Northeast Corridor Travel Markets and Fares
I’m optimistic about the ability of fast trains on the Northeast Corridor to generate healthy ridership at all times of day. This is not just a statement about the overall size of the travel market, which is naturally large since the line connects three pretty big metro areas and one very big one. Rather, it’s also a statement about how the various travel markets on the corridor fit atypically well together, creating demand that is fairly consistent throughout the day, permitting trains with a fixed all-day schedule to still have high occupancy – but only if the trains are fast. Today, we’re barely seeing glimpses of that, and instead there are prominent peak times on the corridor, due to a combination of fares and slowness.
Who travels by high-speed rail?
High-speed intercity rail can expect to fill all of the following travel markets:
- Business trips
- Tourism and other recreational trips (including visits to friends and family)
- Commute trips
The key here is that the first two kinds of trips can be either day trips or multiday trips. I day-tripped to the Joint Mathematics Meeting in 2012 when I lived in Providence and the conference was in Boston, about an hour away by commuter rail, but when there was an AMS sectional conference in Worcester in 2011 (from which my blog’s name comes – I found Worcester unwalkable and uploaded an album to Facebook with the name “pedestrian observations from Worcester”) I took trains. I probably would have taken trains from New York to Worcester for the weekend rather than day-tripped even if they were fast – fast trains would still be doing it in around 2:30, which isn’t a comfortable day trip for work.
Non-work trips are the same – they can be day trips or not. Usually recreational day trips are uncommon because train travel is expensive, especially at current Amtrak pricing – the Regional costs twice as much per passenger-km as the TGV and ICE, and the Acela costs three times as much. But there are some corner cases: I moved to Providence with a combination of day trips (including two to view apartments and one to pick up keys) and multiday trips. Then, at commuter rail range, there are more recreational day trips – I’ve both done this and invited others to do so on Boston-Providence and New York-New Haven. Recreational rail trips, regardless of length, are more often taken by individuals and not groups, since car travel costs are flat in the number of passengers and train costs are linear, but even groups sometimes ride trains.
We can turn this into a two-dimensional table of the various kinds of trips, classified by purpose and whether they are day trips or multiday trips:
| Purpose | Day trip | Multiday trip |
| Business | Work trips between city centers, usually in professional services or other high-value added firms, such as lawyers traveling between New York and Washington, or tech workers between New York and Boston; academic seminar trips | Conference trips, more intensive work trips often involving an entire team, and likely all business trips between Boston and Washington or Boston and Philadelphia |
| Tourism | Trips to a specific amenity, which could be urban, such as going to a specific museum in New York or Washington or walking around a historic town like Plymouth, or not, such as taking the train to connect to a beach or hiking trail | Long trips to a specific city or region, stringing many amenities at once |
| Other non-work | College visits, especially close to the train stations; meetings with friends within short range (such as New York-Philadelphia); advocacy groups meeting on topics of interest; apartment viewings; some trips to concerts and sports games | Visits to friends and family; some moving trips; some trips to concerts and sports games |
| Commute | Long-range commutes | — |
Day trips and different peaks
The different trips detailed in the above table peak at different times. The commute trips peak at the usual commute times: into the big cities at 9 in the morning, out of them at 5 in the afternoon. But the rest have different peaks.
In particular, multiday tourism trips peak on weekends – out on Friday, back on Sunday night – while multiday business trips peak during the week. This is consistent enough that airlines use this to discriminate in prices between different travelers: trips are cheaper if booked as roundtrips over Saturday nights, because such trips are likely to be taken by more price-sensitive leisure travelers rather than by less price-sensitive business travelers.
Nonetheless, even with this combination of patterns, the most common pattern I am aware of on intercity rail, at the very least on the Northeast Corridor and on the TGV network, is that the trains are busier during the weekend leisure peak than during the weekday business peaks (except during commute times). Regardless of the distribution of business and leisure trips, leisure trips have a narrow peak, leaving within a window just a few hours long on Friday, whereas business trips may occur on any combination of weekdays; for this reason, leisure-dominated railroads, such as any future service to Las Vegas, are likely to have to deal with very prominent peaking, reducing their efficiency (capacity has to be built for all hours of the week).
But then day trips introduce additional trip types. The leisure day trip is unlikely to be undertaken on a Friday – the traveler who is free during the weekend but not the week might as well travel for the entire weekend. Rather, it’s either a Saturday or Sunday trip, or a weekday trip for the traveler who can take a day off, is not in the workforce, or is in the region on a leisure trip from outside the Northeast and is stringing multiple Northeastern cities on one trip. This day trip then misses the leisure peak, and usually also misses the commuter peak.
The business day trip is likely to be undertaken during the week, probably outside the peak commute time – rather, it’s a trip internal to the workday, to be taken midday and in the afternoon. It may overlap the afternoon peak somewhat – the jobs that I think are likeliest to take such trips, including law, academia, and tech, typically begin and end their workdays somewhat later than 9-to-5 – but the worst peak crowding is usually in the morning and not the afternoon, since the morning peak is narrower, in the same manner that the Friday afternoon peak for leisure travel is narrower than other peaks.
The day trips are rare today; I’ve done them even at the New York-Providence range, but only for the purpose of moving to Providence, and other times I visited New York or even New Haven for a few days at a time. However, the possibility of getting between New York and Providence in 1:20 changes everything. The upshot is that high-speed rail on the Northeast Corridor is certain to massively boost demand on all kinds of trips, but stands to disproportionately increase the demand for non-peak travel, making train capacity more efficient – running the same service all day would incur less waste in seat-km.
The issue of fares
All of this analysis depends on fares. These are not too relevant to business trips, and I suspect even to business day trips – if you’re the kind of person who could conceivably day trip between New York and Washington on a train taking 1:45 each way to (say) meet with congressional staffers, you’re probably doing this even if the fares are as high as they are today. But they are very relevant to all other trips.
In particular, non-work day trips are most likely not occurring during any regular peak travel time. Thus, the fare system should not attempt to discourage such trips. Even commute trips, with their usual peak, can be folded into this system, since pure intercity trips peak outside the morning commute period.
The upshot is that it is likely good to sell intercity and commuter rail tickets between the same pair of stations for the same fare. This means that people traveling on sections currently served by both intercity and commuter rail, like Boston-Providence, New York-New Haven, New York-Trenton, and Baltimore-Washington, would all be taking the faster intercity trains. This is not a problem from the point of view of capacity – intercity trains are longer (they serve fewer stations, so the few stations on the corridor not long enough for 16-car trains can be lengthened), and, again, the travel peaks for multiday trips are distinct from the commute peak. Commuter trains on the corridor would be serving more local trips, and help feed the intercity trains, with through-ticketing and fare integration throughout.
Pedestrian Observations 
How to offer fare that different for different time and different travel purpose, while maintaining flexibility of just hoping onto the next train coming instead of airlines style ticketing?
Also, I think many operators of high speed rail charge significantly more than conventional rail, not because they want to encourage riders use conventional rail instead of high speed rail, but is because fare for conventional rail were being capped politically that they cannot hike them easily, whereas for high speed rail, at the start of service, rail companies are free to set the fare as high as they wish to recover the cost
The correct answer is to abolish assigned seating, along with every other practice of trying to make Amtrak cosplay as an airplane.
The practical answer, which is observable today in Japan, is to have the first however many cars be unreserved seating. This would be your hypothetical Acela Coach Class and could be the Regional Coach Class today if Amtrak adopted a normal fare structure or even actually adopted airline fare policies to the extent that it would actually sell an unfilled seat on the train departing in 10 minutes at the base price point instead of insisting that the price is still (quickly checking Amtrak’s ticketing system as I write this post) $261 and perfectly happy to run that seat empty instead of selling it at the “wildly discounted” rate of the $73 that ticket costs on the same train six weeks from now. An actual fares-as-demand-management system would gladly sell that ticket for $73 or even $40 because both of those numbers are infinitely larger than $0; but because Amtrak is not actually interested in filling every seat, it leaves the station empty in the hopes that maybe someone in a desperate situation with unlimited budget will capitulate and fill the seat halfway up the line at $164 and the consolation prize that at least they don’t have to report “giving away” a 70% discount.
Who manages two people wanting to buy the $8 seat ten minutes before departure? Amtrak or do they slug it out in front of the ticketing windows? What happens when it’s because Metro isn’t running and there are a dozen people competing for the single seat? And one of them decides to buy it at the higher price 12 minutes before departure? This all works out wonderfully in railfan fantasies but the real world would be different.
In the real world, more than one seat is actually empty because in the real world, people live in multiple different places throughout a region, and 30 people boarding a sold out train to New York in New Carrollton Maryland means that either 30 people also got off there or that 30 seats were empty when the train left Washington Union Station.
It is only in the world of your personal delusions that people living in Alexandria Virginia or Stamford or Trenton or or any number of other places that aren’t exactly Boston/Manhattan/Philadelphia/Washington will not want to board the train in the place where they live.
Half the time they are boarding a train where-they-don’t-live. On their return trip home.
There’s gonna be more than a few people hoping to catch a train to New Carollton when the Orange line isn’t running, for whatever reason. Who manages the mayhem for the cheap tickets. A few times a year because the other days of the year the Orange line is running and people don’t want to pay a premium fare?
Trains should not sell out. Maybe once in a rare while there is a special event, but in general you have plenty of track capacity and so if the trains are even close to selling out you should buy another train (or more train cars if there is platform space for longer trains) and run more. If you are running 4 trains per hour (which the NEC should be able give) you should generate more than enough money from fares to pay for all the new tracks and trains you need just be the additional riders.
Where are talking about the Northeast here – there is plenty of population. If we were talking about a Kansas city to Winnipeg train or some other marginal line that won’t get an ROI that we might build for political reasons – I could see it just filling 2 trains per hour, but not enough above that to that we can afford to build the tracks needed to build the additional passing track needed to run more, or expand the platforms to support more cars. But this is this is the NEC, a good return on investment on running high speed trains frequently should be trivial to pull off with minimally competent management.
The cost of running an empty train car is very different from the costs of an empty airplane.
a lot of trains in Japan are now moving to 0% unreserved seatings, to maximize capacity for seat reservation.
Of course, Japan isn’t perfect and is just as capable of taking steps backwards as everywhere else on planet Earth. That they are eliminating unreserved cars to maximize reservations does not mean it is a good idea to do so nor does it mean that it was a bad idea to have such in the past. We can learn from the things Japan does well without having to similarly adopt the things it does badly, nor do we have to make its same mistakes.
It’s worth considering why Japanese operators feel comfortable/justified in making this change:
@John, unless the train I want to catch is full, and then there is an enormous penalty if the trains are all-reserved.
If you are traveling along it doesn’t really matter which seat you get. As long as there is a seat in your ticket class you are fine. Are if you traveling as a group you really want to get seats all together, and then you want reserved seats. If the train is less than 70% full this isn’t a big deal (I’m not sure the real number 70% feels right but if someone has better information I won’t argue – the real number depends on group size) – odds are you can find seats together, and if not you can get close and ask a few people to move. However as the train gets fuller you need to have reserved seating just for groups.
Most system should be running empty enough that you don’t need assigned seating – large groups will just learn that they should not travel at the peak times. There is plenty of space on other trains as most systems have plenty of track capacity to add more/longer trains if they start to get full. However when you run up against limits of track capacity assigned seating lets you get closer.
Not allowing unreserved passengers is super dumb. What happens if I miss my train and still want to get home and all the trains are full?
Depends on the implementation. On the Chuo Line limited expresses, for example, it’s a European style setup where you can still ride without a reservation (座席未指定券), either standing or using vacant seats whose occupants are yet to board. On Shinkansen services like the Hayabusa, you can get a specific ticket for standing (立席特急券) if the reserved seats sell out.
Ok but if I miss my Shinkansen at a busy time then I am not getting home that day.
That’s a terrible experience – far worse than having standing passengers on the train.
There are also seatless reserved ticket for trains without unreserved seats, which allow using unoccupied reserved seats until people who reserved the seat come
@Phake Nick, OK that sounds decent.
Corollary to the above post: in Washington DC the Regionals wait around for an hour to change engines and crews before continuing on to the next station. If, for whatever reason, the local metro in Washington DC is experiencing problems and there is an Amtrak train with an empty seat going to the same place that is sitting in the station leaving in 10 minutes, selling a ticket to the next station for $8 is just about the closest thing to “harmless” for the bottom line that you can get; even on a sold out train, some riders are boarding at those next stations thus the seat could never be occupied by a higher value rider traveling to a farther destination anyway.
Yet, Amtrak will still insist this ticket is $26 or $40 or the ticket counter will point you to the Metro entrance knowing fully well that the only reason you’re even trying to pay $8 to take the equivalent of a $2 subway ride is because Metro is down. Alternatively, at the other end of the line, Amtrak will refuse to book any fares to Boston from Route 128 or Back Bay because of the presence of a parallel commuter operation even as plenty of seats empty out (or leave Boston empty) due to traffic bound for those stations specifically.
Hey Alon, you once mentioned perhaps doing a video on the problems of pricing Mass Transit for families since a nuclear family using reduces the inefficiencies and cost of private motor vehicles.
What should operators do? Off-peak family discounted tickets alongside traditional children discounts?
This also applies to small groups of adults.
I’m not sure what Alon things, but I have long said that all local transit systems should be rated on % of riders using a monthly family pass. The pass means they are not paying for an individual trip, but for the right to ride when they feel like it. The family means two adults and all dependents living at the same address. (dependents may be elderly parents, or adult kids still at home – while it is cheating somewhat it is better to accept this than try to come up with fair rules). Local means service in a single CSA. You are allowed to have zones, but zones always are around your address. If you have fare gates or other check in systems then the computer should always adjust things to give your family the lowest rates that apply.
If you travel to a different city than the cities should work together to ensure your fares are either prorated or just covered by the other city pass.
Which is to say there is a easy to understand maximum price you pay for your family and the closer to home your family stays the lower your costs.
Amtrak is cross CSA though. People don’t ride them often enough in general to need a monthly pass. So I’m fine with charging per trip. (they should offer unlimited passes to frequent travelers, but I don’t expect the majority to use them).
I fully understand why making local transit free is politically toxic.
This proposal might as well just be “make local transit free.”
Actually, this is even worse than just making transit free – if you make transit free, then everybody just pays for the cost of the pass as a part of their property taxes or their rent, you rip out the fare gates and you’re done. No need to worry about cross-city prorating agreements, who counts as a “family member,” the computer program to adjust the rates or any other infrastructure, and you’ve arrived at the same intended goal of tying fares to properties instead of people.
The only benefit to a scheme like this is the hope that the smoke and mirrors will distract the usual crowd who shows up to oppose fare elimination by insisting that no no, we’re still collecting a fare, look at all these extra steps we’re taking to ensure that we’re not tagged with the dreaded label of “free.”
‘This proposal might as well just be “make local transit free.”’
If you are selling a monthly family pass, how is that making transit ‘free’? By definition things you are selling are not free. You could say the marginal cost of a trip on a pass is free (you don’t pay extra for that particular trip) or that extra trips beyond the break even point are free (you could theoretically ride for hundreds of times a month for the same cost as riding 50 times) but in either case the purchaser of the pass is still paying, not getting anything for free.
Every transit agency I am aware of, everywhere, offers some sort of yearly or monthly or annual pass – does that mean that every single transit agency is ‘free’? This proposal is no different, its just that instead of a single person buying unlimited trips for a month, it would be a group of people buying unlimited trips (presumably at a cost of more than two individual passes, but less than 3-4 individual passes).
There is a tendency in a lot of jurisdictions to offer free passes to youth, but I have always thought it should be the opposite. Instead, at the beginning of every school year students/parents should be given the opportunity to buy a discounted yearly pass for kids (there can be payment plans, further low-income discounts, etc.). During the same month, yearly passes for adults should also be sold (either the only time during the year yearly passes are offered, or at a special discount).
The idea is to build a habit of both transit ridership and buying a yearly pass each September when you have a captive audience (every teenager in the region lined up in one gym or another during registration). As the kids grow up they keep buying the yearly pass each September as adults, providing the agency a larger more stable revenue stream, while also encouraging greater transit use (and thus support) among the general population since people with passes ride more often than people paying each ride.
The description of the scheme with its use of the term “rated on” and complex description of how the pass would work in practice does not seem to imply a purchase except in the most technical sense that I suppose filling out the enrollment form technically constitutes a purchase, but this sounds like “enrollment” into the local electric utility and not like the voluntary “enrollment” to an internet service provider.
And on a similar note, your own implication that a yearly pass would be available in September and at no other time along with the vivid industry of all the school children lined up in the gym to complete their registrations also sounds pretty mandatory and thus puts us right back at “could just be free, but instead has a bunch of extra steps and pageantry so opponents of free fares don’t notice.”
yearly passes are a bad idea unless there is also a monthly pass that works out to the same price after interest. A small city with a tiny system is getting close to $1000 for the yearly pas, for two and large cities close to $3000. That is more than most people can budget for, but across 12 months is reasonable (less than their car)
yearly passes mean you need to figure out how to discount those who move in the middle of the year (who with moving expenses have less budget for this surprise). you also need to refund those who move away in the middle of the year. gyms that make their money on people who sign up for a year and never go can treat customers like garbage, but a transit system that wants people to ride needs to do better.
september is when parents already have to deal with back to school expenses so one more reason not to do then.
the goal is to set kids on a lifetime ride transit habbit so I’m fine with them free if their parents pay.
@TransitHawk “complex description of how the pass would work in practice”
@HenryMiller74’s description was: “The pass means they are not paying for an individual trip, but for the right to ride when they feel like it.” This is pretty simple, and is exactly like how every other type of pass for every type of transit system and many other types of non-transit and non-government owned services work – you pay a flat fee for unlimited use instead of a per use fee.
“does not seem to imply a purchase except in the most technical sense that I suppose filling out the enrollment form technically constitutes a purchase” Nothing about @HenryMiller74’s description in any way represents or suggests free transit. Nothing suggested that enrollment is mandatory, or even that you had to ‘enroll’ instead of just buying the pass a month at at time. His description litter uses the description ‘not paying for an individual trip’ so we don’t need to technically constitute a purchase, because ‘paying’ is the word that explicitly means ‘to purchase’.
“but this sounds like “enrollment” into the local electric utility and not like the voluntary “enrollment”
Except that enrollment into the local electric utility is voluntary, no one makes you do it, you can live in a house without electricity, or put solar panels on your roof and make your own electricity. In some jurisdictions there are actually multiple electric utilities you can pick from, with shared infrastructure but different rates based on electricity source (fossil fuel vs renewable) or level of customer service. Perhaps more relevant, even if you consider using a single local utility to be a de facto requirement as a natural monopoly, enrolling in their service is in no way “free” since you have to pay an electric bill every month, nor is it like a transit pass since you have to pay per kWh used, instead of paying a flat fee for unlimited power.
@TransitHawk
“your own implication . . . also sounds pretty mandatory and thus puts us right back at “could just be free,. . .””
Nothing about my proposal said anything about mandatory. I am suggesting making buying a yearly pass convenient (by offering it to all school kids at the same time), desirable (by introducing an unlimited transit pass to kids before they are old enough to drive, thus letting them experience how transit benefits them) and ‘sticky’/habitual (by offering discounted yearly passes to adults at the same time of year they are familiar with getting them as kids). This are not the mechanisms of a mandatory government program, on the contrary they are the marketing strategies of successful private businesses. Many such businesses are very good at gaining and maintaining customers for life, transit agencies should emulate them.
How could my plan ‘just be free’? I am specifically saying people should be buying yearly passes. If someone is giving you money for something, that is the absolute opposite of free. If you are thinking that transit should be made free by increasing taxes on everyone, then you are correct this is something that will be widely opposed, and could actually lead to lower transit funding if people vote to lower or scrap the taxes because they feel it does not benefit them because they do not ride transit. If, however, people choose to buy yearly passes because it is convenient, a discount/good deal, and a regular habitual expense, then they will not object to their own choice and the transit agency will end up with more money, and more ridership, and more political support because more citizens are using it.
@HenryMiller74
Re the cost of a yearly expense, I did specifically say “there can be payment plans”.
Prorating the cost of yearly pass is simple, make the cost a multiple of 12 and have customers pay for the remaining months. Or also offer monthly passes and let newcomers buy monthly until the yearly purchase window comes around. But 12 monthly passes should cost more than a yearly pass, the whole point is to encourage people to buy the yearly pass during the sale period. It might seem logically counter-intuitive but offering a “deal” part of the time gets more participation than a lower price always. Businesses have known this forever which is why special sales (Labor Day Sale, Black Friday, etc.) are so common. The occasional business that counter-programs by pitching ‘low prices year round’ only proves the rule.
Your point about people moving to the area does make me think that my proposal shouldn’t be to only offer the yearly pass once a year, but offer it at a discount then.
People moving away is a more difficult issue, but your own plan involved people paying for a monthly pass that was transferable to other cities, which is much more difficult than just offering a refund for a yearly pass turned in early.
If you want to build transit as habitual, giving kids free on a monthly family pass doesn’t help for two reasons. 1) If the kids are not involved with purchasing the ticket as teens (even if just with money their parents give them, or with a voucher provided as assistance) then its not something ‘sticky’ to build a habit on. You want them thinking that “its September so its time to get my transit pass.” Programs that have some level of involvement (‘Look for the gold sticker and put it on your order form’, etc.) have much higher response and follow through rates than things given away for free. The first is perceived to have value and important to do, the other is disposable and not worth the time. Foreign aid organizations have learned this and often switched from giving away things like mosquito nets or birth control to selling them for a nominal fee.
2) The problem with ‘free kids then family passes’ is that people go through a stage in life when they are not kids but don’t have a family. If they are not used to buying an individual pass for themselves then you will loose them as young adults and they won’t pick up the habit of buying the family pass when they are older. ”Buy your yearly pass for a cheap price on September 1st” is something that can apply to older kids, to single adults, to families with young children (candidates for buying a group pass like you described), or families with older kids buying multiple individual passes. The habit is lifelong and never changes.
a lot of options exist and I would not expect every transit agency to choose the same ones. some cities don’t need zones which simpifies things. Some cities have several operators which makes things complex. Often a complex fare system con be simplified to ‘I can ride this far and no farther unless I pay more’ or ‘i can only ride these colors unless I pay more’
the important take aways is I want the adults to have the transit pass even if they drive everywhere. That gikes hope they will eventually try transit. Of course a single child family (very common) will probably buy the kid an adult pass in that case but the ‘kids ride free with the parents pass’ means some parents will have the pass the monthly pass and this no reason not to try transit. Getting someone to try a new habbit is the hard part and this gives time to do that.
figure out your scheme. Just make sure you cap the monthly price so you can advertise never pay more than x. x should be less than a used car payment or average gas prices or something else that you can use to show transit is affordable.
@Onyx: The first line of Henry’s explanation, the only one you quoted, was fine. It was the entire rest of that first paragraph, in particular the sidebar on dependents and the “Local means service in a single CSA. You are allowed to have zones, but zones always are around your address. If you have fare gates or other check in systems then the computer should always adjust things to give your family the lowest rates that apply.” that takes it out of the realm of what would have otherwise been a perfectly normal proposal to sell discounted monthly passes and into the realm of “feels like a weird scheme to avoid just sending a bill to a house.”
If you’re just trying to sell a discounted monthly pass to a spouse after the first household member buys one, you can just do that. If you’re trying to get kids on transit, you can – and many cities already do – just issue them student passes.
If you’re trying to get buy-in from the whole household, or more pertinently, if you’re trying to ensure that everyone benefiting from transit access is paying into the system, we have a tool for that already too: it’s called a tax. You should either tax everyone for transit or you should just sell passes. You should not try and invent some weird scheme to make students and their parents all sign up together at the start of the school year or some other weird scheme to issue “family passes” based on home address.
Either sell a monthly pass and say every additional pass you add to the order is 50% off but you have to buy them all right now, or rip out the faregates and charge everyone a transit operations tax. There are benefits and drawbacks to either of those two paths forward. A path that tries to present the transit operations tax in the form of a pass sale drive has all the drawbacks of both with the only benefit being that you can say “look, they’re passes, that means it’s not free transit.”
@TransitHawk
You have a fair argument that @HenryMiller74’s proposal was too complex, but none of those details have anything to do with your incorrect assertion that his proposal in any way related to “free transit.” Lots of transit systems have zones or distance based fares, or different fares for seniors/students/tourists (i.e. more expensive but bundled with museum access), etc. Lots of systems also now have fare cap policies/software that charge you by the ride until you pay as much as a monthly pass and then the rest of your rides that month are free. In other words lots of systems have features similar to what you refer to in his comment (zones, dependents, ‘adjust to the lowest rate’). None of these systems are a-way-around-saying-its-free-transit because all of them require you to pay to use them.
In much the same way I don’t care how complex it is or how he phrased it, you cannot say his proposal is avoiding sending a bill to a house when he literally said “they are not paying for an individual trip, but for the right to ride when they feel like it” – ‘paying’ by the customer means ‘sending a bill’ by the agency. You state that the options should be a tax or just selling passes – HenryMiller74 was very clearly advocating just selling passes, since ‘the right to ride when they feel like it’ is the exact definition of a pass.
Your terminology/wording may be different, but your repeated references to a discounted/50% off pass to every household member after the first is just restating HenryMiller74’s position. I don’t know if 50% was his intended discount but qualitatively if not quantitatively you are suggesting the same thing as him. You can’t complain about issuing passes based on home address and then advocate for discounts for further household members – home address is what defines household members. You demonstrate an astonishing lapse of logic in suggesting to do something that is functionally identical to something you just criticized as the wrong thing to do.
Re: student passes, the objective is not to get kids on transit but to get adults on transit by building lifetime riding habits, including regularly purchasing a pass. I have explained how free student passes can counter-intuitively work against this goal.
Note: I never suggested that anyone “make” students and parents sign up at the same time. I am suggesting encouraging and rewarding buying a yearly pass at the same time every year, tied to an prominent recurring event, to make pass buying an activity that everyone remembers and most people want to do every year, thus maximizing both ridership and stable, recurring transit revenue. There is no need to provide a discount for families, each pass can be priced and sold independently (again, with a cheaper cost for students as the de facto “family discount” and vouchers for low income groups so that they don’t have to pay face value, etc.).
I took the liberty of repackaging Henry’s idea into something much simpler that can be easily programmed into a vending machine, which is where the simplified proposal of “buy a pass and be offered a second pass for 50% off if you say yes and pay right now” came from. And in actuality, the difference is that in this scenario, it’s just a product on sale that you can buy whenever you want.
If you can buy the pass whenever you want, or be automatically opted into the pass once you hit a fare cap, that is fine. I can get on board with that. But if you want to discuss “astonishing lapses of logic,” I’d suggest you start with your initial reply, where you suggest and I quote:
You cannot use the literal words “captive audience” and evoke the kind of imagery that “every teenager in the region lined up […] during registration” conjures up, then turn around and say “oh no, it’s actually completely voluntary.”
You can’t have it both ways. You either want to compel people to acquire the right to ride whenever they feel like it, or you don’t. If you want to compel people to acquire the right to ride whenever they feel like it, “encouraging and rewarding buying a yearly pass at the same time every year, tied to an prominent recurring event, to make pass buying an activity that everyone remembers” and the corollary of punishing everyone who didn’t is a great way to go about doing that while consoling yourself that everyone you rounded up into the gym had a choice to walk right out the door and accept their consequences.
Personally, I don’t have any happy childhood memories, so I guess the problem is me, but mandatory school assemblies and “optional” school events, including things like elective selection and getting permission slips signed? Those certainly are not things I look fondly back upon. I can’t imagine I’d look fondly back upon Transit Pass Purchasing Day, either. I’m also not a parent, but being handed a Transit Pass Order Form along with all the other paperwork I’m sure to be bombarded with as part of the start-of-year process is also not something that I can imagine would spark much joy in me if I was. Actually, I would look at the whole song and dance and ask myself what the point was. I might even say “isn’t this just a tax with extra steps?”
And then I would realize that it was, and either become an extremely vocal proponent of just rolling it up into tuition or my tax bill so that there was one less obnoxious annual to-do, or I would become an extremely vocal opponent of this thing you claim is an optional purchase but is actually an experiment in using timeshare sales tactics to pressure people into opting in to a “voluntary tax.”
Just be upfront and issue the tax, or just let people choose when and how to pay their fare.
My proposal was complex because there are many options and there is no one size fits all. Each city needs to figure out the right answer. A transit severing a small rural town can have a very simple system and thus a simple fare structure. A system serving a large metropolitan area should have a much more complex system and will probably conclude for various reasons that a simple fare structure is wrong. Each region needs to figure out what works for it. Some systems will be more complex than others, no system will support all the complex options I outlined. If you live in a large metro area then you will need to figure out the system they come up with. If you design a fare structure for a metro area you need to make the system as simple as possible – but no simpler.
A good chunk of people travelling by train in Switzerland have some form of monthly or yearly pass. This applies to both inter-city and regional trains (there is no fare difference between them).
Commuters can have a pass for a certain subset of the network that covers their commute. In Zürich, there is an off-peak version that is only valid starting from 9 in the morning. There are also passes for young people who are only valid after 19:00 (to go out). You can also get a pass for the whole Swiss network, it is quite expensive. Some companies offer them to their employees to simplify travel expenses (it’s a nice perk).
One pass that is quite popular basically gives you half fare on the whole network (soldiers in uniforms get that one automatically). Kids can have yearly passes that are very reduced, but only valid if the kid is with a guardian.
Generally steep discounts for kids is a good way to encourage families, I remember traveling with the TGV in France, and paying 5€ for my 5 year daughter in first class. Infants being free is also a nice when compared to airplanes…
Yeah we need detailed price/performance studies across the various pay traditions. Particularly the Central European pass/proof-of-payment tradition versus East Asian tickets-and-faregates. The most obvious trade off is in profitability i.e. SBB and OBB would look quite troubled by JR or Kintetsu’s standards. But the former two are public corporations with hard budget constraints and the latter are for-profit companies.
I don’t think the tradeoff is profitability at all.
Ignoring politics for a moment, there’s no reason why you can’t throw tons of taxpayer funding into a profitable system of faregates, cheap single tickets, and low value unlimited passes. The only reason why you might not be able to increase fares until region-wide profitability in a subsidized system of proof-of-payment, expensive single tickets, and high value unlimited passes system, is that you might drive so many people towards driving that farebox recovery peaks then goes down.
With that in mind, the fact that JR East or Kintetsu even make a profit is an argument in favor of faregates, cheap single tickets, and low value unlimited passes. They are in a sense playing fare collection on hard mode, being forced to convince people they are directly getting more value from transit than the transit costs to provide.
Amtrak sells monthly tickets. Philadelphia-New York and Cornwells Heights-New York are supposedly popular.
The leading metro areas for transit usage like Tokyo, Osaka, and Hong Kong all have relatively limited unlimited passes that offer relatively little discount from single fares.
High value transit passes are forcing people who derive relatively little value from transit to cross subsidize people who derive a lot of value from transit. It’s actively driving away people thinking about taking transit, to reward people who already use transit without thinking.
Really, instead of expanding transit passes with convoluted rules, just make single tickets reasonably priced. If there isn’t funding for reducing the price of single tickets, find it from making transit passes more limited and/or more expensive.
@Sassy
I am not familiar with fare details in Japan/HK, but their high ridership combined with low use of passes should give pause, similar to how all of the highest fare recovery ratio agencies in the US use distance based fares, even though they are more complex to understand than flat fares or simple zone plans.
On the other hand, cause vs effect could be reversed here. The metros you mention are all extremely dense and owning a car is very expensive (high taxes and fees in Japan, extreme parking costs in HK due to lack of space). In that environment you will get high transit ridership regardless, and you don’t need the carrot of an unlimited pass to encourage non-commute trips. Because ridership is so high, you can also keep individual fares low because the volume provides adequate revenue.
The fact that, as Alon has noted several times, ridership on Deutsche Bahn went up when they introduced the flat rate monthly D-Ticket, and that only 5 million out of 11 million subscribers had previously had a monthly pass, suggests that passes not no decrease ridership in all cases.
Japan is also a great example of why fares should be distance based. Distance based and non-radial zone based fares are the only fare systems that really make sense for local/regional transit that covers a wide area. Non-radial zone based fares are almost as complicated, and far more convoluted in presentation, vs distance based fares.
In addition German fare and fare collection not only punishes irregular transit riders by having expensive single ticket fares, but makes paying for fares with single tickets annoying and convoluted, requiring the use of vending machines instead of just gates/validators. Making transit cheap can of course increase transit usage, however, would the cheap monthly pass actually outperform cheap, easy single ticket fares? The D-ticket monthly price exceeds the average monthly per person spending on transit in Japan by a pretty decent margin, and exists within an environment of expensive and hard to purchase single tickets.
The regions with the least car usage in the developed world basically all have transit systems with distance based fares, unlimited passes that offer relatively small discounts and just on relatively small parts of the system, and primarily use NFC/RFID payment. Instead of trying to argue why the German system of convoluted, punitive, hard to use single fares is better despite performing worse, it would do well to recognize the German fare system for what it is, actually so terrible it is exactly the problem that the “solution” of wide area unlimited passes was searching for.
Japan has extensive point-to-point passes, basically you get 1 month to annual pass between two stations (every linear point inclusive). Its relatively simple and actually encourages ridership since it means your work pass subsidises your off-work trips to city centre or getting to city centre for inter-city terminals. Either its a different management context, if DB breaks even its doing fine, if JR East breaks even, its doing something wrong.
Parking is expensive in Japanese relative to elsewhere yes, but parking still has de-facto tax subsidies which matter outside megacities (nothing huge just parking is taxed less than business or residential). Japan has areas with 60-80% car modal split such as Hamamatsu, Okayama or Niigata. Even if you have the most pro-transit/anti-car regulatory/taxation/TOD framework not having an appropriate operations-infrastracture mix will kill you. “Run EMU back and forth frequently between places want to go to” is the basis of a 1st world mass transit system.
@Borners
Hmmm, if Japan uses point-to-point passes like you say, then that would seem to undercut much of @Sassy’s point about single tickets. A point to point pass for a distance based fare system is relatively analogous to a unlimited use for a flat fare/zonal system.
@Weismann made the point above that passes are very common in Switzerland, and their mode share is great too.
I’m questioning passes less than I did when @Sassy first brought up his single-ticket point.
Tokyo has monthly unlimited transit passes, but they are notably more limited, as Borners mentions, being only for a fixed route, but also significantly less of a discount. For example, prior to D-ticket Berlin AB monthly broke even vs single tickets at 21 trips, and with D-ticket, 14. Standard monthly tickets in Tokyo break even at 30-41 trips per month.
People in Tokyo end up paying for single trips or single day tickets a lot more often than people in Germany, partly because commuter passes are expensive with the median commuter pass being almost twice that of D-ticket, but also because single tickets are cheap. A Tokyo Metro base fare, considered on the expensive side in Japan, is well under half of a Berlin Kurzstrecke, and a Tokyo Metro 1-day fare is just a bit more than a Berlin zone AB single ticket.
When I was an infrequent transit user (digital nomad) in Freiburg, transit felt kinda pricey even on US salary (though that was pre-D-ticket). As an infrequent transit user (walk only commute) in Tokyo, transit feels reasonably priced even on a local salary. If I was a transit commuter, that would be flipped, but most people are not transit commuters, even in regions where most people don’t drive.
D-ticket is good and all, but I still think infrequent transit users would be more convinced by a massive cut in single ticket prices, at least if the process of buying single tickets in Germany wasn’t so painful.
@Sassy
“prior to D-ticket Berlin AB monthly broke even vs single tickets at 21 trips . . . monthly tickets in Tokyo break even at 30-41 trips per month.”
“A Tokyo Metro base fare, considered on the expensive side in Japan, is well under half of a Berlin Kurzstrecke”
If Tokyo tickets are half the cost of a Berlin ticket, but Tokyo passes are twice the cost of a Berlin pass, then Berlin and Tokyo passes are roughly the same price (not sure if its exactly the same when accounting for purchasing power, etc., but close enough for a blog comment). This once again seems to undercut the notion that single tickets vs passes lead to better metro usage, although I would be open to investigating if making single rides cheaper while holding the price of passes constant makes a difference.
Once again, I have to wonder if Tokyo transit usage is so high because it is Tokyo (dense, not a lot of freeways or parking, huge rail network going everywhere) and thus transit operators can set a pricing/ticketing scheme that takes advantage of this (single tickets can be cheap because fixed costs are spread over so many riders, passes can be limited because commuters will use the train no matter what) rather than the other way around.
The point of a pass is to place a maximum price on transit. That is a loyalty discount for your users who ride the most. This makes budgeting easier for everyone – I will never pay more than $x so I don’t have to worry about costs of an trip. Even if individual trips are reasonably priced, they still add up.
There is one other advantage to a pass: if I buy an individual ticket to work at 9am every day and other home at 5pm every day (that is very typical commuter patter) – some of my ticket price should go to subsidize the 3 am transit that I might never take, but that fact that I can should I have to work late gives me confidence I can take transit for the regular trips. Similarly even though my I only visit my uncle across town on Christmas, all my regular trips on transit that don’t cross town should subsidize those routes I rarely use.
Most of the costs of a car are fixed (payment, insurance, license). The ones that are not are still per distance not per passengers. If you make even one trip impossible on transit that means someone needs to have a car (if it only one trip per year you can rent, but the costs of renting a car are generally high enough that most it quickly is better to just buy), once someone has a car the incremental costs of more trips in the car is low, and the convenience of the car going when and where you want to go is nice – soon you want to move to the suburbs where there is less traffic and plenty of free parking.
A pass ensures a maximum price, and because you have the right to ride anywhere anytime you are more likely to ride transit for optional trips.
Why should there even be a maximum price on transit? There isn’t a maximum price on plane tickets, gas, or tolls. I don’t think the people using transit the most should be given a major discount for “loyalty” since heavy users of transit need the least convincing to use transit.
Individual trips adding up if you take a lot of them, is a feature, not a bug. Why should a downtown office worker get to take transit anywhere they want for free, while a waiter at a suburban restaurant who bikes to work pay more for the few transit trips they make, and a waiter at a suburban restaurant who drives to work just be driven away from transit altogether?
I agree that downtown commute trips, the trips that make the most sense with transit anyways, should cross subsidize other trips. However, high value transit passes work directly against that. A transit pass that breaks even at just 7 round trips per month is leaving money on the table. An office worker that has 3 in office days a week makes 13 office round trips per month, and full in office 21. The transit system could be getting 2-3x as much money from the downtown office commuter if it offered less of a discount for unlimited passes. In fact, due to accounting simplicity and incentive to work in office, some companies might be willing to reimburse a 15 round trip commuter pass even for office workers that are expected to work in office fewer than 15 days a month (mine is btw), which is literally free money for the transit operator.
And longer downtown commute trips should cost more. Living deeper into the suburbs should result in higher transportation costs, both to discourage such lifestyles, and to capture more money from downtown workers that choose that lifestyle.
The best way to convince people to use transit when they could otherwise drive is by making transit better, and you don’t do that by turning away money from the people most willing to pay it. Ultimately, making transit free for everyone has shown to not significantly reduce car usage. Why would making transit free for only a fraction of the population while more expensive for everyone else, be expected to do a lot better?
And the goal should be to significantly reduce car usage. While transit modal share is a decent proxy for car usage, an improvement in transit modal share without a decrease in vehicle kilometers driven, is at best doing nothing to address the climate crisis. In real world situations with non-100% renewable/nuclear energy grids and non-100% electrified rail networks, it is actually making the climate crisis worse.
A maximum price is good customer service and advertising. Car owners never know when their car will break and need expensive service by placing a maximum price you can directly target that (if the bus breaks it isn’t your problem to fix it). It also rewards your more loyal customers who are spending the most. Those customers who ride the least are the most expensive as they will be asking questions, holding doors, getting on late, demanding to be let off early when on the wrong bus, and otherwise making simple mistakes that cost you money.
Selling single tickets is expensive. You need vending machines for each sale, servers in the backroom. Most people will likely be paying by credit card which has high per transaction fees for small dollar value transactions. Bulk transactions for multiple rides are thus cheaper because it needs so much less infrastructure and this savings should be passed on to encourage customers to buy in bulk.
Maximum price for family is good because not only does the car look attractive and then since you have it the parents won’t ride transit either. It is also good marketing to get kids using transit early, trying to establish a lifelong habit, but parents are less likely to let gets get rides if they cost money.
There are lots of options, but in general you can say that the lost income from a well priced unlimited pass is more than made up with goodwill.
To be clear credit card charges are pretty low in Europe. When receiving them myself the charges have been super low – like 1%.
People who choose to drive choose to do so despite driving being expensive and full of unpredictable costs. Transit is already much more reliable in terms of costs than driving. Literally making transit free, which is the most reliable cost, always zero, doesn’t significantly reduce driving. Why would making transit free for only a fraction of the population while more expensive for everyone else, be expected to do a lot better?
Reasonably priced single tickets are worth selling, considering almost all for profit transit agencies follow that model. The savings from selling fewer tickets is far from enough to justify the discounts those tickets offer in German style fares regions. With a transit card model, the cost of fare collection can actually be negative, as in through float and non-transit merchant integration, the fare collection system pays for itself and then some, even ignoring the actual fares collected.
In the real world, it doesn’t seem like the goodwill from cheap unlimited passes actually makes up for not being able to run as good as service, and/or not being able to offer cheap single tickets for infrequent transit users. Again, while the comparison is kinda shaky, the lowest car usage regions in the world have reasonably priced single tickets, and relatively poor-value unlimited passes.
If anything, it’s the goodwill from cheap single tickets that matters more. Frequent riders should already have tons of goodwill because they benefit a lot from transit.
I think the cheap German monthly ticket is getting more people to use regional trains right?
D-ticket did not lead to significant reduction in car use as per TUM report last year. Considering how fossil fuel heavy Germany’s electric grid is, and how much of German transit still relies on in-vehicle combustion of fossil fuels, significantly increasing transit usage without reducing car usage is actually counterproductive.
D-ticket was overall a positive for public transport in Germany, but it’s hard to imagine the program is the best use for over 4 billion EUR in additional subsidy a year for impact to transportation choices and greenhouse gas emissions.
If making local/regional public transit free doesn’t lead to a significant reduction in car usage, it would be expected that making local/regional public transit 49 EUR per month also doesn’t lead to a significant reduction in car usage. Many parts of German public transit are quite a bit better than Luxembourgish/etc. public transit, so there is reason for optimism, but it seems like the optimistic scenario isn’t the one actually playing out.
Re: families, I know that many Scandinavian trains have “kinder cars” designed for families. There are amenities like a larger bathroom with a baby changing table, a small number of seats are given over to a playroom with a small slide and toy trains, etc. I wonder how they handle pricing for families?
Larger bathrooms and pram spaces also double up as wheelchair access etc. Those are good.
But playgrounds are for luxury trains on low frequency services. You don’t want them anywhere near trunks i.e. great for rural Scandinavia or even Switzerland. Trashfire bad elsewhere. Just like dinner cars.
Pointedly for-profit rail companies in Japan are doing “barrier free” stuff while avoiding dinner cars/play spaces outside special services geared towards relatively low frequency geared towards luxury market.
Pre-pandemic, the Acela between DC and NY was much busier on weekday mornings and afternooons than on weekends. Currently is the other way around. Business day trips were the driving force. I suspect that a lot of those shorter business meetings are now on zoom.
I’m confused by the “intercity trains are longer” point. They’re not! At the least, the NJT NEC trains are 10-12 cars long, compared to the Amtrak NEC train which are 8 (that’s what the Airos will be) and the Keystone (5 cars). Do you mean that they could and should be longer?
Correct – Amtrak should be moving to lengthen all of its trains, especially where capacity is limited on the busy Northeast Corridor but also on the network of terrible once a day long distance trains that somehow sell out anyway. Unlike a plane, you don’t need to send a second one when the first one fills. You can just attach more train to the end of your train.
Tragically, this is forbidden knowledge to Amtrak, who aren’t even capable of buying enough spare cars to have a healthy reserve supply, nevermind enough to add more cars as demand requires.
(PS: Of course, lots of other reasons exist to run more trains. This comment is not meant to imply that we should make trains longer instead of running more of them. This comment is meant to state that we should make trains longer instead of demand managing, shrinking seats, or purchasing bilevels.)
“This comment is meant to state that we should make trains longer instead of . . . purchasing bilevels.”
Very much this. Everytime I look at capacity of bi-levels I am always surprised at how a single level train a few cars longer can carry the same or more people. Bi-levels lose so much capacity to stairs, lifts, electrical equipment closets in EMUs (that would otherwise go under the floor in a single level), space above trucks that cannot be doubled, etc. In the US context, just going to trains that are AAR plate width (10′-8″) versus the standard width for Bombardier bi-levels or base European models that are less than 3m/10′ can make a big difference by allowing for 3+2 coach seating (in a commuter context) or 2+2 seating for business/first class (in any context).
That attitude is all well and good if you can actually add carriages without modifying all stations that are serviced by the train. But it’s not sensible once congestion and capacity have been exceeded, especially to fulfil demand when it is at its peak. Most modern bilevels add 40% passenger capacity which is nothing to scoff at if it is an option, say compared to adding extra trains. Take SE England which has peak hour train hell, and train frequency has been reduced allegedly to improve reliability. Conventional bilevels cannot be used but a German design outfit (used to doing aerospace design) has a bilevel design that would fit existing tunnels and infrastructure but the Brits have rejected it. True, it increases capacity by only 27% due to the loading gauge constraints (imposing 2+1 seating on the top level) and costs more than standard bilevels because of the need for more precise engineering and hi-spec steel etc. But that would be extremely useful to millions of UK rail travellers where apparently they are out of other options and the cost looks trivial compared to the billions of subsidies to those private rail franchisees.
I’ve highlighted the relevant line here, bolding the most important eight words in the comment.
Purchasing bilevels is a policy failure. They’re your best option in this circumstance because – and only because – every other sensible option has been eliminated already. In actuality it isn’t impossible to run 27% more rail service on the lines where bilevels are suggested, nor is it impossible to make trainsets 27% longer. It may genuinely be impossible to re-nationalize, of course, and HS2 shows us that embarking on a singular transformative project seems just as impossible on both sides of the pond.
However, a 27% increase in train length means adding two more railcars (or three if the train is already 9 cars long) and possibly a guard to control for if the new cars end up hanging off the back ends of the platform. This is, of course, just a matter of putting up some jobs listing and adjusting the number of railcars you’re intending to buy upwards. Similarly, 27% more train service is simply a matter of extending peak hour service levels further into traditionally “off-peak” times, or adding a single (1) additional train per hour. Again, just a matter of hiring some more staff and adjusting up your next railcar purchase.
Of course, I’m a stupid American and don’t know about UK trains, so I’m just speculating that there’s probably 27% of additional train service you could run by simply reverting to pre-COVID schedules or enough capacity in places that see 2 or 3 tph right now to add one more. But because I’m a stupid American, I also see the push to purchase bilevels through the lens of stupid American railroading decisions, where trains are routinely shorter than the platforms they service and where 100% more service everywhere in the country that is not the Northeast Corridor means adding three, five, or six additional departures… daily. (In fact, for a depressingly large swathe of the USA, adding one additional daily departure would be enough to double service.)
If you can’t run “peak service” for a longer stretch of the day, can’t extend the off peak operating hours any earlier or later than they currently are, and you also can’t add any more train to your trains, then yes, bilevels become the most sensible option. They are still a bad option and still a policy failure: how can trains genuinely be full enough that something must be done to add capacity, yet every sensible option to add line capacity instead of shoving more seats in is deemed impossible?
@Michael, you aren’t having the 45 second stops with bi-levels that you get on commuter trains in Britain, so the capacity doesn’t help you much when the dwell time is important.
Plus in general the railway lines into London itself are pretty busy and that is where a lot of these people want to go – so it’s not totally straightforward. I am however sure with more competent planning that there are a bunch of cases, especially away from London, where more capacity could be added relatively easily.
@Matthew Hutton @Transit Hawk
You guys are such jokesters. And it’s good to see you aren’t the least bit inflexible or dogmatic on these matters.
/sarc
But it does seem to be an Anglospherian railfan thing. I reckon this somewhat illogical, seemingly emotional aversion correlates with a dysfunctional approach to rail in general. None of the Anglosphere have HSR (I don’t count HS1, part of the Eurostar link.) And their long-distance rail, and city transit is nearly as bad. Though the US has plenty of bilevel commuter trains and Sydney has had them for more than 60 years.
We know that the only reason the UK hasn’t deployed them is their network’s PG1 loading gauge problem. The reason for HS2 is mostly to relieve capacity constraints on the WCML. Likewise the London commuter lines are beyond a joke, particularly <a href=”https://www.theguardian.com/commentisfree/2016/jul/19/southern-rail-failure-crush-unions”>the Southern network</a> which has among the highest commuter traffic in the world. TH should realise from what I wrote that they cannot increase train frequency to fix it (indeed they reduced train frequency because it was causing cascading network wide failures), at least not without massive investment in signalling and trains to which they are totally allergic. And they obviously have been at train length limits for the last century. They, like many rail systems worldwide, have staff recruitment issues that means adding extra trains is problematic independent of ability to sustain the higher frequencies.
When considering trains that would use HS2 but primarily serve parts of the non-HSR network north of Birmingham (eg. Edinburgh) they commissioned the German company to see if they could come up with a bilevel design that worked on all of the network. (There was no particular need for them on HS2 itself; the fact that continental bilevel trains would be able to operate on HS2 was not the issue; in fact all-new lines in the UK can take such trains–go figure, terrible policy huh?) Such a solution could provide real relief for a lot of the capacity-constrained network everywhere. The Germans produced a good design but it was rejected, for reasons I haven’t seen elucidated anywhere.
The argument about 45s boarding times is b.s. or worse (it is a bad-faith argument). Almost all these trains have most of their pax at the termini where it is a non-issue. Anywhere else it is trivial; unless of course you somehow believe pax on the top level are too dumb and unprepared to get themselves to the boarding level in time for their efficient exit. Which actually does seem the patronising or indulgent or self-serving attitude of rail managers in the Anglosphere.
@Michael, Britain and the US both have high speed rail. All the lines at 125mph and above that are improved existing lines meet the definition.
Also I don’t really think you can say British long distance service is bad. If it is bad I have no idea where you think is good – as there is nowhere where the long distance service is clearly and unambiguously superior to British service.
Certainly for long distance service where stops are longer than 45 seconds I do think bilevels are worth considering if they will fit on the classic network.
The sarcasm is unwarranted, Michael, in particular considering that you yourself admitted that the existing design only adds 27% capacity and that the aspirational goal of most modern bilevel designs is 40% – and I would also put forth that you are on the inflexible and dogmatic side here considering that your argument, just like every argument for bilevel carriages, hinges on first discounting every possible alternative such that your only remaining choices are “purchase bilevels” or “do nothing.” It’s easy to select bad options when the only options are bad options, and it’s also easy to claim inflexibility or obstructionism on the part of opponents to bad options when they insist that actually, no, we should consider good options instead.
Indeed, further on down in your own comment, you write:
And, indeed! I do realize that when it’s impossible to invest into signaling and impossible to invest into capacity and impossible to hire anyone, then “what if we found a way to cram more people into trains of the same length” becomes the “best” remaining option. I realize this and reject it, for probably the same reasons the proposed design for UK bilevels got rejected, for two simple reasons.
First: As is insinuated in the Guardian article you have linked from 2016, Southern does not actually have a capacity problem, it has a behavior problem; it’s reducing frequency to “improve consistency” because it cannot or will not employ enough staff to run a more robust schedule, and it is enabled by a UK government which is disinterested in holding it to account. The solution is as obvious as it is “impossible” – but actually, it’s not impossible, since you have crush loads of angry passengers demanding a resolution, which means that the will exists to do the “impossible.”
Second: If you’re allergic to spending money on new trainsets, this probably includes spending money on procuring and maintaining an entire parallel set of bilevel trainsets, which certainly aren’t going to cost less than purchasing more of the same stock you’re already buying.
You’re free to call it a dogmatic opposition or emotional aversion as much as you like, and you know what? It is. I’m very emotionally averse to terrible ideas such as purchasing bilevels in the hopes that spending money on actual problem resolutions might be avoided then.
However, you can’t call it illogical. There’s actually a very clear logic to refusing to accept a bad proposal in lieu of any good proposal, such as figuring out how to hire more rail staff.
I disagree with this definition of high speed rail. I do understand that it is one of the (many, competing) possible definitions of high speed rail but I just don’t think it’s that useful when several “conventional” electric locomotives exist that are capable of 125+ and when it’s more than possible to deliver a line that sees 125 mph operation over long sections but still manages to average half or less that speed. (Amtrak’s high speed Acela averages 70 mph even with its segments of 150 mph operation and substantially longer segments of operation between 100 and 135, which is the main reason I don’t consider it to be HSR.)
I believe actual high speed rail should be judged by average speeds, not top speeds on a given segment. I would put forth that the average speed should be 125 mph for a line to be considered HSR. However, I’m willing to take the lower 93 mph average speed definition that is sometimes used instead.
If you say that something has to average 125mph or 200km/h then only the French, Spanish, Japanese and Chinese have high speed rail as no one else hits that average between city centres.
Taiwan manages 124mph I believe so it might actually hit 200km/h depending on rounding.
I am surprised that the list is that small.
I don’t know if I’m surprised enough to adjust down to the more achievable 93 mph waterline.
@Michael
Please stop with that ridiculous “bi-level in British loading gauge” vaporware project. The interior dimensions are so tight that almost everyone is sitting in a gallery car seat (can’t stand up at the seat) but it’s worse than a gallery car because gallery cars only have that problem downstairs (that design would have it upstairs too), gallery cars at least get the 40% seat increase, and with gallery cars you can stand up in the aisles while that design was so tight some people couldn’t stand up straight to get to their seat. Face it, British loading gauge isn’t large enough to allow bi-levels.
It’s only a 27% increase but also “more expensive.” How much do you want to bet its more than 28% more expensive and thus more expensive on a per-seat basis and thus would require more a subsidy?
You claim the Anglosphere has a dysfunctional approach to rail, then you scoff at 45 sec boarding ties when 45 (or 30!) sec boardings at stops is exactly what non-Anglophere countries do regularly.
Having all pax get off at the terminus is even more of a problem with bi-levels because you end up with crowding as everyone tries to use the stair at once, instead of the “one person getting off walks downstairs before their stop” paradigm. Also, there are busy non-terminus stations in the UK (Newcastle, Birmingham New St, etc.) where this would be an issue, and Alon has noted how the RER had to reduce frequency to deal with this problem reducing the capacity benefit of bi-levels.
If you really needed bi-levels north of HS2 the proper solution is to upgrade the relevant lines to GC gauge like HS1/HS2 and other new build lines. It carries benefits for freight and non-bilevel passenger trains, and has to be done at some point, either through upgrades or building new ROW (i.e. HS2). Of course, there is no “capacity constrained network” north of HS2 because British cities get smaller and thus traffic gets lower as you go north.
The problem with Southern is that the Underground barely exists south of the Thames. If the Underground reaches Upminster, Epping, Cockfosters, Chesham and Uxbridge then it should reach Dartford, Chelsfield, Croydon, Woking and Staines. If it did, then Southern wouldn’t have among the highest commuter traffic in the world because the commuters would be on the Underground, like they are to the north of the Thames.
No thanks, not as long as you pour out the b.s. which is your description of the German design. The thing about termini is b.s. too. Perhaps the Knights & Lords who decide such things in the UK want zero compromise on their Rolls Royce service (!) but I’d like to see a poll conducted on the millions of passengers who have missed trains or those who have paid £3-4k for a season ticket to only have to stand the entire journey. Whether they think a 30% increase in capacity with a slightly reduced head height (for 22% of seats) is preferable, especially when 99+% of the time is spent seated and those seats are actually luxurious compared to the lower level (due to the 2+1 config.). “can’t stand up at the seat”: you mean like most airplanes especially billionaire’s Learjets etc (Concorde was notoriously tight too)? The design is far superior to American gallery cars (indeed I recall Chicago is buying the new Alstom RER bilevel?)
Get real about priorities.
Instead you propose:
After a century or more of this problem what sign of any of that wishlist happening? Your suggestions are absurd. It would cost more than building HS2–and how’s that going? And it would take forever. Essentially the bilevels won’t happen because of quite ridiculous purists such as yourself, and no real desire to improve the system.
As to the comments on RER-A, Alon Levy has no monopoly on wisdom on that. The fact is that any system trying to operate very high frequency trains combined with very high pax loads risks a single incident catalysing cascading chaos across the line. RER-A has the highest pax load of any commuter rail outside Asia, and the intensity is mainly a problem in a handful of central stations. The boarding issue is also exaggerated and largely solved by the new design of wider double-doors and large foyers etc. The fact is that RER-A successfully carries its immense pax load, which is apparently more than CrossRail can do right now with less than half the pax load.
And here’s the thing. The French have been working on solutions for at least a decade and they are about to kick in. RER-E will take a large pax load off RER-A when its new cross-town tunnel opens this year, and extended further north-west by 2026. Bilevels are being introduced to all RER lines. When the Paris-Lyon TGV got so popular they increased frequency (headway down to 3min at peak; OMG Onux how do they cope at the termini?) and doubled the train length, then they went to bilevels (which actually added 45% of seating). But you know, “poor policy”. As to your Underground extensions, Paris has 5 RER lines that do exactly that while London has one CrossRail (operating for one year versus 46y for RER-A and -B).
So, what is in prospect to solve all the capacity constraint problems in the UK system? Absolutely nothing. Including from you or anyone here. Wall to wall b.s. and denial.
Oh, and while talking about RER-A (which you brought up). As a mathematician Monsieur Levy never quite got to grips with the math:
(RER-A pax capacity per hour, standing + seated):
Train type………………30tph ……..25tph
bi-level MI09 trains: ….28,440 ……23,700
1-level MI84 trains: …..18,720 ……15,600
That’s still 50% more pax at the lower tph. The reason why Victoria line (or others with very high pax load) try to run 36tph is that they have no choice if they want to try to deliver the pax.
Here is another Pedestrian Observer (my abstract of his post; he’s talking of multi-branched RERs):
@Transit Hawk, another problem with a 125mph average speed requirement is that even in the countries that have it only some express HS trains achieve it. On the Tokaido Shinkansen the Hikari services only average 106mph and the Kodama only 78mph. I noted elsewhere in the thread that when the Shinkansen opened average speed Tokyo-Osaka was only 101mph. Better tell the Japanese it wasn’t real HSR!
In that regard, if Amtrak ran one train a day non-stop DC-NY in 1:48 would you suddenly consider it HSR because it now achieves 125mph average?
Instead, I’ll point out that I was and am willing to adjust down to the lower 93 mph average speed figure (in particular since it seems like there’s actually far less rail hitting the higher watermark here than I had believed) – but I will insist on continuing to use average speed as the benchmark by which to measure whether a line is HSR or not simply because it makes more sense to define these things holistically.
I could make a joke about time traveling to 1964 here as a lead-in to a comment on standards changing as technology improves, but I’m going to try to be less hostile to everyone other than adirondacker12800 going forward.
I really don’t think it’s all that useful to base HSR definitions on what the highest speed limit is on the line, at least, not without defining how long the stretches of top speed need to be to “count.” Otherwise it is very easy to introduce “HSR” at the prices such projects command but then put forward a train which only ever reaches its top speed for about 11% of its running route and cancels that by spending far more of its running route at far lower speeds.
To that end…
That run specifically and no other run, yes – the remaining Acela trains I would continue to consider not HSR.
I’d also consider it a spectacular waste of money to have that single Acela Nonstop; but considering my stated position on what level HSR has to be at to be worth investing in, that really goes without saying.
@michael, there is still scope the lengthen at least some trains in the UK so that can still be done.
And the other disadvantage of bi-levels is that the French service frequency is terrible away from Paris.
With regards to Crossrail London has the equivalent of RER B and D (Thameslink), RER A (Elizabeth line) and RER E (Metropolitan line) at this point. Plus while it doesn’t have an RER C equivalent it does have Overground which is better and Paris has no equivalent to.
Yeah, it’s not hard to lengthen them – the Acela stations from New York south are all much longer than eight cars, I think the shortest is Wilmington at 10 and then the next shortest is 30th Street at 14, both expandable to 16? New Haven is 8, expandable to 16 (nontrivially).
It’s “not hard” but they haven’t done it and if they want to do so they’ll have to buy new train sets, right? So what’s a realistic time frame for longer intercity trains? 10 years?
You can order standard train cars from major manufactures for delivery in two or three years. I believe the FRA requirements that made standard cars not viable were eliminated years ago, but if someone looks up regulations I’ll stand corrected. Getting all new trains might take 10 years, but really Amtrak should be buying a few trains/cars every year to ensure manufactures stay in business and have reason to invest in better manufacturing. Amtrak should be able to figure out which trains sell out most often and add more cars to those in 2 years, then trains that sell out sometimes and fix those in year 3, then in year 4 replace the older cars/trains that are less reliable…
Amtrak gets funding from Senators who think trains are a Communist plot to sap and impurify the precious bodily fluids of Real Americans(tm). Who drive everywhere.
As soon as they can buy (imported) rolling stock.
there’s nothing special about assembly lines in Germany, Switerland or France. It’s a lot cheaper to ship parts to an assembly plant in North America than ship whole cars across oceans.
The special thing they have is experience building trains in a competitive environment, rather than in a politicized environment. Remember, the one time NYCT bought trains without Buy America rules – the R62 – it imported them from Japan.
they’ve been getting stuff from Kawasaki in exotic Yonkers instead of Japan since then.
Yes, because Buy America made them. And this worked while the rolling stock remained conservative, but over time technology has moved on, and in the last 10-ish years there has been sharp growth in NYCT (and other American) rolling stock procurement costs, which has not happened here. NYCT has struggled to procure things that don’t look like R110s.
Things that fit in subway tunnels will tend to look alike.
The R211 cars …will fit on the “B” division. “A” division cars would fit”B” division tunnels but aren’t a good choice for revenue service. It’s unclear from the Wikipedia article what the cost per car will be.
https://en.wikipedia.org/wiki/R211_(New_York_City_Subway_car)
And even though Kawasaki is assembling them they will be using Alstom inverters and traction motors. It doesn’t say who is making the brakes. Hmm. Or the CBTC Hmm.
New York to Worcester for the weekend rather than day-tripped even if they were fast – fast trains would still be doing it in around 2:30,
Why would it be 2:30? I come up with 1:45-ish to account for the slower part between New Haven and Springfield. Or the slow part between Providence and Worcester. The high speed tracks are going to be there for other markets, no reason they can’t be used for secondary ones.
New York-Boston plus Boston-Worcester. Doing it via Springfield is about the same most likely.
New York- Boston via Springfield will be slower than New York-Boston via Providence because it doesn’t make a whole lot of sense to spend billions and billions of dollars to make Hartford-New Haven or Hartford-Springfield a few minutes faster. There are enough people along the Inland Route to have a train an hour that goes beyond New York. The East SIde of Manhattan is enough of an origin/destination that there could be enough demand for Boston-Grand Central once an hour. 30 minutes after the Penn Station train would mean things like Worcester-Hartford have every half hour…. Two more trains to amortize the cost of the tunnel to Framingham.
Railfans such as yourself love having bespoke trains on a million different routes and pitching once an hour or once every half hour.
Unfortunately, in the real world, there’s going to be some number of trains going between New York Penn and Boston South on the compromise alignment that is ultimately agreed upon and zero trains going between Grand Central and Boston South via anywhere, nor will there be any such “inland route” or parallel line or your choice of 27 different services. You will pick between taking one of several per hour Montreal-bound trains out of New York Penn and changing to a different several-per-hour train that stops in Worcester wherever it intersects with the Toronto-to-Boston line, or you will take one of several per hour Boston-bound trains out of New York Penn and switch to a local train to Worcester wherever such a local service conveniently intersects. Maybe that will be Providence. But it will probably be Boston.
It would be the same trains as the main routes. That can generate enough traffic to have a train. A side effect of having the ticketing system analyze things.
If there is enough demand for a Boston-Toronto route the train on the Boston-Montreal route can use the Toronto route to Albany and use the same tracks as the New York-Montreal trains to get to Montreal. Without making people change trains or running empty seats all over the place.
The Swiss, Germans, British etc literally have a million different routes pitching in once an hour or once every half hour.
@Adi, how do you get to 1:45 NY-Worcester? What route and what average speed/time per segment?
The high speed train uses the high speed tracks between New York and New Haven, the more conventional speed tracks between New Haven and Springfield and the high speed tracks between Springfield and Worcester. It’s more people than Providence and a bit less than the sprawl of Baltimore. On tracks that would be there for other reasons.
It’s New England, gotta do things to gin up demand to pay for the expensive tracks to places Not-New-England.
This wouldn’t get to Worcester from NY in 1:45. Assume an all-high-speed route NY-Bos would take 1:30, very best case (averaging ~250kph). This train would be an express, stopping only in New Haven, if at all. Worcester is a 2nd tier city, so the train that stopped there would be a limited, with a stopping pattern similar to today’s Acela. If the express is averaging 250kph, the limited will be averaging closer to 200kph, which theoretically means 1:32 to cover the 308km to Worcester. But if you are assuming conventional speed tracks from NH-Spr then you will not get a 200kph average for the journey. Also, 250kph average speed is very fast, to get it you have to have a max speed of 350kph for the route. More practically if your express does NY-Bos in 2:00 (for an average speed of 185kph using 250kph track) then your limited is averaging 150kph, which means Worcester in ~2:05, but only if you build NH-Spr as HSR. If you use conventional track for that route you will be slower still.
The non stop tilting service from Euston to Warrington Bank Quay averages 105mph with a top speed of 125mph. That is the same ratio as averaging 250km/h with a 300km/h top speed.
Fully agree you would need a non stop run to achieve that though, and a 320km/h top speed would help.
The fastest journey time I can find for Euston to WBQ is 1:44, which is 95mph average, which puts you back in the 330kph+ zone to average 250kph. Beijing to Nanjing averages 318kph on with 350kph top speed, but that is with a non-stop journey of 1,000km, which is impossible on the NEC, let alone just NY-Bos. The ratios I gave are consistent with the Nozomi/Hikari/Kodama average speeds vs top speed on the Tokaido Shinkansen, which is a good proxy for Wash-Bos in terms of length and frequency of stops.
There are lot of people in southern New England. All densely packed together. There can be TWO high speed routes out of Boston. One pointed at New Haven through Providence and one pointed at Albany through Springfield. Once the New England train gets to Albany it can use tracks that will be there for the markets to New York. Just like once the train gets to Manhattan it uses tracks that are there for the market to New York.
The very high speed service between New York and Boston will be going through Providence. It’s a lot flatter that way. Which means it will be cheaper and easier to build.
The “inland route” train leverages tracks that will be there for other reasons.
40 minutes to get between New York and New Haven, 45 at conventional-ish speeds between New Haven and Springfield and 20 between Springfield and Worcester. 40+45+20=105. 105 divided by 60 is 1.75. Or an hour and 45 minutes. Changing to an MBTA train in Boston isn’t going to get you to Worcester in 15 minutes after your 90 minute ride on the super express. Or even if you change to a high speed train.
There are lot of people hiding in those obscure secondary cities. 830,622 in what was Worcester County, 697,382 in metro Springfield and 891,720 in what was Hartford County. Roughly the same size as Denver, Baltimore or Saint Louis. All using tracks that will be there for other reasons.
This is not a problem from the point of view of capacity
Yes it is. The anecdotes are that the Clockers ran empty between Philadelphia and Trenton and standing room only between Metropark and New York. It’s why SEPTA and NJTransit don’t want to run combined services. The train would be almost empty between Philadelphia and Trenton and SRO between Metropark and New York. It’s why, when Amtrak was providing capacity for NJTransit by running the Clockers, Princeton Junction and Metropark made it onto the list of Amtrak’s busiest stations.
There is enough demand for service to Toronto from Buffalo and all points east of Buffalo that the single high speed route to Toronto will go from Albany to Toronto. There is enough demand for high speed service to Montreal from the northeastern United States, but only to justify one corridor. Where shall that corridor run? The options are Boston to Springfield to Burlington to Montreal, Boston to Concord to Burlington to Montreal, Boston to Albany to Plattsburgh to Montreal, New York to Albany to Plattsburgh to Montreal, or New York to Springfield to Burlington to Montreal. One of these five routes will see 8 express trains each way each hour. The other four will see zero.
Luckily, because a train can stop in multiple places and some of those places can even have two trains next to each other at the same time, it’s fine if zero Montreal trains leave Boston. They can take a high speed train to Albany and switch trains there, where their seat on the Toronto-bound train will be filled by someone from New York who got on the Montreal-bound train in order to make this exact same switch.
And if it should turn out that more people want the Montreal train to connect to Boston instead of New York, then the residents of New York can take the train to Boston and change trains there. And if the residents of whatever Backwoodsia towns are on the other routes want a train, they can build and run a local train that connects up with a transfer at the nearest junction station. They cannot, however, have the once per hour train that goes to their local station on a special service pattern designed to provide a bespoke one-seat ride to everyone.
That might be how it works in Railfanlandia, but in the real world, you can’t give everybody 1 TPH 1 seat rides. In the real world, some people get 8 TPH 1 seat rides and some other people get 8 TPH 2 seat rides. Maybe some people have truly esoteric routings that leave them with a 3 seat ride – guess how much service they have? Yep, still 8 TPH. I know, changing trains sucks, but them’s the breaks I’m afraid. At 8 TPH, you’ll barely even have time to notice the scenery around your transfer points.
The Vermont Agency of Transportation commissioned a study and it determined there isn’t enough demand in New England to build high speed rail across the difficult terrain of New Hampshire and Vermont. The train from Boston to Montreal can use the tracks that are there for the Boston to Toronto trains and then at Albany use the tracks that there there for the New York to Montreal trains. Without anybody changing trains because it’s Boston and Montreal.
If there’s enough demand to run 8 trains an hour between Boston and Montreal, then Vermont is wrong and there is enough demand to build the high speed rail across the mountains of upper New England.
But I don’t think Vermont is wrong. Actually, there’s not enough demand to run 8 trains an hour between Boston and Montreal, so you run 8 trains an hour between Boston and somewhere else and the group of people trying to go to Montreal take one of those trains and then switch to the train from somewhere else to Montreal. “Somewhere else” is almost certainly New York. There probably isn’t 8 trains worth of demand for Boston to Toronto either, but there is demand for 8 from New York to Toronto – but not 16, and not 16 from New York to Albany, so the best possible solution which services the most markets with the least resources is that the Toronto trains leave from Boston and the New York crowd switches in Albany.
That is the essence of compromise. Everyone loses something. In this case, the loss is that some people need to change trains, instead of the loss being that everyone gets less service everywhere to accommodate making trains switch lines instead of making people switch trains.
There isn’t. IT’S NEW ENGLAND.
The once an hour train can use the same tracks as the once an hour train to Toronto, the once an hour train to Ohio and the once an hour train to Chicago via Detroit use. If there is a whole trainload of people in New England that want to go any of those places why do any of them have to change trains?
Because that whole train load of people have the pesky habit of not all wanting to leave from the same place at the same time and because there’s many many more whole train loads of people trying to go to many many more other places. Some of those places are in a neat little line. Others, you can’t draw the neat little line because there’s a pesky angle in the way.
So you can send 8 trains an hour down one line and 8 trains an hour down the other line and have some people mad that they had to change trains where the lines crossed but lots of happy people with 8 trains worth of service.
Or you can send 1 train an hour down one line, 1 train an hour down the other line, a third train an hour that goes to the junction and then turns, a fourth train an hour that skips the stops we hate and only makes the stops we like, a fifth train an hour that doesn’t stop at all, and so on and so on.
Now instead of a simple “get on train A, then change to train B” with a train showing up at any time you care to leave, you instead have to wait for when the one train that brings you right to where you’re going shows up while a bunch of other people wait around and still other people are wasting theirs and the employee’s time arguing over why this train has a “premium” fare and that one doesn’t when they’re both using the same tracks and making zero stops between New York and Stamford and an entire cottage industry of spreadsheet-wielding psychopaths founds the railroad equivalent of FlyerTalk to “maximize value” from the railroad airline loyalty programs while a different cottage industry of influencers and so-called “creatives” finds a niche performatively screaming about how cool and rare and legendary this or that special esoteric route is while thousands of fans show up to take pictures.
Personally, I’m not interested in catering to FlyerTalk, the YouTube algorithm, or the kind of person who just won’t go if a transfer is involved. I’m interested in delivering as much service as possible to as many people as possible in an equitable way. Unfortunately, because we don’t live in Railfanlandia, that means some people lose out because we can’t run infinite trains to everywhere. They can’t have a single seat ride to every possible destination across the continent, but they can have trains every 7.5 minutes going in the general direction they’re trying to go with easy transfers along the way. That’s going to have to be good enough.
There isn’t going to be 8 trains an hour because it’s NEW ENGLAND. And masochists who want to change trains in Albany will be free to do it. Other people will look at their smartphone or perhaps even a watch and take the train that doesn’t involve changing trains.
If there’s not going to be 8 trains an hour then there’s going to be 0 trains an hour, and the one train load worth of people can get on a bus or drive instead.
There is no scenario where you get 1 train an hour to Montreal and 1 to Toronto and 1 to Detroit and 1 to Montreal (the other way) and 1 to Chicago and 1 to Albany but it makes the podunk village stops and 1 to Buffalo noted as “first stop Albany” and 1 to Concord and 1 to…
Because you can’t run 27 different service patterns without creating conflicts that reduce capacity and ultimately mean less trains everywhere. People walking across the platform is easy and has zero capacity penalty. Trains switching tracks is harder to do, no matter how much you scream and project and insist that it isn’t.
Pick a route. That route gets high-speed tracks and 8 express trains an hour. Every other route you didn’t pick gets local trains and you have to transfer. You can also pick nothing, and get a bus or a car. I think New England can have the line that goes west from Boston and the line that goes south from Boston. Lots of places are west or south of Boston.
Most of them that you somehow seem to believe warrant dedicated hourly service are actually all in one of those nice straight lines which means that instead of having 1 train each with 5 different stopping patterns and 5 different destinations, we can just have 8 trains that all go to those same 5 places and even more places besides those 5 places! Why is that so hard for you to understand?
If you’re spending money on high speed rail to deliver 1 train an hour then you are wasting money and your funding should be sent to the places that can deliver 8 trains an hour instead. I think all of New England can come together for 8 trains an hour going west. But if you can’t justify 8 trains an hour to (Worcester, Springfield, Albany, Syracuse, Rochester, Buffalo, and) Toronto (all of those stops in that order) then you can’t justify spending any money for high-speed tracks on the line that goes west, because a pile of disconnected once per hour services is too wasteful to justify building for. If that’s the case, hope that the local DOTs figured it out on the local train that gets you as far as the high-speed line which was justified… or just get a car. After all, New England’s empty, so I’m sure all those freeways are empty too, and when they start filling up, we can revisit finally building the HSR line … in 2075.
New England isn’t empty. Rhode Island is the country’s second most densely populated state, Massachusetts is the country’s third and Connecticut is the country’s fourth. The three of them have roughly the same population as Ohio in less than half the space. If it doesn’t make sense to build high speed rail in southern New England just where does it make sense to build high speed rail?
“If there’s not going to be 8 trains an hour then there’s going to be 0 trains an hour, and the one train load worth of people can get on a bus or drive instead.”
This is so foolish. Even just an 8 car intercity train can have 525-600 seats. That would take 11-13 standard coach buses at 48 seats each to carry the same load. That is a bus every five minutes – each direction! – to replace a train an hour, and the only thing you do is get people where they are going slower at greater cost. If you are running 16 car trains you can double that. Even at only a 60% train load but 100% bus load, a train would be so much more efficient/cheaper, on top of considering the reduction in pollution from an electric train versus a diesel bus, on top of giving people the tangible benefit of faster travel.
If you just want people to “drive instead” of taking HSR, why are you arguing for any HSR at all? The goal should be to maximize the number of people using the HSR line, by providing a well thought out mix of services that reasonably (but not perfectly) matches capacity to load and maximizes quality of service (i.e. speed) for the largest number of riders. The goal is not to maximize the number of trains using the line, by setting some silly standard of “only 8 tph or 0!”
No, the goal is to maximize the use of the line. Ideally, we’d be maximizing use of the lines that we already have by spending dramatically more on operations than is ever conceived, but the cynic in me has accepted that it’s impossible to get that without ribbons to cut and new lines to promote. So, given that every dollar of increased operational funding needs to be snuck in attached to capital megaprojects, the goal becomes investing our capital dollars in places that let us best run more trains overall, serving more population segments, and reducing private automobile dependency. That means a service that is frequent and reliable and takes you where you want to be – crucially, it does not necessarily mean fast, although a key part of reliability is not slow.
It’s a nuance, but my position doesn’t shift and hasn’t shifted. As Matthew has said in other comment threads and I will paraphrase here, we need to learn how to walk before we can run. We need more local train lines. We need more suburban routes. What we actually need is, pardon the gauche terminology, “metro-like commuter rail” operations and intercity corridors that are actually serviced by anything other than the train or bus of last resort. I want 8 trains an hour between Albany and New York City, and if the only way I can get those 8 trains is by having all of them run on a $160 billion greenfield “Empire State HSR” line then you had better believe I’m going to show up to demand 8 tph or don’t bother building.
Because I’m more than fine to not build HSR at $160 billion a “line.” I’ll gladly take $4 billion to purchase a bunch of new trains, hire a bunch of new staff, and run trains once an hour between New York and Albany, Albany and Buffalo, Buffalo and Cleveland, Cleveland and Chicago, and the railfans who want to take the train to Chicago are welcome to take the Lake Shore Limited or transfer at any (or even all) of those natural service breakpoints. I would gladly go back in time and redirect every federal dollar spent thus far into a raft of local projects designed not to create a single showcase state-of-the-art HSR line but rather to ensure that any city pair within 200 miles of each other that has active rail service today has it hourly or better.
I am not able to do that, and we are seemingly unable in this country to do anything that isn’t a megaproject out to redefine what rail can be through drawing a “transformational” … line on a map somewhere. And as I’ve also said repeatedly, I’m fine with that, but it had better actually be transformational then. It has to behave like a subway and not like an airline, because even in California with all the LAX/SFO air traffic, the train is primarily competing with driving for basically every possible origin/destination pair that isn’t exactly one end of the line to the other one. Once or twice an hour is not good enough, especially not when it’s once or twice an hour and you still end up in a private automobile at one or both ends of your trip.
Eight per hour is good enough for a singular twelve-digit megaproject, and one per hour is good enough if it’s one per hour available within 25 miles of the majority of residents in the country, a goal which would require hundreds of teeny tiny capital miniprojects but more importantly the massive ramp up in operational spending that I want.
I am frustrated and furious and tired of seeing vision maps and crayon line dreams and invitations to imagine trains that are always 10 years from accepting revenue customers. Trains exist all over this country today, and I am begging anybody to just run more of them.
The only place in the country that is going to have 8 or more intercity trains an hour is the stretch of track between North Philadelphia and New York City.
“One of these five routes will see 8 express trains each way each hour. The other four will see zero.”
This . . . makes absolutely no sense. On several levels.
First, the idea that Montreal to the US will have 50% of the ridership as Tokyo-Osaka is laughable (Tokaido Shinkansen runs 16 tph max at peak).
Second, many very successful HSR lines run 2,4,6 tph. The idea that your minimum is 8 tph or don’t build is silly. For perspective many successful air routes (the short/medium range ones that HSR can replace) fly one 737 per hour. An HSR train can seat 2x-5x a 737 (depending on length of train, seat density, and 737 size) so to expect 16x-40x times the passenger load on HSR is absurd. A huge market like SF-LA has about 137 flights per day (all airports, with only 16-34 SFO-LAX) so the idea that you are going to run 96 trains per day Montreal-NY (each seating many times an airliner) is fanciful.
It is worth remembering that even if the line can handle a higher capacity, not all of the slots need to be or should be filled with the longest intercity service. A NY HSR line should see services that are only NY-Albany or NY-Buffalo because the travel markets to them are large enough to support service multiple times per hour, so your tracks south of Albany may see 8-12 tph combined, even if neither Montreal or Toronto are not seeing 8 tph.
Third, although cross platform transfers, and transfers in general, are a successful part of any transit system, they are more a feature of shorter journeys (subways and commuter rail) than intercity, and at the longer distances are a feature of connecting secondary and tertiary markets, not primary ones. You are correct that the strength of rail service is linear routes, not exclusively point to point services, however, branching is also a viable and useful rail practice. From a train perspective, a service Bos-Albany-Montreal is just as much a linear route as Bos-Alb-Tor, even if it isn’t straight on a map. Both routes are metro areas connected in series.
In this regard there is no problem with running Bos-Alb-Mon and Bos-Alb-Tor and NY-Alb-Mon and NY-Alb Tor services, you just need the appropriate switches in Alb, and don’t pay a penalty for it if the trains are stopping there any way. Your suggestion that every NY-Tor passenger disembark in Albany is foolish when NYC-Toronto is the highest possible ridership pair out there.
The cases for transfers in intercity travel are passengers from Washington getting off in Philadelphia to get on the Keystone to go to Harrisburg. Or NYC passengers getting off in Albany to get on a train to Burlington. But you are suggesting something like all trains from NYC go to Harrisburg, and everyone has to get off in Phila to catch the DC-Atlantic City train.
Rail has a very high upfront cost to build in exchange for dramatically lowered operating costs. High speed rail tilts this balance even farther in the direction of initial upfront capital and lower operating costs.
You can stick a train capable of 220 mph on conventional rail tracks. That doesn’t make it HSR. And you can look at all the places with no rail service or dramatically underbuilt rail service and you can either choose to build a brand new high-speed rail corridor or you can choose to shore up the conventional route and settle for “higher” speed rail. That is to say, regional intercity trains.
If your route isn’t built to support the highest speeds then running the highest speed trains over it just to claim it as a “high speed line” is both inaccurate and inefficient; the Acela Express is marketed as high speed rail but should not qualify under any reasonable definition thereof.
Regional intercity trains are actually perfectly fine. You don’t need the highest speed Shinkansen to go from Washington to Richmond or Philadelphia to Harrisburg or any number of other possible city pairs in that ~100 mile distance range. You probably don’t even need them for ~200 mile trips, where an average speed of 80 mph beats every alternative mode of conveyance and is incredibly achievable with 20th century conventional railroad technology. Absolutely, we should be fixing up all the old tracks and restoring this conventional speed regional service.
But if you’re going out of your way to build a high-speed route, buy high-speed trains, and sell high-speed service: that’s not conventional rail. Conventional rail is not good enough. You have to create the high-speed line to high-speed standards which means a lot of expensive track work to get your 220 mph running speed through the mountains and it means either keeping the 220 mph trains separated from the 125 mph trains or managing a lot of complicated scheduled overtaking to let the faster trains pass safely. Lots of capital expense, that pays itself back through low operating costs. Actually seeing the benefit of those lowered operating costs? That means service, and a lot of it.
So, no, given the state of the tracks north of Albany and from Albany to Boston (hint: very bad) – rebuilding them to high speed standards is not worth it if you can’t run 8 trains per hour. And you’re going to have a hard enough time just balancing the schedule of the 220 MPH train to Toronto and the 125 MPH regional train to Albany and the other 125 MPH local trains that run only as far as Poughkeepsie or wherever Metro-North decides the end of local service is without adding in the need to shuffle in trains switching corridors.
The question is, can you justify 8 trains per hour that run at 220 MPH between Boston and Albany? If the answer is yes, then those same trains can continue on. If the answer is no, then you run the local 125 MPH trains on the conventional rail at the more expedient price point and you run 0 express 220 MPH trains until the ridership numbers from everybody transfering are there to justify rebuilding the line to that standard.
And, again, if the answer is “there just aren’t any corridors where that’s true except CAHSR” – then you send all of the money to CAHSR and wait until another corridor ripe for similar investment materializes somewhere else in the country. And you just keep building and restoring the local lines in the meantime. What you don’t do is spend $160 billion to deliver a once-hourly “high speed” train that mostly just runs on low speed tracks surrounded by low speed traffic (NEC FUTURE), nor do you spend $160 billion on a state of the art greenfield alignment supporting 220 MPH through the mountains and across the rivers of New England so that it can see one train an hour and spend the rest of its life empty.
If you’re building anything at these prices it had better be fully utilized. Once per hour to anywhere is a waste of money and unacceptable.
Very few high speed lines have as many as 8 trains per hour. London to Ebbsfleet does, probably the Paris end of the LGV Sud Est does, the Tokaido Shinkansen does, maybe Frankfurt Cologne does.
Even on the classic lines upgraded to High Speed Rail standards the Great Western as far as Reading has that service level, as does the West Coast mainline as far as Rugby and maybe the London end of the East coast mainline.
Frankfurt-Cologne and Frankfurt-Mannheim are 5.5-6 trains per hour (not all stopping at either Frankfurt Hbf or Cologne Hbf), it’s just the same schedule all day without extra peak service.
There are more people along the 450 miles of the Northeast Corridor than there are in the state of California.
Yes, I fully agree with you Matthew – and I apologize in advance if this comment sounds argumentative with you. You are not who I am trying to argue with.
Having prefaced with that:
Chicago is 790 miles from New York by car, 712 by plane, and 959 by the singular direct Amtrak route that serves this city pair today (the terrible Lake Shore Limited – I feel the Cardinal deviating to lower Virginia disqualifies it from being a direct route as even if you’re staying on the same train, you’re functionally connecting between a Chicago-Virginia route and a de facto Northeast Regional which just so happens to not accept passengers in or after Washington.)
I think its actually a tremendous instructive example of the position I hold: at 712 miles, the expected in-air flight time is just over 2 hours; even given an extreme 4 hour city-to-airport-to-city penalty modifier, it’s still 6 hours door to door and comfortably beating all but the most aggressive of crayon line alternatives: the best non-insane land route for HSR between these cities requires traveling via Pennsylvania and likely spending $160 billion punching the Pennsylvanian HSR line through the Alleghanies in order to produce a route that is just about 875 miles long; HSR on some crayon parallel of the existing route through Buffalo is the next best non-insane routing and that clocks in somewhere in the 950s range. Given a reasonable baseline average of 125 MPH (easily achievable on an HSR line rated for 220 top speed, comfortably doable if the speed limit is the 185 instead, and tough but theoretically possible under a 155 mph speed limit) these routes are 7 hours or 7’36” respectively: uncompetitive for all but the most ardent of rail fans (or “no flights ever” eco-warriors.) Even assuming a tough-but-doable-on-220 average speed of 175 mph gives us 5 hours or 5’25” which is still at best a draw with the air alternative.
No, HSR will never compete with flights from one end of the line to the other. HSR could be competitive with segments of this line, certainly: if you split it directly down the middle, Chicago to Buffalo and Buffalo to New York are both attractive high-speed rail options but this once again forces us to confront the unfortunate reality of the required transfer. Lots of other shorter travel pairs are similarly hiding in this long and winding and very bad route: Buffalo can easily justify lower-speed regional intercity service to many of its neighboring metro regions; it even has such service today – but only in the form of a miserably slow, frequently late, and inconveniently timed portion of this very same journey.
Furthermore, with some limited possible exceptions, there is already little to no regular air traffic on any intermediate sections of the line; it’s a convincing argument to say that the HSR crayon line would be competing with Buffalo-based (or Pittsburgh-based in the slightly shorter route option) flights to every other part of the line, but an incomplete one. Actually, the main thing that this HSR crayon line would be competing with is Interstate 90: certainly some passengers would switching off of flights from BUF and onto trains instead, but many many more would be switching from driving. The typical HSR rider on this line is not a railfan foaming at the mouth for the chance to have 7 hours of quality time on a train and it’s not the fantasy of an executive with a ranch in Western New York State and an office job in Midtown Manhattan, it’s the commuter currently driving from Buffalo to Schenectady or the political class driving from their constituencies to Albany or the college students wanting to get off campus for a day or two.
It’s the people who are driving now, and more critically, the people who don’t care what the top speed of the train is because they’re only on it for some amount of miles that they don’t even perceive as miles but rather as either time or “stops.” And for the people who are driving now, certainly a crayon line HSR route would help them out, but actually, lower speed local rail would help them even more: because there’s already a line that could be serving as local rail today, except it sucks. There’s already “service” east out of Buffalo in the morning and west into it in the evening, which is great if your travel pattern lines up that way and useless if it doesn’t, say for example, it you’re trying to attend an evening function in Buffalo but also trying not to pay for a hotel room. (I guess you can stay up until the first of your four eastbound opportunities at 4:30 AM out of Exchange Street.)
You don’t need to draw a crayon line between Buffalo and Albany because the train is already there, you just need to split off that section and run it more.
You don’t need to draw a crayon line between Buffalo and Cleveland and Pittsburgh because the trains are already there, you just need to split off those sections and run them more.
You don’t even need to split off the Albany to New York section because that line already has corridor service: you just need to run it more.
And the things that can help you to run those trains more are the same things that they’ve always been… but they’re the things we refuse to contemplate, because it’s better to “plan” for the New York to Buffalo to Chicago HSR “vision” of “the future.”
And I’d really like us to stop planning for the beautiful future that’s always 10 to 20 years away, and start purchasing trains that we can be running in 2 years instead.
The above comment is meant to be in reply to the comment at: https://pedestrianobservations.com/2024/02/29/northeast-corridor-travel-markets-and-fares/#comment-158057 and was somehow moved into this comment thread instead somehow. I guess clicking on another comment changes where the reply is going.
Rhode Island is the country’s second most densely populated state, Massachusetts is the third most densely populated and Connecticut is the fourth. Where do you proposes building high speed rail? Someplace less densely populated?
The train from Boston to Montreal can use the same tracks as the train from Boston to Toronto. And the train that makes all the stops along the way to Cleveland. Once any of them get to Albany they can use the tracks stations etc. that will be there for the market to New York City and beyond. Just where do you think it would be a good idea to build high speed rail?
Obviously, California, whose project continues to languish in hell in large part from lack of funding leading to endless delays. It’s a drain on actual capital, a drain on political capital, and an ongoing national embarrassment. If you have any other corridor in the country that you can get done instead at the price point set by CAHSR and justifiably support that level of service, then sure, spend there instead: but that’s basically just New York-Washington, not even New York-Boston (nevermind -Montreal or -Toronto or anywhere else).
We have, currently, one ongoing and underfunded HSR project in California that may some day deliver a functional high speed rail corridor to the largest cities on the west coast but currently seems unlikely to do so any time soon. If you want to spend any money on doing anything “HSR” that isn’t accelerating that, the return on investment needs to be as good or better.
In other words, it had better be able to justify 8 trains per hour, the water mark set by CAHSR, or it shouldn’t get built. We can come back and discuss whether a brand new high speed right of way is justifiable to build if it isn’t supporting 8 tph after the national narrative on rail shifts from “can it even work” to “when do we get ours” and a sustained campaign of successful building brings the costs back down to something where it’s reasonable to pursue hobby train lines with thin schedules.
In the mean time, if you want more rail – and I do – the barrier to entry for any type of rail project other than HSR continues to be a lot lower. The state of Virginia isn’t bothering with 220 MPH greenfield alignments: it’s purchasing up and fixing up freight lines and delivering measurable incremental improvements. It won’t be genuine high-speed rail (it might not even be branded HSR), but 110 MPH trains are being delivered to the south at the cost of about 1% of a CAHSR. They need to get serious about electrification and then they can deliver 125 MPH trains at the cost of about 1.5% of a CAHSR. That won’t be 8 trains per hour but the costs are far more in line with where they should be for 1 train per hour.
Further on down the line, North Carolina, who entirely owns its own railroad route already, currently runs 4 roundtrips a day (plus one more that goes all the way up to New York) that’s currently largely held back by Charlotte’s complete failure to finish their new train station along with no movement on improved layover facilities – again, marginal, incremental improvements that aren’t “HSR” but will deliver high”er” speed rail at a dramatically lower cost which means running dramatically fewer trains still makes the math pencil out okay.
You mentioned Ohio. Ohio, like much of middle America, is dramatically underserved by rail and also filled with some of rail’s biggest detractors. It is here in Ohio where we see the next likeliest place that full-fat greenfield 220 MPH HSR is justifiable, and probably the next line to build if we can ever finish up in California, but in the meantime: we don’t need to spend hundreds of billions on HSR there when we can spend singular billions on delivering Cleveland to Columbus in 90 minutes and Columbus to Cincinnati in an hour, running all day long. Local, regional rail. Hell, I genuinely can’t tell whether or not you’re pro-import, but if you’d prefer to build American, Ohio and the south are both great places to build up the factories you can do that with too.
And speaking of places where rail doesn’t exist or only nominally exists, lots of places are within easy rail reach of Chicago on competitive times today, held back only by lack of service. Chicago to Detroit, competitive with driving (5:20 by rail, 4:15 driving with no traffic – figure normal traffic pushes that to 5) but has 3 round trips daily. You don’t need to build HSR to shave an hour off that trip and you don’t need to build HSR to run more trains. Actually, all you need to do is buy more trains. A lot more trains. Leaving the Northeast Corridor and Empire Corridor out, the next best-served line in Amtrak’s system is the line to Milwaukee which sees 6 round trips a day despite being a fully double-tracked rail line that delivers Chicago to Milwaukee in 90 minutes. Not HSR. Able to run hourly trains right now if we had the supply available. Do that instead.
This is the point, and this has always been the point: lots of existing rail links in America are bad. Lots of obvious 100 mile and 200 mile trips that should be obvious rail links in America are missing. If you want to prioritize showcase high-speed rail lines at US construction prices, you had better be delivering a showcase: 8 trains an hour. Otherwise, it’s a waste of money.
What’s not a waste of money is taking the $160 billion we project to spend on one single showcase HSR line and delivering 100 regional rail lines for about $1.6 billion each. It’s the most prudent and expedient thing we can do for rail in this country, but railfans hate hearing that, because it means less ribbon cuttings and less room for their imaginations to go wild, covering the map with crayon lines for this train and that train and oh the many places we could go if we just built more HSR. They don’t really want to hear about cost control either, and they don’t want to hear about fixing up what you’ve already got before pursuing the shiny new toy.
And that’s honestly fine, there’s room for CAHSR in all of this too, but again the other point is: 8 HSR trains per hour, or 0. We have better things to spend our money on than hourly HSR service between Boston and Montreal. Such as local service to Springfield. Or local service from and west of Albany.
California was promising 12 trains per hour, but that was incredibly optimistic, on the theory that everything would eventually fill. (North Atlantic Rail has the same assumptions with similar problems.)
The LGV Sud-Est runs about 12 peak trains per hour and has an economic (not just financial) rate of return of 30%. It had atypically low construction costs, which boosted its ROI; but conversely, some of those trains per hour only became viable after subsequent lines were built, not included in the ROI. Trying to work this from first principles with TGV data, with $30 million/km in construction costs rather than the much lower figure for the LGV Sud-Est, I’m getting a financial-only ROI of 1.5% per train per hour, so you need around two to three to break even on risk-adjusted government debt or even a bit less, not eight (or 12).
Every successful HSR project we deliver lowers the barrier for entry to every subsequent project; it’s why I’m leaving open the door to revisit lower frequency lines after we deliver CAHSR and eight (or six) an hour on the NEC. I’m also hopeful that part of the solution on out of control costs is to just start building a lot more in these incremental places where the magnification doesn’t hurt as bad and where building a strong state-based engineering core can then pay dividends when it comes time to tackle the next big HSR corridor – the same reason I think fixing up Charlotte – Atlanta today so that it gets just a little faster (and gets hourly or bi-hourly service instead of daily service at a terrible time) is valuable so long as you’re actually just fixing it and not trying to sell a brand new HSR line at current US HSR prices.
We can break even a lot more easily on these smaller projects, but conversely, just breaking even on CAHSR may never have been good enough and certainly isn’t good enough now. We need some wins. Let’s take some of the readily available and very cheap and easy wins first.
Given the US has less state capacity than we do in the UK I think getting some wins from a US north east corridor high speed line will be very very hard.
Upgrading the whole line to 125mph with tilting trains and doing over 100mph average on the whole thing like the West Coast Mainline – which strictly would be high speed rail per the definition – is the most deliverable in that corridor. And yes it’s 105mph between London and Warrington Bank Quay as it’s 96mph between London and Preston and there are stops on that bit.
I forgot to include the reason. Look at the fuss in the Chilterns where HS2 failed to provide any wins at all. Well North East Corridor High speed rail would have the same problem except that even if you did do the fixes they could have done in Buckinghamshire there would be too many surburbanites who never use trains.
So, I don’t think the politics works out the same way. Against the ability of the US to do this, there’s way worse decentralization, a longer tradition of the heckler’s veto, and much higher incomes in suburban Connecticut than in the Chilterns. But in favor of it, there are mitigating factors:
* Decentralization as far as is relevant to federal policy is to the states (via influence peddlers in the Senate) – and no matter what Darien thinks about takings, there are going to be stops at Stamford, New Haven, Hartford (on a branch), and New London.
* The timetable padding on New York-area commuter trains is so extreme that even the white flighters benefit from modernization. Darien today gets about a train every half hour at rush hour, getting to Grand Central in 1:07-1:09; the timetable that Devin and I are developing looks like the same stopping pattern plus a New Rochelle stop would be around 36 minutes. This may shift the politics to be similar to that of Boston, where, due to the trip time benefits of electrification and high platforms, there has not been any push from 9-to-5 suburban riders against Regional Rail, only from fiscal conservatives who worry about the upfront costs of wire.
* There is a lot of federal willingness to spend political capital on the Northeast Corridor – it’s an issue of personal importance to both Biden and Buttigieg, and there’s bipartisan consensus around it so far (John Mica was at least in theory in favor of this, after the 2010 midterms), though of course as we’re seeing with Sunak’s behavior, such consensus may fray if trains turn into a culture war issue.
There are long stretches of track, between New York and Washington D.C., where the speed limit is 135. It’s not higher because the catenary needs to be upgraded. And the track. A few curves need to be straightened.
Unfortunately for the second- third- and fourth- most densely populated states in the country, there’s actually 45 other states in the continental United States of America where people also live and all those people have wants and needs as well. Quite a lot of those people currently have wrong conceptions about any number of things including rail, and quite a lot of those wrong conceptions come from having no access or exposure to real rail. All of those people also get pesky pesky voting rights, and as long as the narrative is “pump all of the money into the one place with a functional rail line already and maybe someday a railfan’s crayon line will deliver you a connection too” they are going to keep right on using their voting rights to stop and obstruct meaningful progress anywhere. The railfans will tell you that the answer is HSR picking up Ohio on the way between Chicago and New York or any number of other similar “wouldn’t it be nice if…” crayon lines drawn on a map. Some of these maps even get official branding from Amtrak or professional lobby groups.
What the railfan doesn’t support, doesn’t like, and doesn’t want to hear about: functional regional corridors that exist as subsections of the existing long-distance routes today getting more trains than once per day. Existing state sponsored corridors with 6, 5, 3, or even 1 daily roundtrip despite being capable of supporting any level of service that can be described using the word “hourly.” Setting goals like 10xing staff rosters, 10xing equipment supply. Electrification – real overhead wires, not battery crap. Boosting average speeds to around 80 mph instead of 30 mph. All of these things that we can and that we should do to deliver a better Amtrak then has ever existed and that can open eyes to the potential of rail investment in places like the rural midwest and the deep south and snowball rail investment. Not necessarily mutually exclusive with “HSR,” but also not “sexy” – ribbon cuttings don’t happen for a targeted curve straightening or an additional 6 roundtrips a day to Milwaukee or Charlotte or wherever else. You might be able to finagle a ribbon cutting for a new maintenance facility. (Definitely for the factory that gets built to deliver the 10x railcar supply.) And because it’s not “sexy,” it gets ignored: better to imagine what it would be like to have Amtrak connect us or the future of rail travel and accuse anyone who balks at the price of being backwards or not getting it, then turning around and asking ourselves how we can convince people to believe in trains again or lamenting about how bad trains in America are as though the thing that keeps getting shot down is the manifestation of their crayon line dreams and not the reality of “one single extremely overcosted high speed line for one of two possible groups of coastal elites and 50 more years of bad rail service for everyone not in that group as we send token amounts of study money to ‘identify’ future rail corridors.”
But, I restate: it’s not necessarily wrong to do that. It is not automatically a terrible idea to pick a single corridor and say ‘we are spending all the money here and we are aiming for the moon and we’re going to deliver a Statement and that Statement is that rail is the future and the future is right here’ in California or New York or maybe even in Ohio. That’s a choice you can make if, and only if, you’re willing to commit to it, to go all in, and to consider anything less than full fat service from day 1 a complete failure.
So. You can try to spend $160 billion and deliver actual high speed rail to one coast or the other one, justifiably run 8 trains an hour on that line and maybe hopefully get everyone to start believing in it because the conception of rail in America stops being Amtrak and starts being that instead. But if you want that, if you’re trying to land a moon shot, if you’re saying HSR is the 21st century space program, then you have to deliver exactly that and nothing less.
And if you can’t deliver that, then you shouldn’t deliver less, because that same $160 billion pays for 100 different projects throughout the country, lots of which we should have done 30 years ago and certainly should be doing today. The other benefit to that is that if you only get a quarter of the money, that’s still 25 projects you can do, all of which are far more worthwhile than a showcase HSR line that sees one (1) train an hour at the same price tag.
8 or 0.
The definition of “rural” is places where few people live. Places where few people live don’t generate a lot of votes. Or passengers either. Because there aren’t many of them.
Yeah. Sucks that those few people count more than you do. Or I do. Or any of us here, really. So, what, you want divorce? Maybe we’ll have a better shot at getting anything done if we only need a dozen states rowing in the same direction?
Not sure that “bring on the national divorce” is a winning strategy either though. Personally, I think the winning strategy is actually to get everyone on board again, which means less screeching posts about the behavior and voting patterns of senators from parts of America we hate and more investment across the country, yes, even into the bad parts of it.
Your concern-trollery is becoming obvious.
Once again, you’re projecting.
@Matthew Hutton
“And yes it’s 105mph between London and Warrington Bank Quay as it’s 96mph between London and Preston and there are stops on that bit.”
Can you identify where Lon-WBQ is timetabled at 1:34, which would be 105mph for the 165mi trip? I can find nothing faster than 1:44, which puts it at the 95mph average I referenced.
@onux, the distance by rail from London to Warrington Bank Quay is 179 miles not 165.
Sorry it’s 182 miles by rail as per https://www.realtimetrains.co.uk/service/gb-nr:C27694/2024-03-03/detailed#allox_id=0
@alon, I think you have kinda missed my point about the opposition.
The upper middle class older people who live in the Chilterns who run every political campaign ever will catch the trains into London every few months for a leisure activity because the train works for that. I am far from convinced that those same people catch the train into New York City for leisure trips.
I also think those people would be much more likely to take a train trip to the north of England or Scotland a couple of times a year even with just an hourly train from Marylebone to Milton Keynes – and certainly if it was run every 30 minutes that wouldn’t be bad by any means. The fact that the rest of the US rail system is doesn’t even run hourly trains makes that less likely in the US.
Yes, exactly this. I will admit my position is more extreme than most, in part because I’ve lived through 25+ years of aborted HSR proposals and returned funding and every shiny glossy promise of The Future that’s surely coming someday while I wait for a future of rail travel that involves trains that I can see running today running at times that are useful to anybody for anything other than peak hour commuting.
The fact of the matter is that the entire country of the United States needs to be shown that rail is a real possibility and that it will work in their communities, and that starts with delivering them functional train service. Not a “vision for a nationwide HSR network” we are nowhere close to even being in a position to build, and not crayon drawings of a future network spanning coast to coast, but simply a vision of everywhere that Amtrak runs today having local corridor service to somewhere. A vision of the US where rail is a viable option for anyone other than the Northeast Corridor and a somewhat limited subset of feeder services into it, where proposals to activate new conventional rail lines to places like Scranton come with hourly service by default instead of “3 trains daily,” where rail can actually occur to any random person off the street as being a viable travel option.
And I’ll confess to, in the past 4 years especially, continuing to feel a ton of frustration over the fact that there’s tons of active passenger rail lines in the US today where rail could easily be viable if it was serviced at an appropriate level but the only initiatives that ever see airtime are what new HSR lines people can dream up instead. Or which local governments can grab $500,000 of FRA “corridor ID” money to identify whether or not maybe a train makes sense between any two cities while CAHSR continues to go nowhere quickly from lack of funding just the same as how all these places studying whether a train is a good idea or not already have trains whose main problem is that those trains are useless.
It is for this reason that I’m so insistent that the next HSR lines have to be good enough to overcome the reality of miserable US train service, or the focus needs to shift away from HSR and onto 21st century rail modernization instead. And it is for this reason I’ve been saying and will continue to say that any new HSR line needs to see 8 trains per hour or it shouldn’t get built at all.
I think perhaps the US should stop running before it can walk.
Fixing the suburban commuter lines so they work for leisure travel would be a good start. As would connecting Chicago to New York so you can do it in a day without a sleeper or with a sleeper without lots of extra time.
There is a comment I made three minutes ago that was meant to end up in this reply thread but somehow ended up in a different reply thread. I contemplated copying and pasting it but elected not to.
@Matthew Hutton:
Interesting, every source online (including Avanti’s own website: https://www.avantiwestcoast.co.uk/travel-information/train-times/london-euston/warrington-bank-quay) gives the distance for Euston-WBQ as 165 miles.
On your previous link I noticed that distance was given as Miles:Chains, which at first made me think the station and junction “milages” were an original Victorian chaining made obsolete by route improvements and cutoffs of the years (it is not uncommon on older lines that if a tunnel cutting two miles off the route opens, that the mileposts go suddenly from 37 to 39 – that way the location of the station at MP40 and every subsequent station, switch, bridge etc on every subsequent branch do not have to be relabeled). But I traced the route on Google maps, and it agreed with the distance to WBQ and all intermediate stations I checked.
Then I noticed that the train in your link was taking 2:11 not 1:44, so I wondered if perhaps the express took a straighter, shorter route from Rugley to WBQ, instead of going west to Crewe then back east. But checking on a Euston-Glasgow train that makes the run express (https://www.realtimetrains.co.uk/service/gb-nr:C27706/2024-02-29/detailed#allox_id=0) shows that it indeed follows the WCML primary route. In fact, that train made the run in 1:42 or slightly less, for an average speed of 107.14mph/172.5kph.
Unless Avanti has some trick to shorten the WCML by 17mi, you are correct on the speed of the Euston-Warrington express. Making 107mph station to station on a 125mph line is top quality railroading, the 85% ratio is equivalent to what China achieves on the Beijing-Shanghai trip, and just behind the world’s best 90% ratio on the Beijing-Nanjing non-stop. All the more so in that the British example is much shorter than the runs to Nanjing or Shanghai and uses a route originally laid out close to two centuries ago not two decades ago. Of note, a year after opening, the Shinkansen was making 101mph average speed Tokyo to Osaka.
In the context of the original comment and average speed on the NEC from NY-Bos, I would still hesitate to plan for 250kph average out of 300kph line speed. Unlike these non-stop runs, I don’t see a situation where the express/Nozomi level service does not stop at least New Haven (for transfers) and possibly Providence (Alon’s plan) or Hartford (my plan) as well (as the largest city between NY and Bos on the respective routes; although not as large as other express stops like DC/Balt/Phila.)
The point of a Nozomi is that it an EXPRESS train. EXPRESS trains don’t stop at every wide place along the tracks. If you want to go to New Haven to change to Metro North or SLE take the Hikari from Washington D.C. or Kodoma from Philadelphia.
The Nozomi has 5 stops over 515km, about one stop every 128km. For the NEC this would mean 7 stops (rounding up from 6.7). DC-Balt-Phila-NY-N.Haven-Prov-Bos would be the same stop frequency, not stopping at “every wide place.”
If you want your express stops to focus on the major MSAs (2M+) you can skip Providence, but you still need New Haven. The point isn’t to transfer to MNRR or SLE in New Haven, the point is to transfer to/fromfrom a Limted/Hikari or Regional/Kodama. Unlike the Tokaido the NEC effectively splits at the north end. Passengers to/from Hartford, Springfield, New London, etc. will all want to go to places to the south, and will want the option to connect with an express to get them there faster. Taking the limited from DC or the regional from Philly adds 30-60 minutes to a trip.
There are only three and half million people in Connecticut. There aren’t any stations in Connecticut that serve 2 million people. There aren’t any that serve a million. …It’s New England not Tokyo-Yokohama.
The Hikari isn’t going to meet a Nozomi in New Haven because it’s NEW ENGLAND. In some wildly successful scenario there are six intercity trains an hour between New Haven and New York. The southbound New England HIkari can stop in Newark, Newark Airport, Philadelphia, Baltimore and Washington DC. If you want to go to New Brunswick or Trenton use the Philadelphia Kodama.
Like airports, few people have an origin or destination of the train station. There are going to be lots of people changing to local trains at stations along the Northeast Corridor because the local trains actually exist and get used.
Either there’s enough people in New England to run a train there or there isn’t.
If there’s enough people in New England for the high-speed train from New York to Boston then it will stop at Penn Station, Stamford, and New Haven on its way to Boston.
If there’s not enough people in New England, then you will take a regional and connect to your faster train in New York.
I’m sorry that it’s so difficult for you to grasp the concept of sharing with others but unfortunately the price of getting your fast train to Boston is that you do actually have to share it with Connecticut, a place which either has enough people for a train or doesn’t no matter how many times you repeat “ITS NEW ENGLAND” in all capitals to try and force two contradictory things to be true at the same time.
The Acela Nonstop was a terrible idea in 2019 and will still be a bad idea in 2039, no matter how many times railfans repeat the word “Nozomi.”
There are enough people in New England to have EXPRESS trains from Washington D.C. and trains that make more stops.
Sorry! Capitalizing the word EXPRESS repeatedly isn’t going to manifest your dreams of sailing past people waiting on the platform in New Haven, either. Your train’s stopping there if it exists at all.
Certainly there is no point in not stopping express trains at a major stop where there is a slow corner in the neighbourhood of the stop which applies to a fair few of the stops on the NEC.
Perhaps in Britain it would be worth speeding up the slow tracks at Carlisle to do a London-Preston-Glasgow service which would knock about 30 minutes off journey times. But then a lot of the intermediate places are pretty small and/or the line is pretty damn fast past them.
@Adirondacker
“There are only three and half million people in Connecticut.”
There are 3.5 million people in Shizuoka Prefecture . . . and six Shinkansen stops. The one thing @Transit Hawk is correct about is that you are wildly and incoherently flipping between “there won’t be service X because its just New England and there are not enough people” and “New England is large enough for multiple service patterns to DC and its own HSR line to upstate NY and the Midwest”.
“The southbound New England HIkari can stop in Newark, Newark Airport, Philadelphia, Baltimore and Washington DC.”
That is too close to a Nozomi stopping pattern. The Hikari pattern (in terms of stop spacing and concept) would be the current Acela, including Metropark and Wilmington; you may add Newark Airport if you wish.
@Transit Hawk
“If there’s enough people in New England for the high-speed train from New York to Boston then it will stop at Penn Station, Stamford, and New Haven on its way to Boston.” / “The Acela Nonstop was a terrible idea in 2019 and will still be a bad idea in 2039, no matter how many times railfans repeat the word “Nozomi.””
Are you suggesting there should only be a single stopping pattern for all HSR services on the NEC? If so you are even crazier than Adirondacker. If you stop at every station along the way then your speed will be no faster than today’s Acela (the all-stop Kodama in Japan averages ~125kph, below your minimum 150kph threshold for HSR, well below your 200kph average preferred threshold.)
Approximately 70% of services on the Tokaido Shinkansen are Nozomi. You do realize that people repeat Nozomi because the Japanese have been running one of the most successful and well used HSR systems for generations, don’t you? The travel demand between major metropolitan areas is much greater than between major metros and smaller metros, or between non-major metro areas. This isn’t to say other areas shouldn’t get service, but your busiest pattern will be likely be the express service stopping DC-Balt-Phila-NY-(NH?)-Bos. To suggest that you wouldn’t run this pattern and instead make every train stop in Stamford is just nonsense.
Perhaps before commenting further you should take some time to research how HSR actually works in places that have successfully implemented it, before making further comments on how it should be designed in the US. It isn’t just Japan. AVE runs non-stop trains from Madrid to Barcelona, as well as ones that stop in Zaragoza. In Italy you can get a train that goes Rome to Milan non-stop, some that also stop in Bologna and Florence, and some that also stop in Piacenza and Parma.
There are 15 million people in New England. Unless you want to ban automobiles and airplanes there is a limited amount of demand for railroad services. I understand how local and express service works whether it’s on the subway, commuter trains or Amtrak.
If they skim off a trainload of people at 5:15, in Washington D.C., that want to go to Philadelphia, New York and Boston it frees up seats on the 5:20 for people who want to go to Trenton, Newark, Jamaica, New Haven and Providence. 15 million people there can be a super express in the morning and late afternoon. To free up seats on the trains that make more stops. there isn’t going to be enough demand at noon time. Because there are only 15 million of them, there is approximately nothing north or east of them and it’s going to be 15 million of them until worldwide population begins to decline late in this century.
…. Wilmington is almost exactly halfway between New York and Washington D.C. The station has a tight curve on either side of it. Which would be very very expensive to fix. It would be okay if Wilmington had two or three slower trains an hour and express trains were blowing past on a bypass out in the swamps near the freight bypass and I-495. The through runnning railfans can plotz over the thought of a commuter train service from Springfield Mass. to Wilmington Del. Which would give them an additional train.
(Pre-empting the comment system sending this comment into some other thread by stating in advance that this is a reply to the comment at https://pedestrianobservations.com/2024/02/29/northeast-corridor-travel-markets-and-fares/#comment-158087)
This is a problem that is rising from inconsistent definitions, I think. Part of that is due to an informal insistence on talking about this in terms of the Shinkansen, even though the Shinkansen is not a perfect 1:1 match with the Northeast Corridor (actually, it’s not a match with what US HSR is, could be, or in my opinion even should be. More on that in a second.) And I don’t mean this as a way of saying “ha, gotcha, I was actually arguing against a strawman I built out of creatively reinterpretation your comments” but rather to say that:
In my conception, and I think in the general conception of most people using the three-tiers terminology, the actual stop distances on the real Nozomi are not relevant: the concept is and the concept of a Nozomi-equivalent is a super-express that would only stop at the largest “terminal/core” stations: Washington, possibly Philadelphia, New York City, and Boston. This was deployed as the Acela Nonstop between Washington and New York in 2019, is the stop pattern that the word Nozomi immediately calls to my mind, and is the train that shouldn’t exist on the NEC (or anywhere else). The concept of the Hikari-equivalent, a limited-stop express train, calls to mind today’s Acela and is mainly the format I envision for “HSR trains”, with the Kodama-equivalent to be the Regional, or some future permutation thereof.
I would ask you to keep this in mind as I talk about HSR going forward, in particular when it comes to the other half of this comment, because:
Yes, I have been suggesting that there should be a single stopping pattern for HSR on the NEC, because HSR on the NEC to me isn’t the entirety of the Shinkansen or the conceptual Shinkansen-equivalent. It’s just the Hikari-equivalent, or the service called “Acela Express” (or whatever other name), and if you have a named service such as Acela Express, every single train that shows up under that banner should make the exact same stops. Similarly, the regional intercity train (which is what I truly want more of) would conceptually equate to the Kodama, but does not need to be nor do I necessarily consider it to be HSR. It is the “Regional,” which depending on stop density derives so little value from a very high speed ceiling versus a train designed for any other purpose that having a standard train model for “HSR” routes and a different standard model for “Regional” routes makes sense. I don’t feel strongly enough about that to oppose having a single model for all intercity train services, although I maintain that it is critically important to say that the Express train only makes the Express stops and the Regional train makes the Regional stops. (And, again, that the nonstop train – which is what is still in my head when I read Nozomi and I apologize for that – shouldn’t exist.)
It’s also critically important to have local services run by the local operator, which should utilize the same sort of binary local-or-express stopping pattern on a standardized model of railcar geared towards local rather than intercity service. This is already four different stopping patterns to juggle. Further granularity very quickly diminishes both capacity (through juggling many different stopping patterns) and usefulness (through forcing people to consult an app or a spreadsheet to figure out whether the train pulling in goes where they want to go instead of visually identifying ‘ah, it’s the Regional to Penn Station, I know I can take it to Stamford’ and just getting on.)
That’s critically important, but it also doesn’t exist – never mind the rest of the country, not even the NEC is fully covered by local services and there is zero integration or standardization between any of the operators serving as the local services today, and that’s a major source of my frustration. As I’ve mentioned before, I’d much rather see all the HSR money get spent on local services; and to restate my position I’m generally opposed to constructing new greenfield HSR lines to replace things like the Empire Corridor because my understanding (and, I suspect, the common cultural understanding in the US) of what a greenfield HSR line is going to deliver isn’t the entire Shinkansen but just one subcategory of possible Shinkansen services – thus, my absolute and uncompromising insistence on 8 tph or 0; which, given the revised understanding of the differences between our definitions, I want to clarify means I really want 8 Hikari-equivalents or sending all the money to Kodama-equivalents instead.
If you don’t have the all stops trains use the high speed tracks you aren’t going to get 8 trains an hour between Philadelphia and New York. Really, you have to keep your concerns coordinated.
Really, my concerns are perfectly coordinated. If there’s $160 billion in money to spend on any kind of passenger rail at all: it can, should, and must be spent locally. $160 billion buys transformative change to rail in America, meaning everywhere in America and not just the winning corridor. Divided out over the entire country via a modernization and jobs program that never once uses any three word phrases that abbreviate to HSR, it’s enough to bring hourly or better rail service to virtually all of Amtrak’s network in the form of new and enhanced corridor services, fully fund regional rail initiatives in several major metro regions, and is the best way to demonstrate that Rail Can Work both in and for its current largest detractors in the deep south and middle America.
If there’s $160 billion in money to spend on raild the only thing we can do with it is high speed rail, then all $160 billion needs to go into a service that runs 8 trains an hour – probably CAHSR, possibly one of the handful of alternatives we can just collectively refer to as “upgraded NEC,” really unlikely but not impossible it’s many (or any) wacky routes “identified” by rail fans who may or may not have a corner office at FRA/Amtrak HQ such as the ones in New York and maybe to Canada. That singular HSR line has to be as transformative for rail in America as the option that I actually want would be, which means 8 trains an hour on what functionally is the subway line from SF to LA but is also more importantly good enough everywhere in between to convince someone currently driving I-5 or California Route 99 to instead park their car at the closest garage and hail a ride out of the closest station to wherever they were originally driving to, and even more importantly good enough to convince everyone in the other 49 states in the Union that it’s actually something they want a copy of badly enough to drum up a second $160 billion to get one for themselves.
If there’s $160 billion to spend on rail, and the stipulations are “has to be HSR and will only have 2 trains per hour,” well, that’s a problem of backwards priorities and the only solution there is to “Vote No.”
@Transit Hawk, if you want good value for money you need to buy things from abroad.
America has no history of building trains at this point so buying them from Europe or Japan is sensible.
When I say spend it all locally that primarily means on local projects: a targeted curve straightening here, a station modernization there, hiring local to staff up hourly trains running between the closest two big cities on any given line out in the country where Amtrak runs today (but only technically because it “runs” as a single long distance train). This is in part because that’s going to do far more for rail’s cultural perception in the United States than being able to point to a world-class HSR line somewhere (again, with the caveat that maybe a legitimate best in class world beater of a HSR line does the same thing – which is why my posts are “8 or 0” and not “defund CAHSR”) and in much larger part because of that lack of history, a problem to which the only real solution is to start writing a new history.
And yes, the most sensible thing to do right now is to import from the companies with a long history of building because they never stopped, and we need to leverage the institutional history and standards and lessons learned from those places. Spending all the money locally doesn’t mean prying the Budd IP loose from whichever conglomerate has it now and founding a plucky rail startup under that brand. It also doesn’t mean insisting that the US not being Japan or Europe means that we need to reinvent the wheel.
But it does mean ground-level investment with an eye towards rebuilding that history; it means acknowledging that delivering a train in 2 years requires importing it but that the goal is getting to the place where we no longer have to, and to a certain extent, to borrow phrasing from Alon’s most recent post – to transfer waste from consultancy fees into “waste” from bettering the neighborhood. (Mostly, in this case, by way of the dreaded and always controversial “jobs program.”)
Adirondacker12800 will be by shortly to take the slam dunk on me that I’m laying up for them so I might as well put that in such an easily “gotcha!” quotable format. (You’re welcome buddy!)
It means (re-)learning how to walk before trying to run – but also it means, because I love tortured metaphors in the morning, actually going to rehab and doing physical therapy and putting the work in to re-learn how to walk instead of just giving up and deciding we’ll never walk again.
That’s what I desire most for the future of rail in the USA. I want to be standing in 2 years, walking in 5 years, and just maybe running in 15. But, if we are going to give up on ever standing on our two feet again, then I at least want the wheelchair we’ll be stuck in for the rest of my life instead to be the #1 coolest and most powerful wheelchair in the world.
No matter how you slice it 15 million people have a limited amount of urges to take trips of any sort. They have more destinations within range of HSR than any other place in the country other than the Northeast Corridor. 15 million people would be the country’s third biggest metro area after New York City or Los Angeles. It’s going to be difficult to scare up ridership in places with less people.
Go ahead, toy with a few lines. Chicago-St Louis? Metro Chicago is smaller than Southern New England and metro Saint Louis is smaller than the Albany-Buffalo corridor. Which will have high speed tracks for the market to New York and beyond. It’s almost as big as Dallas-Houston. With Toronto, Detroit or Ohio beyond Buffalo. Dallas has Oklahoma and Houston has the Gulf of Mexico. Almost anyplace that’s not the Northeast Corridor or someplace that connects to Los Angeles.
All the trains will be able to go to all the platforms. They can have the reservation/ticket sales computers suggest which service patterns make the most sense.
Worcester, Springfield and Hartford have almost as many people as metro Baltimore. The reservation/ticket system can suggest that there be an express to Washington D.C. once an hour and an all stops to Wilmington or Harrisburg a half hour later. Or not. In 2090 when it’s actually completed.
While it is possible to run a successful HSR system with only 2 TPH, I believe that is only because they are not counting sunk costs. Correct for them, but not useful for discussion of a new system which has no sunk costs to ignore. If you are building a new HSR then you need to consider what it will cost to build the line at todays prices, not just the cost to run the line that is already built at yesterdays prices. Once you realize that I’m convinced you will discover if you need extreme cost control to make 3 TPH work out, and you probably need 5 TPH in the real world. Dense areas like the NEC probably need 6 TPH just because of all the takings driving up costs (it should get more than that). Of course if you can upgrade any existing right of way those sunk costs will drive costs down.
If you have an individual proposal we can analyze specific construction costs, expected operation costs and then try to figure out how ticket prices will affect ridership to figure out if a line is cost effective. However for discussion if we do not expect at least 6 TPH in the US then we shouldn’t be serious about HSR there for now – if we get experience in building and running it we can probably start looking at 3TPH lines in the future but not now.
Acela costs three times as much.
It’s a good thing that there are enough rich people along the Northeast Corridor that Amtrak is able to grift them for business class fares in coach and first class fares in business class. With the side effect that it keeps entitled rich people off the Regionals.
Commute × multiday: I understand this is very common among university students in Europe (may not be in the US), and while much less common, not unknown among “adults” either. On weekdays, live in some form of small and cheap accommodation (university dorm, or with flatmates, or an HMO, etc., whatever one finds convenient/available; it’s entirely unsurprising if this changes one year to the next), on Friday afternoon travel home to family (one would typically be legally resident there), on either Sunday afternoon or early Monday morning travel to the metropolis. This may have been included under the tourism/leisure category, but I would argue that its traits are better described by the commute category. (Above all it is highly regular. Further aspects, less relevant for transportation planning, are that the description of hitting a string of distinct amenities doesn’t fit, as well as people’s conception disagreeing with the model — mostly they consider their primary residence to be with their family, while the metropolitan pied-a-terre is ephemeral, rather than considering themselves to live in the city and merely visiting their family.)
One potential lever to pull to increase Vegas ridership could be to build a people mover or something to the Vegas airport, assuming we can get Vegas to Union Station in closer to 2 hours. Vegas has a ton of domestic flights, and often to smaller markets than those served by LAX (e.g., Midland Texas). More importantly, Vegas is often significantly cheaper to fly to than LAX, being dominated by leisure travelers. It could very well be more cheaper and/or more convenient for someone to fly to Vegas, particularly on an LCC, than fly to Chicago/Houston/Denver and connect.