Metrolinx has updated their report on daily Union Pearson Express ridership to include the last two months of 2015.
The chart below shows the original ridership projection (blue), the day-by-day figures (green) and the seven day moving average (red).
No additional comment is necessary.
Barely a year and a half after its launch amid high expectations, Target made the decision to pull out of Canada after realizing it wouldn’t reach profitability until at least 2021. It had spent $7 billion to that date. How long will it be until Metrolinx pulls the plug?
At least most of the sunk costs are salvageable. The Weston Corridor, even the trainsets. This isn’t exactly a white elephant. But how can we expect Metrolinx to come up with a credible fare integration strategy when it can’t even get GO Transit fares right, never mind the UP Express?
Since we’re all so fond of back-of-the-envelope assessments in this town …
Even if this is some seasonal blip and ridership averages 2500/week for the year, and we assume that all riders pay the MAXIMUM fare (totally unrealistic yes), that works out to no more than ~$68,750 per week or $3,575,000, against expected operating budget of $68-70 million (from Dec 10 2014 Star article). So it costs taxpayers at least $64.5 million for 130,000 UPX trips, or a subsidy of ~$496 per rider.
Suddenly $0.88 a rider on the TTC doesn’t look so bad !
It is comforting to know that these transit professionals at Metrolinx, with the Ontario Government in full support and backing, have done such a wonderful job with the UP Express. It’s just uncanny how their predictions of passenger traffic have worked out so well. (but, to make it look even better, lets throw in a long weekend of free travel on the UP Express to get the numbers up!) With the UP Express success under their belt, they can now jump right in and see what kind of a smooth integration they can dream up for SmartTrack and RER. Cost should not be a consideration, just back our transit expert Mayor Tory.
They are running 5 trains a day with 3 person crews, engine person, conductor and ticket checker plus usually at least 5 people at Union and 5 at Pearson running around trying to look useful. Add to this the cleaner who gets on and sweeps away the detritus from the 8.5 passengers and the maintenance, personnel, OH and don’t forget the Vice President and her underlings and I am sure that this is bleeding money faster than a gambling addict at a slot machine.
Is it possible to get all the wage costs or is that considered proprietary?
Steve: This should show up when they publish their financial statements (which are broken down by division), but they won’t be out until the middle of the year (yearend is March 31).
I am pessimistic about UPX, but I’ll play the Devil’s Advocate.
Excluding the first two week “novelty” surge, there seems to be some interesting flux regarding when holidays fall.
For example, the Friday and Thursday before Canadian Thanksgiving were the highest outside of June. In fact the Wed-Fri prior averaged 3667 passengers and the Wed-Fri after averaged 3122 passengers. On the flipside, the lowest total occurred on Christmas Day, when Fridays are normally the strongest day of the week.
The extra Wednesday in September and Thursday/Friday in October tended to boost their perceived average ridership.
They are still well short of 5K, but I’m beginning to think it might be achievable. If they drop it to TTC/GO fares, they’ll get there, but the whole premise of 5K will have been forgotten in the meantime.
Steve: And the hopes of breaking even will recede even further from credibility. Remember how in the early UPX days they claimed they had to have a high fare to discourage riders lest the trains be too crowded?
As you say… “No additional comment is necessary”
Seriously Steve, I think much comment is required. This continues to show, month after month the critical waste of precious track especially in the USRC, the incredibly bad service design, and lack if imagination that was initially involved. This really needs to be rethought using the track for service through the airport area, with the circulator coming out to meet GO at a real transit hub. Bringing the Crosstown, and Finch LRTs to the same hub, along with MiWay and ZUM service, would make for a real connection for the airport district, that would truly have an effect on airport area traffic, have the potential to serve commuters as well, and be transit connection that was highly effective and Toronto and region could be proud of.
Looking at the particular dates, the 4093-rider peak on October 9th was that 14-inning 5-hour Blue Jay Game 2 playoff game at Skydome. The preceding day was 3802 was Game 1
As I suspected, much of the peak is post-game traffic. The preceding day was 3802 was Game 1. The other playoff games are higher than normal on those days. So that’s much of the late-August hump. I wouldn’t be surprised if much of the September hump was Blue-Jay related.
And what this means, is this is total traffic, not Pearson traffic.
Shame we don’t have the Pearson-only traffic from Union. That probably would easier to discern trends from.
Dear Metrolinx, please turn this into just another go line with go fares and add stops at Queen/Dufferin, Mount Dennis, and Woodbine Racetrack.
Get rid of the UPX/SmartTrack branding.
UPX is just a spur of the Kitchener go line.
Shut the line down until ridership can be built with density, jobs, and air traffic growth.
I am astonished (or maybe not) that neither Metrolinx nor its critics understand the most valuable thing about UPX, which is the direct connection between Canada’s largest airport and the region’s busiest transit hub. Metrolinx’s insistence on marketing the service to business travelers was, both at the time and in hindsight, idiotic.
I predict UPX will be closed inside 12 months, and a conversion to transit will be promised but later cancelled, because that’s the way things go around here. Great work everybody. This is why we can’t have nice things.
$68M-70M is per year, not month and 2400 per day, not week.
Using the average daily ridership, it’s 887,494 riders in a year. At $70M in operating costs, that’s $78.87 per rider or $51.37 after an assumed full fares. However, in that same December 2014 article, Metrolinx stated that it’d be 3-5 years before the operational breakeven point.
Why are the quoted operating costs so high ($7,990.87 per hour 24×365)? They obviously need some trimming (say 2-men crews instead of three).
Beyond that, operating costs and profitability shouldn’t be the main focus of a transit network (it should be a consideration because of opportunity costs of using that money elsewhere).
I’ve taken the train a few times for business and for personal trips, but mostly to experience the novelty. (I work at a smaller mid-size company ~50 people.)
The biggest value is that I can take out my laptop and work during the ride. That’s not quite so easy on the bus or the subway.
But the trade-off is that it’s not so convenient to get to. If you don’t work in the financial district, then you have to subway to Union or Bloor/Dundas West and walk those long connections. When I check Google before I leave, it often tells me that UPX is only a few minutes savings over taking the Bloor line and 192.
Given that I’m not saving a ton of time, but just putting time to better use, the price is kind of funny. People more senior than me still take a cab because it isn’t an order of magnitude more expensive. And if two people are going together to a flight, a cab is a no-brainer. If UPX were considerably cheaper, it might push the company to encourage people to change their habits.
I wonder if they’ve done any surveys/analysis of the types of people who take the UPX and what their thinking is.
Steve: Given the degree to which they have been pushing “family” use recently, I can’t help thinking that they might sense the business market is tapped out. This implies a major flaw in their original market analysis.
There were 60 of 209 days with Jays games (10% of those play-off games).
Without controlling for other variables you get:
However, if you eliminate the playoffs and June 6 (UPX opening day changes the Saturday delta by +690.9), the picture is very different…
If you also remove Christmas Day from the analysis, Friday changes to 2,703.3 average and 2,494.3 non-Jays for a delta of 731.6 (change in delta of -262.8).
Looking just at the Fridays…
The peaks and valleys tend to bridge beyond just the Jays home games… Maybe need to add the Leafs and Raptors into the mix.
Steve: I can’t help noticing that this is not the sort of market Metrolinx thought it was going after.
They need to find out the actual demand for this service. It could be high as they say, and only being limited by high prices. Or it could be low, masked by high prices.
I propose a month long apology for the high prices at the start. Run it for free for a full month…after a day or two, and definitely by the end of the month you will have got way more people using the system and have a better idea of who your customer base is. You can survey them, talk to them, and get them used to the service.
Have big signs by the taxi stands at the airport…people with balloons whatever.
Once you know your customer you can figure out how much they are willing to pay, and can maximize your revenue.
LikeLiked by 1 person
While I like a good back-of-the-envelope as much as the next guy, it looks like you’ve inadvertently misread the ridership numbers. The numbers plotted in the chart are *daily* ridership, not weekly–so your revenue estimates are low by a factor of 7.
If we take the year-end average of about 2000 riders per day (not per week) and assume they’re averaging Presto fares ($20-ish) then we’re looking at $40k per day from the farebox, or about $15 million per year. If annual operating costs are around $70 million, that’s an operating subsidy of $55 million, or just under 80% of operating costs–not the 95% you calculated. (A 20% farebox recovery ratio is still ugly, but not quite as epically bad as you calculated.)
If they cut fares in half and doubled ridership, the recovery ratio would still be terrible, but at least the per-ride subsidy would improve. And it’s possible that they would attract more airport employees that way.
@Build Scarborough subway now (Re: “Shut the line down until ridership can be built with density, jobs, and air traffic growth.”)
I admit that I have great difficulty reconciling your comment with the policy position embraced in your username.
Steve: Airport employees already have a $10 fare, and so unless Metrolinx cuts this in half too, a cut on the regular fares would have no effect on this group of riders.
I wouldn’t say idiotic and instead say naive because there are a few factors in the Canadian economy right now which might be contributing to the poor performance of the UPX. Mind you, those factors would only amount to the difference between flat and declining ridership and the fares were certainly not targeted to modern business travellers.
No comment? How about “Oop$”?
Given the depth of knowledge of both Steve and commenters, it’d be really fun to get the dirt on how well major transit expansions of the last say 40 years have met the pre-build hype. To be clear, the original Spadina extension, and the Sheppard Stubway – how do they match the numbers that were projected?? And let’s watch the second Spadina/Sorbara extension, tho at least up to York University made some sense, just the way the contracts were structured, no, can’t manage to do it in a linear way that might allow for the line north of York to be scrapped, or so I’ve heard.
So just to make sure I understand:
A line running between one of the large airports of the continent and one of the large train stations of the country which is incidentally right in the densest part of the city should be shut down until density and potential ridership rise, rather than being adjusted to make proper use of the investment that has been made.
Meanwhile, a fantastically expensive single-stop subway extension running to a medium-density node in the middle of low-density suburbs whose demand projections can be handled easily by an LRT line should be built now?
Please, let me know how I have misunderstood.
LikeLiked by 1 person
The suggestion above (from Asher) that the UPX needs to understand who their current customers are seems like a very useful one. If they know who (and why) their current customers are using UPX they may find out how to attract the huge number of non-customers. I have taken UPX a couple of times (and my neighbours more often) and none of us have been surveyed.
For what it’s worth, I live about 10 minutes from Union so can easily walk there and I normally take the TTC to Pearson but sometimes (rush hour or after a very long flight) it is safer/more comfortable to take the fast (and empty) train. However, even at $19 it is really not worth it for me to take it all the time and if I also had to take the TTC to get to Union I would certainly just go to Kipling and take the very efficient 192 bus. As most UPX riders will have to take the TTC to get to it, at the very least they should give Presto users free TTC rides at both ends of a UPX trip – this could surely be programmed into Presto?? (I assume it would be virtually impossible to do this for non-Presto users.)
Maybe, Metrolinx should abandon the idea of breaking even on UPX, and instead maximize the usage.
Unfortunately it is not possible to run regular GO trains on the airport spur without a massive reconstruction. Therefore, the price cannot be lowered to a regular GO or regular TTC price (small UPX trains will get overcrowded).
But they can lower the price by say 20% from the current level, and see how it affects the ridership. If there isn’t much effect, then lower by additional 20%.
Another possibility is greater discounts for families. Some have pointed out that a cab is comparable in price, or even cheaper, for a group of 2+ people travelling together. The UPX pricing should reflect that.
I would suggest to you that the entire notion of having a matched set of tracks (1 each way) to support a single destination, is madness where track space is so tightly constrained, unless it actually generates a huge (20k peak) demand. The track should be reconsidered, so that the airport spur, is special, is used for local service, where there is a station for the main line, or dropped entirely. If the UPX was 1 train in 2 or 3 on this rail set it would be easier to take. However, the fact that there is no room in the USRC, really means that the track there and the space in Union needs to be reconsidered.
I would put to you, this in 3 car trains every 5 minutes, where 1 in 3 was a car with a baggage allowance, would be more reasonable use of track. While that would likely only net a capacity of only 3,000 passengers/direction/hour, that would likely be used (overused) as regular high frequency GO service, and if properly aligned/integrated in platforms with a periodic full sized GO train make the combined services more attractive. That is in station you would board the high frequency train to your left, the regular GO to your right, so that the full sized GO could empty the platforms as the ones on the inner part of the line get overloaded, much like Yonge/Bloor subway does. While I will concede that such a high frequency service extended out as far as Brampton GO would likely be overloaded, by the time it reached the Malton Station, it would actually make use of the tracks and platforms dedicated to it. If the platform at Union could somehow be extended to allow for 5 or 6 cars, this service would likely be enormously successful at taking cars off the road (or at the least reducing the number of new cars getting on).
You don’t even know the half of it. 3 man train crews are one thing, but that’s kind of hard to get away from considering Transport Canada wants passenger trains to have at least 2 rules qualified people on each train. And to be fair to MX, they did have SPO in mind since UP consists were designed for a one man operation. The conductor doesn’t even have a seat on the C car, he sits on an electrical cabinet with his feet on the cooler hah!
The GSR is basically a ticket checker, but at least they’re only being paid a fraction of the costs of the in house guys that work the GO trains. But check this out, right now they have a full time employee (contracted by PNR) positioned at both Union & Pearson stations whose only job is to make sure the the platform screen doors open and close properly. Every once in a while, can’t say how often exactly and that’s another matter, they won’t open or close even after the train is spotted up properly. His/her only job is to manually flip a switch in those cases. You can see them sitting off to the side at a small table usually with a laptop. And they’re getting paid about 30$ an hour to do this, that’s about 10 grand a week lol. It’s a good job if you can get it.
Check this out – UPX President Kathy Haley is speaking next month in Germany at a conference of fellow Kool Aid drinkers on – wait for it – customer experience! Well I guess the experience is pretty nice when the train is near empty and there are more employees than customers. Even if they hit their target of 5000/day, that’s still only, what, 30 per train? And what about the customer experience for lowly GO passengers? UPX trains get priority so GO trains (which are actually full of paying customers) routinely have to stop for a couple of minutes to watch that wonderful UPX experience sail by.
Steve: Metrolinx managed to get an award from the Global AirRail Alliance in 2013 proving that you can get an award for an idea that is little more than hot air. This is a perfect example of a problem I saw originally at the TTC where, quite literally, the former head of planning would hunt around for awards they could be nominated for and “win” thereby validating how wonderful management were. This kind of BS is too prevalent, and Metrolinx should be ashamed of it.
In a way I wonder if I am the only one out there who would ever push hard enough for this city to finally have some sort of inter regional rapid transit system.
Of course there are other metropolitan regions that have that sort of infrastructure, like the MTR in Hong Kong or the recently built Overground in London.
With the Union Pearson train one travelling from the GTA from the suburbs has to go to the downtown at Union Station and then jog back to the suburban Pearson airport. And now the Union Pearson train is losing money of all things.
There will be so much heated debate about transit and none of it concerns putting in place a comprehensive suburb to suburb Metro rail system.
The electrification of some of the GO Train lines will help, but there is still no debate on having a sort of GTA Metro rail system like the MTR or like the Bay Area Rapid Transit system.
There is still a lack of that kind of big thinking. And of all irony it was the Progressive Conservatives back in the early 1980s that would have built a sort of BART system through the then GO ALRT Program.
Regardless of which at some point there should at least be some discussion given to the idea. All I hear about is some silly named project called “Smart Track”. What does the city of Toronto have to do with planning and building something the provincial government and GO Transit would be more concerned about.
At the very least we will be lucky if someone perhaps more intelligent like Kristyn Wong Tam were to wade into the next mayoral race. We lack political leaders who would concern themselves with thinking big, not laughable mayors or deputy mayors trying to look off like media personalities..
Presumably there will be a huge spike because the UP Express was free this weekend. The long lines clearly indicate that the fare has to be significantly higher than TTC fare (given the low capacity of the trains) to prevent overcrowding. I managed to avoid waiting in line when I took it Sunday morning around 9am to the airport, but there were long lines in the other direction a few hours later and I took a (relatively empty) bus route 192 back.
I severely doubt if ridership will go up very much when the price goes up to the usual $19. Clearly there is demand for this service, but $19 ($27.50 without Presto) is outrageous. I think that if the price is cut to $9.99 it will be well used. The UP Express should not be far more than airport rail links in the US (if you ignore the effect of the low Canadian dollar).
I would suggest that the spike this weekend is caused partly by the price and partly by the novelty. If they simply stopped charging fares permanently you wouldn’t necessarily see that much traffic all the time. In fact I was considering bringing the kids and taking a round trip on the UPX, possibly with a side trip on the LINK train. At no time did I consider getting on a plane this weekend. How much of this weekend’s traffic was similar sightseers?
Having said that, it’s pretty clear fares could be dropped a lot. Maybe by a factor of 5. If the trains then filled to 50% (a ridership increase by a factor of about 5), the revenue would be the same. Since that would be approximately a $5 fare, I think a 5-fold increase in ridership is entirely believable.
Really they should just use the SFPark algorithm to set the fares. Done. No arguments — just keep the trains close to but not quite full at all times. Then nobody can argue the train is benefiting few people. Of course it might turn out that the market price of the service is embarrassingly low and the subsidy correspondingly high but only economists and accountants and similar sorts of experts would really notice since the trains would at least be reasonably full.
Reportedly 2 hour lineups to get on UPX this weekend. Seems like people would love it if the fare drops.
There totally was, as least on Monday. Could have been anywhere from 500-700 people in line at Union — I thought they were going to the aquarium. Since this isn’t the latest roller coaster ride, I skipped I decided to wait for another government-sponsored freebie occasion. Since the maximum capacity is only about 170 passengers, I’d have had to wait about an hour just to board and shuddered to think of the wait for my return trip.
I wouldn’t be surprised if Metrolinx was using the occasion to take a lot of pictures of full trains for their website. I was touched to see so many families out to experience a new rail service; there’s something heart-warming about it. But the unintended lesson for many, I fear, was that this is not a service that can handle commuting numbers. That’s the joke of this half-billion dollar boondoggle: too few riders most of the time, far too many when the price is right.
Shut down UPX. Run the free airport people mover along the old UPX airport spur. People mover trains are light enough for the spur bridges. Move the Etobicoke North station west to the racetrack where it will be the terminal for the airport peoplemover. Increase GO service to 15 minutes along the Kitchener line between Union and Etobicoke North using the old UPX tracks. Now, GO passengers coming from east or west along the Kitchener line can stop at Etobicoke North and connect to the airport through the free people mover. Problem solved. Why can’t the gov’t figure this out???
The airport people mover is basically a horizontal elevator. I don’t know if it would go that distance.
Steve: There is also a question of its capacity when it meets a train with a large crowd.
I should really just write this all up and post it on the internet somewhere and then just reply with a link every time it comes up.
The existing APM is 1.473km long with 3 stations, 2 six-car trains, 250s headways, 36s dwell time, 36 passengers/car, 180 passengers/train, and/or 2150 pphpd. The operating speed is 43.2 km/h (12 m/s), acceleration of 0.55 m/s² and deceleration of 0.80 m/s².
The Pearson subdivision is 2.77km (1.72 miles) long from the Signals at Mile 13.6 at Goreway Drive.
The system maximum is 9km, but typical installations are under 4km.
If you use the current speed of 12 m/s, you have 33s acceleration, 27s deceleration, 216s dwell time, and 706.8s travel time, so 982.8s round trip. With 4 train sets, you have 352 pphpd seated capacity (1319 pphpd crush capacity) and 245.7s headways.
If you increase the speed to 14 m/s, you have 605.9s travel time, 392 pphpd seated capacity (1470 pphpd crush capacity) and 220.5s headways.
However, your transfer capacity would be much less as you have 4 trains per 15 minutes (4.08) on the APM and 1 train per 15 minutes on the new UPX. Seated capacity is not sufficient to meet current peak-load travel on the UPX.
As a comparison: UPX 3-car consist has 948 pphpd seated capacity and 1908 pphpd crush capacity.
How do you put 4 train sets on? Right now each train set has its own track. Can the system handle some train sets passing? How would it work — multiple cables per track? I know cable car systems can be built with passing sections but it’s not clear to me how it would be done with this particular system.
Anyway the real solution is to run the Eglinton and Finch LRTs to the airport and have them meet up in a no-fare segment from the GO line to T1. At least that’s my made-up-in-seconds solution to the problem.
It seems to me that the problem is at least as much political as technical.
A world class city has a direct rail link from downtown to the airport. Any other transportation between the airport and UPX whether people-mover or as someone else suggested LRT would be seen as a transfer, and this is unacceptable for a world-class transfer to/from downtown.
This may seem ridiculous, but I truly believe this is the way the politicos see it.
Note: I am definitely not saying I buy into this position.
The airport rail station is not even big enough to handle any high capacity passenger trains like a real airport link to such a large airport should be able to. As it is now you gotta transfer anyways if you’re going to Termial 3. So I agree with “Obviousman” in trying to make what we have so far work, transfer at a station on the main line and shuttle from there to the airport. Let the big GO trains do the hauling and give access to the rest of the system. Even VIA Rail could serve Etobicoke North which should be a modern larger station much like Aldershot station.