Great article. There are a few other factors to consider that I can think of, where AVs and Alto are concerned:
—Will AV trucks take freight share from rail, thus potentially freeing up rail lines for faster passenger service that could be a cheaper alternative to ALTO?
—Will AVs help solve the first-mile last mile problem, thus making people more likely to ride ALTO..or alternatively, more likely to fly instead of taking ALTO?
— Will AVs push up the price of fuel, or perhaps push down the price of fuel, which could in turn push up or down demand for ALTO?
…there’s a lot to consider. It’s hard to predict, though I share the concerns you raised in this article.
AV trucks won't take any more freight from rail than driver trucks take now, unless the driverless trucks are cheaper than rail currently is by comparison. Driver trucks already solve the first- and last- mile problem better than trains do, and are usually faster where modern highways exist. Where trains make it up is being cheaper per ton-mile than trucks are. Great if time isn't crucially of the essence. (If it is, like with live lobsters, you use air freight.). Besides, I thought the mantra was that freight was supposed to be moved off of trucks and onto trains, by subsidy if necessary. You know, that climate thing? Why would you want to move the other way when trucks emit more CO2 per ton-mile than trains do?
There is no case on God's green earth that would cause a freight railway to let its lucrative container business -- "Laser Intermodal Service" CN calls it -- slip away to trucks (AV or not) just so it could operate money-losing slow passenger trains on its now under-used tracks. Why would anyone want to compete to provide a service for "budget-conscious travelers" that loses money? That's a death spiral. The freight railways are out of the passenger business forever. Highways and airplanes killed it. It's dead. If the government wants to subsidize passenger rail on its own tracks, "Fine," say the railways. "Just don't ask us to have anything to do with it. Particularly don't ask us to give priority to your trains on tracks we own and pay to maintain with our freight profits. Those profits, by the way, come from promising our customers that our trains carrying their goods will arrive more or less on time, not delayed by taking the hole every few minutes to wait for a passenger train to go by."
I was unclear, I meant solving the first mile last mile problem for passenger rail, not for trucking.
So you don’t think there is any chance that AVs will have any impact on rail’s freight share…
I would have thought that there’s at least some small chance that more freight will move to AVs, and more passengers will use AVs to travel the first/last mile to train stations, to ride passenger trains that, even if not high speed rail, can still travel at much faster speeds than AVs.
No, I don't think AVs will capture any of the high-volume freight from the railways that trucks don't already have. Once the first-mile-last-mile problem fades out as the total distance increases, the railways will win on price. Remember a load of automobiles assembled in Oakville might still get loaded on a train -- they in fact do -- because they aren't just going to Toronto. The trainload gets reclassified in MacMillan Yard (no relation!) in Concord to be put into other trains going to customers in Montreal, Halifax, Winnipeg, or into New York or Michigan or Chicago. The cargo never needs to be unloaded until the final rail destination. The train cars just get reorganized as necessary that each customer gets his stuff. AV trucks won't get any of that freight. And even if they did, CN's not going to bring back its own passenger trains. If train density took a nose dive, it might charge VIA *more* (not less) to rent its tracks because its dispatchers would have to stop the passenger trains in sidings a bit less often. VIA should be willing to pay higher rent for faster routing.
Where I think passenger train advocates (and HSR advocates in particular) are going wrong is that you want passenger trains for their own sake because you like passenger trains. (So do I, as a matter of fact.) You think that policy makers ought to come up with transportation solutions that give a prominent place to passenger trains. (Which I don't.) The reality (I think) is that only a few romantic enthusiasts actually *want* to take an intercity passenger train. The rest will settle for a passenger train because it's cheaper than flying and less grotty than the bus (if there are even any intercity buses anymore) and can be more pleasant than driving if it's not like four hours late. If it becomes almost as fast as a plane, then the train passengers should be willing to pay almost as much for a ticket. But they won't. They will feel entitled that their good social behaviour in taking the train should be rewarded with a larger subsidy to keep ticket prices artificially low, "to get more people out of airplanes and onto trains!" If ALTO ticket prices are double VIA (as I've heard suggested), most people will just give up and drive. Montreal's not worth that much to the casual traveler or McGill student that they need to get there two hours quicker.
That's what will kill ALTO, as the author has explained. To make the train pay, ticket prices have to be high, but to get stingy entitled Canadians to ride it, tickets have to be cheap. You could force business travelers to use the longer segments that maybe someday will go to Toronto and Quebec City by curtailing the number of air flights permitted between Toronto and Ottawa/Montreal, or Quebec City and Ottawa, but Air Canada's not going to just sit there and take that arbitrary attempt by the government to rig the market in favour of its pet Crown Corporation.
AV also wins on time if the train departures are infrequent. But travellers may prefer the train because it is smooth enough to get the laptops out. Time in an AV will be noisy and jittery.
There’s also a case for slowing down. I regularly take a painful early morning flight from Montreal to Toronto and look forward to the day I can take a “red eye” style AV over, sleeping in the car doing a smooth 80Km/h on the freeway so I can arrive in Toronto for 8am meetings after a solid 8hrs of sleep.
Not for that distance, unless you want to park the trains outside Toronto or Montreal for a few hours. In steam days, Toronto-Montreal night trains did take 10 hours or so to make the distance. Diesels shortened the run to where you would have left Toronto at 22:30 and arrived in Montreal at 04:00. What are you going to to in Central Station at 4 a.m.? Sit on your suitcase until the coffee shop opens? So they did park the trains on sidings so your arrival would be a reasonable hour to be getting out of bed. Some people can't sleep on trains, at all, no matter how comfy the sleeping berth is.
But sleepers mean the hotel costs of porters and changing linens. On the railway they'll be union labour, not the minimum wages that real hotels pay. Women traveling alone (which they didn't used to do) are going to want lockable compartments that they can lock from the outside to keep predators out while they are out and about, as well as from the inside while they're in there. Sleeping car accommodation was, and is, very expensive. I think it's priced to keep the riff-raff, who get drunk and rowdy in coach on night runs, from buying sleeper class.
I got thinking about cost functions when you realize that ALTO will have to leave from and arrive at some terminal (and intermediate stations) that people don't necessarily live near anymore. (When Toronto's Union Station was opened in 1927, -- building started in 1914 slowed down by the War --there were horse-drawn streetcars and Ford's Model A went on sale that same year. The Subway --to Eglinton, the edge of the known universe! -- was 27 years in the future. Train travel seemed space-age, burning the fuel of the future -- coal -- a step up from wood. (In early days, if the train ran out of wood, able-bodied passengers had to get out and reload the tender from strategically placed woodpiles, and then cut down fresh trees so the wood could dry out for the next train that ran into trouble.)
But my main point is different. The reason why over-the-road (or through-the-air) speed is decisive in costing out a transport system is labour cost per passenger-mile. The nuisance of getting from Woodbridge or Uxbridge to Union (to board a train that might go through Uxbridge but doesn't stop there!) is invisible to ALTO because it doesn't bear that cost. Rather, if it makes the trip to Montreal in 2.5 hours instead of six it has less than half the labour cost to pay the train crews for the trip. This is, of course, why airlines can afford to fly: 5 hours of pilot and stewardess wages to get you YYZ - YVR instead of four days of train crew wages, especially in sleeping and dining cars. (And the explosion in airplane technology means that one plane can carry more people with fewer staff than a passenger train could. Especially since if your flight is only five hours you can put up with hunger and discomfort. Four days and nights in coach on the train is an ordeal even if there is more legroom and if your drunk seatmate doesn't snore all night.) So people will expect to pay more for an air ticket for the same distance because it's so much faster.....and it's the very speed that makes it a profit-making proposition for the airline. (Then there is the lucrative "rail-cruise" market that "The Canadian" has become. Much slower than in the 1950s but rich foreign tourists don't mind especially when the dollar is cheap.)
So where does ALTO fit in? The faster they go, the less they have to charge to make back their labour costs per trip, and the more trips they can make in a day. But the fewer stations there are, and the farther people have to drive (and park downtown!) to board the train, the less desirable it is. Drive to YYZ or drive to Union? Hmmm. Yes, the airline also doesn't have to cost the time factor of you getting yourself to the airport and through security. It's sensitive only to how long it takes to get one planeload to the gate and push back out again to Montreal. It will be able to do this faster than ALTO will, just because the conveyance moves faster, no matter how long it takes an individual passenger to get to the gate. Door-to-door ALTO might be faster for downtown livers than Air Canada, but Air Canada will have a time-cost advantage that ALTO will have to meet in order to price its tickets competitively. (Now sure, pilots are more expensive per hour than train drivers, but air hostesses may be cheaper than "service managers" who replaced conductors on VIA trains 20 years ago, who are all members of various operating brotherhoods who know a thing or three about feather-bedding.)
Bottom line: ALTO will struggle to earn enough to justify the subsidy just because it has a time-cost disadvantage compared with flying, being slower gate-to-gate while the employees are on the clock. That it might be attractive on total trip time for *some* passengers (disappointingly few in our suburbanized country) doesn't change the fact that it will cost more staff time to move one passenger from Toronto to Montreal than it costs Air Canada. (Robot trains with no human staff on board -- just cameras to police bad behaviour -- would reduce that cost disadvantage, of course.)
You see what I'm doing here: engaging with this Substack vigorously enough to justify upgrading! :-)
This is *such* an excellent Substack. I would love to upgrade but I'm getting buried in subscriptions and I'll have to discontinue some before I bite your bullet.
Taken as a whole, I don’t think this commentary has a leg to stand on. Just my opinion.
Maybe commercial flight should be grounded and we should scrap the whole autonomous vehicle paradigm because flying cars will render both platforms pointless, useless, worthless.
But seriously…
The reason we have trains is because they can move vast quantities of goods and large numbers of people extremely efficiently and at relatively low cost. Electricity makes that task even more practical, allowing for increased efficiencies and providing even greater savings. Improved transit times mean upped average speeds. Accomplish that, and all of a sudden rail starts to look really attractive as an alternative to both driving and flying. High-speed rail, with its own dedicated, double-tracked right of way, with sidings and crossovers strategically located thereby permitting express trains to overtake the all-stops “cruisers” that serve the intermediate stations, this will not only allow for unencumbered operations, but make the overall service a seamless (or much more seamless) one, especially with the trains running city center to city center.
Furthermore, trains in the grand scheme of things have a considerably smaller carbon footprint, comparatively speaking. And railroading is the safest “automated” form of transportation going.
Meanwhile, in the U.S., recorded every year are upwards of 40,000 highway deaths. Autonomobility (driverless technology) may one day eliminate all human-error-caused roadway-based deaths, but it won’t end all roadway-derived crashes. Car crashes will still occur due to such inevitable factors as vehicle mechanical and/or electrical breakdowns and/or onboard computer glitches that will crop up from time to time, affecting motor vehicle operation and/or performance in some way, shape or form.
You seem to be pinning a lot of hope on the AV platform being the be all, end all. You, and a whole lot of other folks. I’m not as optimistic. The same kind of optimism I have about flying cars en masse ever getting off the ground.
Does the highway network (even with substantial upgrades) have the capacity for the additional vehicle traffic? From what I understand, passed a certain point, capacity upgrades to highways have diminishing returns. And the 401 is, arguably, passed that point already.
Is the assuption that AVs drive in a more efficient, collective, manner that reduces capacity constraints? Does it hold when it's AVs and human piloted vehicles operating together?
I don't know how to go about modelling it myself, so I'm just thinking out loud here and not trying to present this as definitive. But I'd think the highway networks would be a congested mess with most additional trips using the highways for the next 50-odd years.
In fairness to leaving out the capital cost of HSR, it's also assumed the highway will be toll free and it's publicly funded infrastructure the AVs operate on.
But really, HSR already has this competition - it's the existing VIA train. I have no idea how HSR could be expected to operate anywhere near the current trains cost, and I don't see how an HSR that costs a jet ticket could possibly have the volume of passengers needed.
It is disappointing, I could do so much for public transit with $60B in capital spending.
Not quite: everyone gets to use the publicly funded highway, not just the AV service company. The capital cost of HSR goes entirely to the exclusive use of the HSR corporation.
HSR for sure won't be able to operate anywhere near what the current train costs. It will be far more expensive per passenger mile to operate (not just build.) The higher speeds will require more propulsive energy to overcome air and rolling resistance. Electricity isn't free. The track and rolling stock will have to be maintained much more meticulously. You might need a pilot vehicle to go ahead of every train to make sure that vandals haven't fucked with the track and signals, as Idle No More periodically threatens to do.
Finally, ALTO will need to pay a squad of goons and thugs to intimidate the locals into acquiescence about the railway's plans to expropriate their land and cut off their roads to avoid building expensive grade separations.
Great article. There are a few other factors to consider that I can think of, where AVs and Alto are concerned:
—Will AV trucks take freight share from rail, thus potentially freeing up rail lines for faster passenger service that could be a cheaper alternative to ALTO?
—Will AVs help solve the first-mile last mile problem, thus making people more likely to ride ALTO..or alternatively, more likely to fly instead of taking ALTO?
— Will AVs push up the price of fuel, or perhaps push down the price of fuel, which could in turn push up or down demand for ALTO?
…there’s a lot to consider. It’s hard to predict, though I share the concerns you raised in this article.
AV trucks won't take any more freight from rail than driver trucks take now, unless the driverless trucks are cheaper than rail currently is by comparison. Driver trucks already solve the first- and last- mile problem better than trains do, and are usually faster where modern highways exist. Where trains make it up is being cheaper per ton-mile than trucks are. Great if time isn't crucially of the essence. (If it is, like with live lobsters, you use air freight.). Besides, I thought the mantra was that freight was supposed to be moved off of trucks and onto trains, by subsidy if necessary. You know, that climate thing? Why would you want to move the other way when trucks emit more CO2 per ton-mile than trains do?
There is no case on God's green earth that would cause a freight railway to let its lucrative container business -- "Laser Intermodal Service" CN calls it -- slip away to trucks (AV or not) just so it could operate money-losing slow passenger trains on its now under-used tracks. Why would anyone want to compete to provide a service for "budget-conscious travelers" that loses money? That's a death spiral. The freight railways are out of the passenger business forever. Highways and airplanes killed it. It's dead. If the government wants to subsidize passenger rail on its own tracks, "Fine," say the railways. "Just don't ask us to have anything to do with it. Particularly don't ask us to give priority to your trains on tracks we own and pay to maintain with our freight profits. Those profits, by the way, come from promising our customers that our trains carrying their goods will arrive more or less on time, not delayed by taking the hole every few minutes to wait for a passenger train to go by."
I was unclear, I meant solving the first mile last mile problem for passenger rail, not for trucking.
So you don’t think there is any chance that AVs will have any impact on rail’s freight share…
I would have thought that there’s at least some small chance that more freight will move to AVs, and more passengers will use AVs to travel the first/last mile to train stations, to ride passenger trains that, even if not high speed rail, can still travel at much faster speeds than AVs.
No, I don't think AVs will capture any of the high-volume freight from the railways that trucks don't already have. Once the first-mile-last-mile problem fades out as the total distance increases, the railways will win on price. Remember a load of automobiles assembled in Oakville might still get loaded on a train -- they in fact do -- because they aren't just going to Toronto. The trainload gets reclassified in MacMillan Yard (no relation!) in Concord to be put into other trains going to customers in Montreal, Halifax, Winnipeg, or into New York or Michigan or Chicago. The cargo never needs to be unloaded until the final rail destination. The train cars just get reorganized as necessary that each customer gets his stuff. AV trucks won't get any of that freight. And even if they did, CN's not going to bring back its own passenger trains. If train density took a nose dive, it might charge VIA *more* (not less) to rent its tracks because its dispatchers would have to stop the passenger trains in sidings a bit less often. VIA should be willing to pay higher rent for faster routing.
Where I think passenger train advocates (and HSR advocates in particular) are going wrong is that you want passenger trains for their own sake because you like passenger trains. (So do I, as a matter of fact.) You think that policy makers ought to come up with transportation solutions that give a prominent place to passenger trains. (Which I don't.) The reality (I think) is that only a few romantic enthusiasts actually *want* to take an intercity passenger train. The rest will settle for a passenger train because it's cheaper than flying and less grotty than the bus (if there are even any intercity buses anymore) and can be more pleasant than driving if it's not like four hours late. If it becomes almost as fast as a plane, then the train passengers should be willing to pay almost as much for a ticket. But they won't. They will feel entitled that their good social behaviour in taking the train should be rewarded with a larger subsidy to keep ticket prices artificially low, "to get more people out of airplanes and onto trains!" If ALTO ticket prices are double VIA (as I've heard suggested), most people will just give up and drive. Montreal's not worth that much to the casual traveler or McGill student that they need to get there two hours quicker.
That's what will kill ALTO, as the author has explained. To make the train pay, ticket prices have to be high, but to get stingy entitled Canadians to ride it, tickets have to be cheap. You could force business travelers to use the longer segments that maybe someday will go to Toronto and Quebec City by curtailing the number of air flights permitted between Toronto and Ottawa/Montreal, or Quebec City and Ottawa, but Air Canada's not going to just sit there and take that arbitrary attempt by the government to rig the market in favour of its pet Crown Corporation.
HSR requires its own corridors and substantially different engineering to support both speed and safety.
I believe the argument was that CN might open a cheep passenger line on a non-conjested route to compete, rather than more HSR
AV also wins on time if the train departures are infrequent. But travellers may prefer the train because it is smooth enough to get the laptops out. Time in an AV will be noisy and jittery.
There’s also a case for slowing down. I regularly take a painful early morning flight from Montreal to Toronto and look forward to the day I can take a “red eye” style AV over, sleeping in the car doing a smooth 80Km/h on the freeway so I can arrive in Toronto for 8am meetings after a solid 8hrs of sleep.
Seems more like an argument for sleeper trains tbh.
Not for that distance, unless you want to park the trains outside Toronto or Montreal for a few hours. In steam days, Toronto-Montreal night trains did take 10 hours or so to make the distance. Diesels shortened the run to where you would have left Toronto at 22:30 and arrived in Montreal at 04:00. What are you going to to in Central Station at 4 a.m.? Sit on your suitcase until the coffee shop opens? So they did park the trains on sidings so your arrival would be a reasonable hour to be getting out of bed. Some people can't sleep on trains, at all, no matter how comfy the sleeping berth is.
But sleepers mean the hotel costs of porters and changing linens. On the railway they'll be union labour, not the minimum wages that real hotels pay. Women traveling alone (which they didn't used to do) are going to want lockable compartments that they can lock from the outside to keep predators out while they are out and about, as well as from the inside while they're in there. Sleeping car accommodation was, and is, very expensive. I think it's priced to keep the riff-raff, who get drunk and rowdy in coach on night runs, from buying sleeper class.
I got thinking about cost functions when you realize that ALTO will have to leave from and arrive at some terminal (and intermediate stations) that people don't necessarily live near anymore. (When Toronto's Union Station was opened in 1927, -- building started in 1914 slowed down by the War --there were horse-drawn streetcars and Ford's Model A went on sale that same year. The Subway --to Eglinton, the edge of the known universe! -- was 27 years in the future. Train travel seemed space-age, burning the fuel of the future -- coal -- a step up from wood. (In early days, if the train ran out of wood, able-bodied passengers had to get out and reload the tender from strategically placed woodpiles, and then cut down fresh trees so the wood could dry out for the next train that ran into trouble.)
But my main point is different. The reason why over-the-road (or through-the-air) speed is decisive in costing out a transport system is labour cost per passenger-mile. The nuisance of getting from Woodbridge or Uxbridge to Union (to board a train that might go through Uxbridge but doesn't stop there!) is invisible to ALTO because it doesn't bear that cost. Rather, if it makes the trip to Montreal in 2.5 hours instead of six it has less than half the labour cost to pay the train crews for the trip. This is, of course, why airlines can afford to fly: 5 hours of pilot and stewardess wages to get you YYZ - YVR instead of four days of train crew wages, especially in sleeping and dining cars. (And the explosion in airplane technology means that one plane can carry more people with fewer staff than a passenger train could. Especially since if your flight is only five hours you can put up with hunger and discomfort. Four days and nights in coach on the train is an ordeal even if there is more legroom and if your drunk seatmate doesn't snore all night.) So people will expect to pay more for an air ticket for the same distance because it's so much faster.....and it's the very speed that makes it a profit-making proposition for the airline. (Then there is the lucrative "rail-cruise" market that "The Canadian" has become. Much slower than in the 1950s but rich foreign tourists don't mind especially when the dollar is cheap.)
So where does ALTO fit in? The faster they go, the less they have to charge to make back their labour costs per trip, and the more trips they can make in a day. But the fewer stations there are, and the farther people have to drive (and park downtown!) to board the train, the less desirable it is. Drive to YYZ or drive to Union? Hmmm. Yes, the airline also doesn't have to cost the time factor of you getting yourself to the airport and through security. It's sensitive only to how long it takes to get one planeload to the gate and push back out again to Montreal. It will be able to do this faster than ALTO will, just because the conveyance moves faster, no matter how long it takes an individual passenger to get to the gate. Door-to-door ALTO might be faster for downtown livers than Air Canada, but Air Canada will have a time-cost advantage that ALTO will have to meet in order to price its tickets competitively. (Now sure, pilots are more expensive per hour than train drivers, but air hostesses may be cheaper than "service managers" who replaced conductors on VIA trains 20 years ago, who are all members of various operating brotherhoods who know a thing or three about feather-bedding.)
Bottom line: ALTO will struggle to earn enough to justify the subsidy just because it has a time-cost disadvantage compared with flying, being slower gate-to-gate while the employees are on the clock. That it might be attractive on total trip time for *some* passengers (disappointingly few in our suburbanized country) doesn't change the fact that it will cost more staff time to move one passenger from Toronto to Montreal than it costs Air Canada. (Robot trains with no human staff on board -- just cameras to police bad behaviour -- would reduce that cost disadvantage, of course.)
You see what I'm doing here: engaging with this Substack vigorously enough to justify upgrading! :-)
This is *such* an excellent Substack. I would love to upgrade but I'm getting buried in subscriptions and I'll have to discontinue some before I bite your bullet.
Taken as a whole, I don’t think this commentary has a leg to stand on. Just my opinion.
Maybe commercial flight should be grounded and we should scrap the whole autonomous vehicle paradigm because flying cars will render both platforms pointless, useless, worthless.
But seriously…
The reason we have trains is because they can move vast quantities of goods and large numbers of people extremely efficiently and at relatively low cost. Electricity makes that task even more practical, allowing for increased efficiencies and providing even greater savings. Improved transit times mean upped average speeds. Accomplish that, and all of a sudden rail starts to look really attractive as an alternative to both driving and flying. High-speed rail, with its own dedicated, double-tracked right of way, with sidings and crossovers strategically located thereby permitting express trains to overtake the all-stops “cruisers” that serve the intermediate stations, this will not only allow for unencumbered operations, but make the overall service a seamless (or much more seamless) one, especially with the trains running city center to city center.
Furthermore, trains in the grand scheme of things have a considerably smaller carbon footprint, comparatively speaking. And railroading is the safest “automated” form of transportation going.
Meanwhile, in the U.S., recorded every year are upwards of 40,000 highway deaths. Autonomobility (driverless technology) may one day eliminate all human-error-caused roadway-based deaths, but it won’t end all roadway-derived crashes. Car crashes will still occur due to such inevitable factors as vehicle mechanical and/or electrical breakdowns and/or onboard computer glitches that will crop up from time to time, affecting motor vehicle operation and/or performance in some way, shape or form.
You seem to be pinning a lot of hope on the AV platform being the be all, end all. You, and a whole lot of other folks. I’m not as optimistic. The same kind of optimism I have about flying cars en masse ever getting off the ground.
Then again, there’s always the chance I’m wrong.
Does the highway network (even with substantial upgrades) have the capacity for the additional vehicle traffic? From what I understand, passed a certain point, capacity upgrades to highways have diminishing returns. And the 401 is, arguably, passed that point already.
Is the assuption that AVs drive in a more efficient, collective, manner that reduces capacity constraints? Does it hold when it's AVs and human piloted vehicles operating together?
I don't know how to go about modelling it myself, so I'm just thinking out loud here and not trying to present this as definitive. But I'd think the highway networks would be a congested mess with most additional trips using the highways for the next 50-odd years.
In fairness to leaving out the capital cost of HSR, it's also assumed the highway will be toll free and it's publicly funded infrastructure the AVs operate on.
But really, HSR already has this competition - it's the existing VIA train. I have no idea how HSR could be expected to operate anywhere near the current trains cost, and I don't see how an HSR that costs a jet ticket could possibly have the volume of passengers needed.
It is disappointing, I could do so much for public transit with $60B in capital spending.
Not quite: everyone gets to use the publicly funded highway, not just the AV service company. The capital cost of HSR goes entirely to the exclusive use of the HSR corporation.
HSR for sure won't be able to operate anywhere near what the current train costs. It will be far more expensive per passenger mile to operate (not just build.) The higher speeds will require more propulsive energy to overcome air and rolling resistance. Electricity isn't free. The track and rolling stock will have to be maintained much more meticulously. You might need a pilot vehicle to go ahead of every train to make sure that vandals haven't fucked with the track and signals, as Idle No More periodically threatens to do.
Finally, ALTO will need to pay a squad of goons and thugs to intimidate the locals into acquiescence about the railway's plans to expropriate their land and cut off their roads to avoid building expensive grade separations.