High-speed rail has proliferated around the world: some 65,000 kilometers of track have been built so far. So why shouldn't Canada join the club and build high-speed rail as well? The federal government of Canada would like to join the club, and in February 2025 announced to build ALTO, a high-speed rail line connecting Toronto, Ottawa, Montreal, and Quebec City. The price tag? Between $60-$90 billion. When would it be operating fully? No sooner than 2044.

The proponents of the project have not fully worked through the economics of it. High-speed rail of the type envisioned is the wrong solution for Canada. It is too expensive, and with typical cost overruns for such projects could quickly turn into the type of boondoggle that California has experienced with its California High-Speed Rail. Initially approved in 2008 with an estimated US$33 billion price tag, has faced severe cost overruns, with recent estimates projecting the total cost to exceed US$231 billion. It also looks as if the original route, from San Francisco to Los Angeles, will not be completed as originally planned. There are many reasons why the project got stuck. How California's Bullet Train Went Off the Rails was described aptly in a New York Times article in October 2022.

And if you think California's high-speed rail is just an outlier, look to the United Kingdom and their HS2 rail project for 230 kilometers of track between London and Birmingham. The cost of that project is soaring to £103 billion, $191 billion in Canadian money, or $830 million per kilometre of track. At $90 billion for 1,000 kilometres of track, Canada's proposed high-speed rail would be just 11% of the cost of the British HS2 per kilometre. I think we can all be quite sure that even the high-end $90 billion estimate will be surpassed in Canada, if Great Britain is any indicator of building high-speed rail in a highly-developed Western country. Of course, Britain is more densely populated than Canada, so costs will be higher because of that. HS2 is also not being built as planned. Because of cost overruns, phase 2 of the project (Birminghan to Manchester and Birmingham to Leeds) was cancelled in stages in 2021 and 2023. And there is no HS1 either; the 110 km high-speed link from London to the Channel Tunnel was cancelled in 2014.

An April 2026 editorial in the Globe and Mail described Canada's ALTO project as The costly fantasy of high-speed rail. The editorial got the economics right. High-speed rail makes little sense in Canada. Canada needs better rail, but not high-speed rail (HSR). Let's dive a little deeper into the economics of rail networks.

To begin with, the emphasis is on rail networks. A simple line, even if fast, can only serve a very distinct subset of customers—those that live along the line. A rail network, on the other hand, connects many cities in a criss-cross fashion. By forming a network, the system benefits from network externalities as more cities are connected, and some cities serve as central hubs. Look at how rail is organized in Europe, or China or India, and it becomes immediately clear why passenger volume will be much diminished if the rail system does not form a proper network.

What makes rail attractive to customers are four things: (1) high frequency so that you don't have to wait too long for the next train; (2) high connectivity so that you can get from home to home via rail rather than a lot of multi-modal transportation; (3) high reliability so that you can be sure to get to your destination when you need to; and (4) affordable rates. If the service is too costly, too infrequent, too cumbersome, or too unreliable, it will fail.

High-speed rail in Canada is almost from the start stuck in a doom loop. It will be too expensive unless heavily subsidized, due to the high construction cost. To economize, service will not be frequent enough. If trains don't run at least every hour, people won't go for it in large numbers. If the high-speed train stations are not well connected to local subways and regional commuter trains, people will find it inconvenient. With insufficient demand, service erodes as costs need to be cut, and that makes high-speed rail enter into the doom loop of failing or requiring continued subsidies to survive. There is a better way.

The extra cost of going from modestly-fast rail (about 150-200 km/h) to high-speed rail (250-350 km/h) is disproportionate. The faster the service, the straighter and flatter the tracks have to be, which means tunneling, building viaducts, and flattening ground through cut-and-fill grading. These interventions make such a project very expensive in many places. In addition to flatter gradients, high-speed rail needs wider turn radii. Smoother curves also run into land acquisition and preparation issues. To make things yet more complicated, it is also necessary to eliminate level crossings (i.e., complete grade seperation), install robust fencing to prevent collisions, weld and fuse the tracks, install advanced signalling systems, and provide improved tensioning for the catenaries to make sure the train's pantograph doesn't lose contact with the wire at high speeds (in order to prevent arcing and damage). All of that is what makes high-speed rail so much more expensive.

Modestly-fast and frequent rail will provide much better economics: more ridership because more routes can be built quicker, and more frequent service is more useful than the faster travel. High-speed rail can emerge from a well-functioning frequent-rail system when demand is already present, and can be grown on the most popular corridors.

Building high-speed rail without a supporting rail network and well-established demand is pure economic speculation. Anticipated agglomeration benefits are wishful thinking.

If the ALTO proposal goes ahead as announced, expect it to become a fiscal nightmare for Ottawa. Prime Minister Carney should take a hard new look at the proposal and give consideration to the flawed economics. Canada needs better rail, but a rail network needs to grow organically along existing population density. What is needed are new rail corridors that are not shared with the freight operators, who take precedence and use their infrastructure to run their freight trains. Passenger trains that use their tracks often have to wait to let slow freight trains pass. To make passenger rail work, Canada needs to build new passenger tracks alongside existing rail lines, and connect many medium-sized cities rather than merely the most populous cities.

How should Canada move forward? We do need better rail service, but which rail option is superior should be determined by an appropriate Cost-Benefit Analysis. A 2025 C.D. Howe Institute study found that high-speed rail can deliver higher benefits than conventional rail, but not in all scenarios. But when costs are taken into account, the net benefit turns negative. The high-speed rail best-case scenario is estimated as $27B. With a cost of $60B-$90B, the economics case for high-speed rail collapses. The economic case for high-frequency rail looks eminently stronger.

Before the federal government made high-speed rail a priority, it focused on high-frequency rail, ostentatiously because faster travel times were more attractive to public opinion. Of course it is possible to run high-speed rail at high frequency too, but the difference is in the construction cost. Estimated costs per kilometer of track about $50m-$120m for high-speed rail (250-350 km/h), and $10m-$30m for conventional upgraded rail (160-200 km/h).

I frequently hear the question: "if China can do it, why can't we?". The answer is threefold. China's enormously high savings rate makes it possible to over-invest into infrastructure. So part 1: cheap money. China's communist government can also more easily acquire and develop land than is possible in democratic Western countries. So part 2: different laws. And lastly, China's cost structure for construction is more advantageous than elswehere. So part 3: lower costs. It would be nice if we could build many thousand kilometres of high-speed track in Canada at China's cost level, but we simply can't. Canada's economy is simply different from China's. Wishful thinking does not solve our problems. What does move us forward is carefully-considered investment into infrastructure that serves our needs, at prices that we can afford. Let's avoid repeating the mistakes of the United States and the United Kingdom when improving our passenger rail system.