I have a confession to make: I’m a satisfied customer of Elon Musk’s SpaceX Starlink internet. With that in mind, indulge me in an audacious proposal for connecting all Candians with high-speed broadband.

First, some background: Last summer, our family relocated to the Niagara region to be closer to relatives helping us manage the load of lockdowns and remote learning. With two parents working from home and a son always online, broadband internet was our lifeline. 

After six months of using the local internet service provider we needed another option. Our internet plan––capped at a download speed of 25 megabits per second and an upload speed of 10 mbps––would strain or break under the demands of daily simultaneous Zoom meetings. 

Enter Starlink, a constellation of satellites beaming high-speed internet to trial customers in the U.S., Canada and the U.K. SpaceX is hoping its satellite internet business will help cover the costs of its missions to Mars. It unveiled its early access program six months ago, with more than 10,000 users signing up in the first few months.

Out of frustration with our existing provider, I entered my postal code into the SpaceX site in early February and, lo and behold, my location was remote enough to put me among the rural internet users Starlink is currently targeting for its product. 

Several weeks and $649 later, my router and heavily subsidized dish terminal (SpaceX has said it’s absorbing about two-thirds of the cost of each unit) arrived. Installation took just a few minutes, and Starlink’s service started delivering speeds consistently over 70 mbps and routinely well over 120 mbps. That’s well below the 200-plus mbps available with existing fibe-to-home networks, but a big improvement from the roughly 18 mbps I was getting—and with a commitment from the company that speeds will continue to improve. 

Starlink acknowledges its product is still in beta, which means it does have regular service interruptions as it works to optimize the network. But in my experience, drop-offs have occurred at about the same rate as they were with my local provider, and it’s gotten more reliable as the company continues to launch satellites into low-earth orbit.

What’s really made me a convert, though, is Starlink’s customer service. The company gives subscribers a courtesy notice if it’s expecting an outage, and when my router stopped working, they shipped me a new one, no questions asked. If I were an existing rural internet provider, I’d be very nervous. (This is where I should point out that I have not received any money from Starlink and I have never disclosed to the company that I am a journalist.)

There are many caveats, of course. The company currently serves just thousands of customers. SpaceX COO and president Gwynne Shotwell has said that Starlink wants to serve roughly 60 million people in every rural household in the United States; how will the company manage that scale? Should users expect data caps? Tiered pricing? 

These aren’t small questions, but if discussion groups and chat forums are any indication, the early reviews are positive. The unofficial SpaceX Canada Facebook group has more than 7,000 followers who excitedly share screenshots comparing Starlink internet speed test results. 

One company that routinely draws the ire of the Starlink forum is Markham, Ont.-based rural internet provider Xplornet. The company also uses satellites, but they’re at a higher fixed orbit than Starlink’s so can’t offer the same speed or reliability. For example, if there’s a big rain cloud blocking the dish’s view to the satellite, Xplornet customers may lose their connection, while a Starlink terminal hums along with no trouble. (Xplornet did not respond to multiple requests for an interview). 

Xplornet relies on Canadian satellites operated by Telesat. And this is where things get really interesting. Telesat, whose largest shareholders include the Public Sector Pension Investment Board, has its own plans to enter low-earth orbit, recently receiving a $400-million loan from the province of Quebec and a $600-million contract from the federal government to be a preferred customer. 

SpaceX, whose funders include the Ontario Teachers’ Pension Plan, has projected its program will cost US$10 billion, dramatically more than what Telesat says it needs to get its low-earth orbit Lightspeed satellites up and running by early 2023. But while SpaceX owns its entire supply chain––making history last week by launching both a reused rocket and a reused capsule–– Telesat is relying on Blue Origin––owned by Amazon founder Jeff Bezos––to put its satellites into orbit.

Space is a competitive, risky business and the only safe bet is that most of today’s new entrants will fail. Nobody to date has been able to launch a real constellation of satellites without going bankrupt. Two decades ago, major satellite providers Iridium, Globalstar, Teledesic, and Skybridge all ran out of cash in pursuit of the stars. OneWeb, another low-earth orbit satellite company, was rescued last November by the U.K. government and India’s Bharti Global. One insider I spoke with for this column told me that while he was once skeptical that Starlink could rise above the carnage, he’s now convinced––with the company’s head-start in launching satellites, its reusable rocket technology and the recurring revenue it will get from subscribers to its service–– that SpaceX is the one to beat.

Back here on earth, connecting Canadians is an imperative—one could argue it’s becoming a human right. More than 12 per cent of Canadians don’t have access to high-speed internet and it’s available in less than half of all rural communities. This year’s federal budget committed another $1 billion to the Universal Broadband Fund (UBF). The now-$2.75-billion effort to expand “access to fast and reliable high-speed internet” aims to reach 98 per cent of Canadians by 2026, and the remaining two per cent by 2030. To do this, the country relies on a hodgepodge of public-private partnerships with Canadian telecommunications companies to build out expensive infrastructure across the country. 

Meanwhile, there are already some 1,200 Starlink satellites in orbit with more coming online every month. This includes 10 satellites launched this January into polar orbits, which will bring reliable service into some of the most isolated regions.

So here’s a thought exercise: What if Canada’s federal government simply handed the $2.75 billion in the UBF to SpaceX in exchange for a huge number of its dishes, on favourable terms? Could a deal of this magnitude be negotiated to include a Canadian SpaceX facility? More Canadian astronauts in the SpaceX program? Even a share of revenue or a stake in the company? Starlink would reach 4.2 million Canadian households as soon as they were able to produce enough dishes––which its COO says could be in as little as three years. It’s already taking pre-orders for mass shipments expected at the end of next year. (Of course, at $129/month, the recurring cost would not be affordable to all, but the government could subsidize those expenses where necessary). 

To-be-clear, this isn’t an argument against creating a Canadian champion in the sector; Telesat has decades of experience and isn’t even competing directly with Starlink for customers. But if the primary goal is connecting all Canadians as quickly and effectively as possible, and homegrown broadband connectivity is still at least a decade away, wouldn’t it be more prudent to bet on Elon Musk? It’s a liberating exercise to contrast two policy paths: one that is almost guaranteed to work in the next half-decade versus another that requires pouring billions into domestic public-private infrastructure projects that could get more Canadians online by the end of the decade, that might help create a domestic champion, but that could also end up creating another boondoggle. Which path would you choose?

I know this could never happen. There are too many competing industrial interests—the entire Canadian aerospace and telecommunications sector would be up in arms. This back-of-the-napkin exercise tests my own inclinations towards economic nationalism and job creation, never mind the potential issues around national security and protecting Canadians data and privacy. But some of those most in need have already made their choice. Last September, the Federation of Northern Ontario Municipalities championed the arrival of Starlink, telling the CRTC it was the “best option” for bringing high-speed internet to its 110 municipalities and towns in northeastern Ontario. 

The original Canadarm cost $108 million to build in 1975. That works out to roughly $540 million in today’s dollars. For the price of five-and-half-Canadarms every Canadian could have plug-and-play high-speed internet. A fully connected Canada in the next 36-months—now that’s a moonshot.