Debate on the merits of the Scarborough Subway Extension

[jaybe]

^Early transit systems including subways were built by the private sector. They were built where demand would generate a payback. We need some of that kind of thinking with regards to transit planning today. As much as vision and equity need to be part of the discussion, existing demand must be the over-riding priority. The STC is a long, long ways from downtown, almost as far as Mississauga City Centre. No one is proposing a subway be built to Square One. For Scarborough, build the most effective transit that meets demand now and reasonably into the future.... LRT.

 

[jcam]

The annoying part is that it's in the middle of about 40K people's commutes, it's not like they're all going in different directions, they're all going to the same place.

So that means we should pay $83,000 per daily rider to make a transfer easier, rather than redeploy that money on projects that will bring new transit?

For example, Eglinton West LRT gives 26k people new transit for $90k per daily rider, with $1.2B left over.

 

[aquateam]

They'd have to compromise with the union since it is their job to look out for workers' jobs. They'd probably make a case when they open Relief Line North, where operators will need to be shifted over to subway trains so they can take them off the SRT and the TTC won't have to lay anyone off.

If Relief Line South doesn't operate driverlessly, let alone Relief Line North, then someone made a bone-headed decision. The modern design standard is driverless and PSDs: no brand new system opening in Asia or Europe would be manually operated. It might be a North American thing to completely ignore modern design standards and international best practices (e.g. open gangways, ERTMS), but hopefully the DRL won't.

So that means we should pay $83,000 per daily rider to make a transfer easier, rather than redeploy that money on projects that will bring new transit?

For example, Eglinton West LRT gives 26k people new transit for $90k per daily rider, with $1.2B left over.

Both of those are pretty awful projects based on that metric TBH:

 

[Streety McCarface]

If Relief Line South doesn't operate driverlessly, let alone Relief Line North, then someone made a bone-headed decision. The modern design standard is driverless and PSDs: no brand new system opening in Asia or Europe would be manually operated. It might be a North American thing to completely ignore modern design standards and international best practices (e.g. open gangways, ERTMS), but hopefully the DRL won't.

Both of those are pretty awful projects based on that metric TBH:
View attachment 144769

The scarborough subway is projected to receive 60-70K daily riders, more if more stations are added. Whether or not it will reach this level is in question, but the bigger issue with the analysis above is that it doesn't factor in inflation and the rise in construction/operation costs. Throw those in, and cost per rider goes up for almost all those projects. Also, I have my doubts about the spadina subway extension. Every time I've ridden on the train, there are fewer people on it than the sheppard subway, but again, lines take lots of time to develop ridership and it's probably unfair to fully judge a line 10, or even 20 years into its operation.

 

[Rainforest]

People who think Spadina streetcar needs to be replaced with a subway, simply don't understand how Spadina streetcar works.

Sure it gets lots of daily boardings, but most of riders hop on it, ride for a few stops, hop off. Hardly anyone rides this route end to end.

Therefore, the peak demand doesn't exceed 2,000 pphpd anywhere along the route. That's perfectly manageable with the large new streetcars running on dedicated tracks. For the record, that's less than 1/2 of the current Sheppard Subway peak, and less than 1/3 of the projected SSE peak.

If you replace Spadina streetcar with a subway, then you:
a) Cause a lot of construction pains, for no gain.
b) Lose a lot of existing stops. Nobody is going to build a subway with 300-m stop spacing these days.
c) Misuse the funds that could much better be applied to extend the Relief Line further west from University. One can debate Dufferin vs Dundas vs something else as the preferred route, but it certainly needs to go west further than Spadina.

 

[Rainforest]

So that means we should pay $83,000 per daily rider to make a transfer easier, rather than redeploy that money on projects that will bring new transit?

For example, Eglinton West LRT gives 26k people new transit for $90k per daily rider, with $1.2B left over.

1) First of all, "cost per daily rider" is not a useful metrics for capital investments. Noone is going to use those $90k in transit benefits in just one day.

For capital infrastructure, spreading the benefits over 50 years is reasonable. Furthermore, I prefer to count cost per ride rather than daily rider (most riders make two rides every day).

Thus for Eglinton West, we are going to have 52k daily rides X 365 days/year X 50 years = 949,000,000 rides. If the capital cost is 2.34 billion (is it really that much? but anyway), we get about 2.5 dollars per ride. A pretty good price, that's 1/6 of the new minimum hourly wage. If the new line saves just 10 min per ride, it is already worth it.

2) Now for SSE, we aren't building the subway just to eliminate the transfer. The existing line (SRT) is crumbling and needs replacement. Thus, the cost of removing the transfer is equal to the difference between the cost of subway and the cost of cheapest alternative (light rail in the Uxbridge corridor).

Not sure how much that difference is; I assume the ballpark is $2 billion for the tunneled subway under McCowan, or $1 billion for a subway in the same Uxbridge corridor.

For 60k daily rides, we will have 60k X 365 days/year X 50 years =1,095,000,000 rides in 50 years. Capital cost per ride: about 2 dollars in the first case, or about 1 dollar in the second case. Surely worth it.

 

[Streety McCarface]

1) First of all, "cost per daily rider" is not a useful metrics for capital investments. Noone is going to use those $90k in transit benefits in just one day.

For capital infrastructure, spreading the benefits over 50 years is reasonable. Furthermore, I prefer to count cost per ride rather than daily rider (most riders make two rides every day).

Thus for Eglinton West, we are going to have 52k daily rides X 365 days/year X 50 years = 949,000,000 rides. If the capital cost is 2.34 billion (is it really that much? but anyway), we get about 2.5 dollars per ride. A pretty good price, that's 1/6 of the new minimum hourly wage. If the new line saves just 10 min per ride, it is already worth it.

2) Now for SSE, we aren't building the subway just to eliminate the transfer. The existing line (SRT) is crumbling and needs replacement. Thus, the cost of removing the transfer is equal to the difference between the cost of subway and the cost of cheapest alternative (light rail in the Uxbridge corridor).

Not sure how much that difference is; I assume the ballpark is $2 billion for the tunneled subway under McCowan, or $1 billion for a subway in the same Uxbridge corridor.

For 60k daily rides, we will have 60k X 365 days/year X 50 years =1,095,000,000 rides in 50 years. Capital cost per ride: about 2 dollars in the first case, or about 1 dollar in the second case. Surely worth it.

It's really hard to put an actual timeline, but the LRT one should be less than the subway one simply because track replacement for an LRT is significantly more expensive than for a subway train (you have to rip up concrete). For that reason, I'd reduce it's lifespan to 30 years and the subway to 40 years (due to rolling stock replacements).

It's also worth noting that neither of those calculations factor in increases of ridership over time. The actual subsidy per ride would be a whole lot less if ridership is assumed to grow from day one on. Seeing the general trend for the Bloor Danforth line over the past 15 years, I'd say it's fair to assume that usage will significantly increase during the 40 year lifespan.

 

[Rainforest]

It's really hard to put an actual timeline, but the LRT one should be less than the subway one simply because track replacement for an LRT is significantly more expensive than for a subway train (you have to rip up concrete). For that reason, I'd reduce it's lifespan to 30 years and the subway to 40 years (due to rolling stock replacements).

In this case I wouldn't calculate based on the track or rolling stock lifespan, because even when those get replaced, it still costs a fraction of the original construction cost.

And, to be fair to the opponents, LRT should not be more expensive to replace than the subway, because either line would be fully grade separate and the tracks don't need to be encased in the concrete. The concrete part applies to on-street LRT, but we wouldn't have it in this case.

So, I took 50 years as a period after which some completely new mode of transit might be invented.

It's also worth noting that neither of those calculations factor in increases of ridership over time. The actual subsidy per ride would be a whole lot less if ridership is assumed to grow from day one on. Seeing the general trend for the Bloor Danforth line over the past 15 years, I'd say it's fair to assume that usage will significantly increase during the 40 year lifespan.

True. Both Eglinton West and SSE should gradually grow ridership as time passes, making the cap cost per ride even lower.

 

[syn]

^Early transit systems including subways were built by the private sector. They were built where demand would generate a payback. We need some of that kind of thinking with regards to transit planning today. As much as vision and equity need to be part of the discussion, existing demand must be the over-riding priority. The STC is a long, long ways from downtown, almost as far as Mississauga City Centre. No one is proposing a subway be built to Square One. For Scarborough, build the most effective transit that meets demand now and reasonably into the future.... LRT.

Of course not. Mississauga is it's own city, therefore there is a certain level of necessity for fiscal prudence.

Scarborough gets to make the rest of Toronto pay for their demands.

This is why Scarborough should really be it's own city - they can elect all the Fords they'd like, and come up with whatever schemes they want to pay for. They'd quickly find that building subways in their low density environments is simply not something they could afford to do.

 

[Cobra]

Scarborough gets to make the rest of Toronto pay for their demands.

Yet Scarborough never paid a dime for the rest of the city's projects... ok

 

[OneCity]

Yet Scarborough never paid a dime for the rest of the city's projects... ok

It's great to compare Sauga to Scarborough until we see that Sauga has had massive transit infrastructure investment of other forms in the previous decades, has not paid directly into the TTC and has it's own democratic vote. To assume they won't extend the subway down the road is also very naive.

Scarborough doesn't have a fraction of the highways Sauga has so we better build well connected public transit to the City it pays taxes in.

 

[syn]

Yet Scarborough never paid a dime for the rest of the city's projects... ok

Given their tax base, Scarborough is most likely being subsidized by other parts of the city. Suburbs usually are.

 

[jcam]

1) First of all, "cost per daily rider" is not a useful metrics for capital investments. Noone is going to use those $90k in transit benefits in just one day.

Agree its not perfect, but its the simplest way to compare projects apples-to-apples...what you're interested in is a relative comparison of some kind.

2) Now for SSE, we aren't building the subway just to eliminate the transfer. The existing line (SRT) is crumbling and needs replacement. Thus, the cost of removing the transfer is equal to the difference between the cost of subway and the cost of cheapest alternative (light rail in the Uxbridge corridor).

Yes, makes sense. So subway = 3.35B+, LRT= 1.48*(1.02^8) = 1.86. So the difference is $1.8B for the transfer, or $30,000 per daily rider for removing a transfer.

For 60k daily rides, we will have 60k X 365 days/year X 50 years =1,095,000,000 rides in 50 years. Capital cost per ride: about 2 dollars in the first case, or about 1 dollar in the second case. Surely worth it.

You're looking at the project in isolation, which isn't reality. We shouldn't be sayin "that's not much incremental money", we should be saying "I have X dollars to spend on a slew of transit backlog projects, is this good value for that money, or is another project?".

 

[Rainforest]

Agree its not perfect, but its the simplest way to compare projects apples-to-apples...what you're interested in is a relative comparison of some kind.

Yes, makes sense. So subway = 3.35B+, LRT= 1.48*(1.02^8) = 1.86. So the difference is $1.8B for the transfer, or $30,000 per daily rider for removing a transfer.

Capital cost "per daily rider" isn't just imperfect; it is grossly misleading. It shows monstrously high costs for any transit project, not just SSE.

Even if you are installing a new bus shelter for $5k, and 50 riders use that stop every day, you will end up with $100 per daily rider.

Attributing the cost to all rides per a set number of years makes the comparison to user benefits much more meaningful. And, you still can make relative comparisons.

 

[sixrings]

Those downtown elite dang money burning lefties trying to stop fiscal conservative progress for the common folk down at The Star. help is on the way
https://www.thestar.com/news/city_h...ave-paid-for-scarborough-lrt-emails-show.html