can afford to operate in those 15 small communities, he says, is to subsidize them with profits from the larger ones.

How do you get around that? With a monopoly. Seriously. Jeff Philipp, the NWT’s lion of competition, is proposing a monopoly.

It costs sealifts of money to put up communications towers, satellite dishes, and the buildings to support them in Nunavut. Philipp believes that’s a barrier to competition. So he’s suggesting government put out a long-term tender for a company to build redundant backbone infrastructure in each community, and then have companies bid on the provision of retail service in the towns, with bandwidth sold to them at a regulated rate. Philipp concedes he’d probably get that backbone contract (he’s already got so much invested in Nunavut), but it would allow customers to shop around with different providers in town, while the companies pay the same rates on the backbone infrastructure. Such competition, Philipp says, would hopefully bring down prices.

(The following are lightly edited transcripts of an interview with Philipp from December 2015)

On how to bring competition to Nunavut:

Jeff Philipp: Are you going to expect Bell, Rogers, Telus and SSi to each build a 100-foot tower in Grise Fiord? Ultimately, you need shared infrastructure.

If we want competition, we need a nice facility for us to compete in. I will offer [other companies] the same rates I offer myself, but they have to agree to go into every community. They can’t just go into the five largest—they will go into all 25 and if they agree to go into all 25 and lose money in the same 15 I’m losing money in, I will sell them service in all 25 competitively. But if they want me to let them into the top five…  [unfriendly four-letter word] you. They can go build their own [infrastructure] in those five because the reality is that’s never going to work.

This is not in my best interest. Realize that if I was really smart, I would just shut up because I stand to win any contract from now on, unless somebody like Bell really, really, really wanted to lose a lot of money and come after me. Frankly it would be not in their best interest because I could just take the reverse tactic. Let them have the contract and then just focus on three or four big communities. The problem is that strategy [screws] the North. It does not benefit the North. It wins for whichever competitor gets the big markets, but unless you’re willing to cross-subsidize those smallest communities, somebody in the community is going to lose if it gets to this competitive state.

It’s not that a monopoly’s a bad thing. A monopoly at the right level is not a bad thing.

On bringing fibre-optic cable internet to Nunavut:

Jeff Philipp: Long before we agree to invest hundreds of millions of dollars to bring fibre to the very first spot, [Nunavut] will run out of satellite capacity in space. We will run out in 24 months, if not sooner. We will not have fibre in these markets in 24 months, so what are we going to do in 24 months?

What happens when that one fibre that the Government of Nunavut is using into Iqaluit in January breaks under the ice? They’re going to go for months with no government services? Or they’re going to immediately move onto an overloaded satellite network that ran out of capacity three years earlier and expect what will be thousands of megabits to come through a satellite capable of delivering hundreds?

We need a satellite because in two years the capacity in space will be out. Telesat has one launching. If you haven’t already booked space and they sell it to somebody else, then what do you do? Then you’re waiting three years to launch one. What do you do about education in those three years? And health care?

Really, if we’re going to be using satellite in 23 of 25 communities for the next 20 years because fibre’s never going to Resolute Bay or wherever, then I guess we’d better have redundancy, and that is two [satellites]. And hey, by the way, they provide redundancy for the fibre, so when fibre comes into Iqaluit, all the satellite that was being used for Iqaluit, it’s going to get shifted over to some other communities, so great for the 15-year-life of the satellite.

If redundant fibre is everywhere, don’t launch another [satellite]. But launch one in 2018. Launch one in 2020. By 2035, if we have a ring of fibre, don’t launch another one. But until 2035, we [should] have redundant satellite capacity with 24,000 megabits per bird, which means we truly could deliver throughput where the utility of broadband becomes like the utility of power, where you don’t worry about whether you turn your lights on in the kitchen.

On government commitments to boost internet speeds:

Jeff Philipp: I mean, you go from 3.5 megabit to six megabit, that’s doubling [capacity], that’s $35 million to $70 million without increasing the number of subscribers or increasing the amount of usage they get—that’s just doubling the speed. And speed—that’s the other ludicrous part.

It’s a double-whammy. You promise a guy 5 mbps down, 1 mbps up, but when he can’t get it all the time, he’s pissed off, right? It doesn’t matter that his video runs perfectly all the time, he’s pissed off about that [proverbial] car you sold him. The last car, he had the speedometer at 100 and he could get it to 100. Every once in a while, when the road was straight enough and he had enough bandwidth, he’d hammer it. The new car, all you did was give him the same car but you painted 200 on the speedometer and he knows he can never get there. It never gets there. It just pisses him off. You would have been better off to say, ‘No, I didn’t change the speedometer, but I gave you four times as much gas. You can drive and drive and drive and you’ll never stop, because we added more lanes to the highway.’ That’s what I’ve been trying to convince the federal government—don’t promise more speed just to keep up with the Joneses, promise the quality of service that people want from the goddamned thing, otherwise we’re going to continually fail because we’re putting a Band-Aid on it. We’re not actually fixing the problem, which is: Netflix costs about $25 a movie, show, episode. A gig is $24. If you’ve got high-speed, if you actually had five-mbps-down, all that Netflix does is increase the quality level to suck up more usage, so what cost you $20 is going to cost you $40 for the same one-hour episode. So [you’d be better off to] lower the speed to two mbps and increase the usage and add other services.

On government investment to get there compared to the south:

Jeff Philipp: In Nunavut, the next subsidy … [Let’s say,] if it’s for four years, if $35 million covered two [years] even at the same level of service, they’re talking $70 million for four [years]. Now we’re talking $15 million a year for broadband at a sub-par level for Nunavut. Apply the standard rate curve that has happened in the south for growth and usage. And realize that growth and usage in the south—where you’ve got a fibre pipe that is like a ravine and you only get a little trickle at the bottom—you don’t have to build a new ravine. In satellite, when you give them more, you’ve got to put another satellite up because it’s a pipe that’s limited in size.

So when you really look at the ramp up, you realize that ten years from now, if we truly want to deliver a proper service, the government’s going to [need to put] in $35 million a year at a minimum—and that probably won’t even come close to it. That’s just for subsidizing the residential consumer—not education, not health care. The government can’t afford what they need.

On why broadband is so expensive:

Jeff Philipp: Because it’s like we’re buying M&Ms at retail at the counter versus buying the factory. [Imagine the satellite company is] a bus driver—what you do is you buy a payload on the bus.