Cost of taking trains vs cost of taking other modes in the U.S.

Amtrak Unlimited Discussion Forum

Help Support Amtrak Unlimited Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
At the IRS estimate of operating costs for motor vehicles ($0.625 per mile), you just spent $200 for the 320-mile round trip, plus you had the aggravation of a 6-hour drive along I-95. Maybe you should have just spent the $200 on the train ticket for your daughter. Another possibility might have been to just drive to New London and have her take the Shore Line commuter train to New Haven, and then Metro North to Bridgeport, although I'm not sure if the Shore Line runs of Sundays.
Just a note, this is at best a ballpark in terms of the "actual" cost. Fuel obviously varies (at $3.50/gallon and getting 20 MPG, that's $0.175/mile; at 40 MPG, that's $0.0875/mile; at 10 MPG, that's $0.35/mile). Insurance, depreciation, and so on are a lot more variable (it depends on the cost/quality/age of your car and so on - a 2008 Crown Vic is going to show up as a lot less than a 2022 Mercedes). For an electric car, there's going to be rather less cost on the fuel size but the depreciation/repair accounting is going to be rather higher (Teslas are infamous for expensive repairs, for example). I'd actually argue that the IRS calculation is probably somewhat above average to reduce the number of times that folks are going to have to fight through depreciation tables and so on (not to mention that letting folks pocket the difference provides a mild incentive towards using more efficient cars).
 
Coming back to the issue of Amtrak, I think a cogent argument can be made that short to medium itinerary's fares should be removed from the yield management schemes and fix them at level where the service is very competitive with travel by road. Some state supported routes already do this. It would make sense to do more of it since road travelers get quite a bunch of subsidy in various forms anyway.
 
Coming back to the issue of Amtrak, I think a cogent argument can be made that short to medium itinerary's fares should be removed from the yield management schemes and fix them at level where the service is very competitive with travel by road. Some state supported routes already do this. It would make sense to do more of it since road travelers get quite a bunch of subsidy in various forms anyway.

While I fully agree, I think the more limiting factor, particularly on the long-distance services, is on-time performance. This is especially critical if trying to get people out of their cars - in most of the country, road travel is relatively easy and timing relatively predictable, with significant (>30 minutes) delays rare. Meanwhile, those taking long-distance trains basically have to just assume that they'll probably be delayed, and often by multiple hours. Even as a railfan, it's quite annoying, and it makes it really hard to advocate for others to take the train instead of driving, even when the schedule and destination might otherwise make the train a viable option.
 
What happened in the intervening decades is "air travel got cheap" rather than "trains got expensive" (though from a relative perspective, coming to the opposite conclusion on face value isn't unreasonable given how inflation works and how it's sometimes hard to really process stuff like that).

You are correct.

Since deregulation of airlines, airfares have dropped, as the cost of transporting one passenger one mile has dropped significantly. In part due to this, demand for air travel has soared.

The same thing has happened with freight railroads: since deregulation, the costs of transporting one ton of freight one mile has dropped significantly, as have rates. In part due to this, demand for freight rail has increased since deregulation (although precision scheduled railroading has reduced rail traffic).

The private sector and deregulation are far from perfect, and there are many grounds to criticize them. But if there were a way to have passenger rail provided in a deregulated market by the private sector, the experience of airlines and freight railroads suggests that costs and ticket prices would fall and traffic would rise.

At a minimum, I would think that passenger rail in this scenario would have much lower onboard staffs, longer trains and more use of technology to replace labor, and more efficient equipment, as airlines and freight railroads have done.

If only there was a way to ensure long-term profitability of privately-provided passenger rail, I think we’d have a very different system.
 
At the IRS estimate of operating costs for motor vehicles ($0.625 per mile), you just spent $200 for the 320-mile round trip, plus you had the aggravation of a 6-hour drive along I-95. Maybe you should have just spent the $200 on the train ticket for your daughter. Another possibility might have been to just drive to New London and have her take the Shore Line commuter train to New Haven, and then Metro North to Bridgeport, although I'm not sure if the Shore Line runs of Sundays.
The thought crossed my mind. But no, I still would drive at that price. Very few people would pay that price for that particular trip. I considered it and decided it was not worth the price.
 
We don’t ride the train just for fun. As seniors, we are no long up to driving long distances. And because of the many inconveniences and indignities associated with air travel, we avoid flying unless absolutely necessary. (The last time we flew was in 2002.)

Our big long-distance train trip for the year is always to Ohio to visit family members. (It is difficult for them to come to California so we must go to them.) These trips are contingent upon our being able to secure bedroom accommodations, despite their higher cost. (As seniors we’re no longer up to making a long-distance train trip in coach or a roomette.) Our trip this year had to be postponed when Amtrak didn’t have enough roadworthy equipment to honor the round trip bedroom reservations that we’d made and paid for months earlier.

There are undoubtedly many other seniors (including those who are NOT railfans) who would opt to travel long-distance via Amtrak and in the higher priced bedrooms if these were available.

The secret to success in any business is to identify a need and then fill it. If Amtrak management was serious about reducing operating losses and reducing ticket prices, during peak travel seasons it would arrange to add additional sleepers to its long-distance trains to accommodate the demand for roomettes and bedrooms. Instead, we get excuses why this can’t be done.

Eric & Pat
Thanks, Eric & Pat. Rosanne and I rely on Amtrak for the same reasons - we are Medicare recipients as well.

We still like to drive together, but not marathon driving.

For an overnight trip we want a family bedroom on AMTRAK, which we find a bit cheaper than a full bedroom, with a bigger lower berth and the ability to stretch out and nap during the day. Also, if the granddaughters take a roommette they can visit us in a family bedroom during the day. So we have to hope that the reservation we made for next July on the Eagle to LAX and back will be honored, or our vacation to the central coast of Cali will be ruined. We will not drive that many days and expect to be rested for the vacation, and we certainly will not fly, given airport hassles, COVID, and general discomfort.
 
Just a note, this is at best a ballpark in terms of the "actual" cost. Fuel obviously varies (at $3.50/gallon and getting 20 MPG, that's $0.175/mile; at 40 MPG, that's $0.0875/mile; at 10 MPG, that's $0.35/mile). Insurance, depreciation, and so on are a lot more variable (it depends on the cost/quality/age of your car and so on - a 2008 Crown Vic is going to show up as a lot less than a 2022 Mercedes). For an electric car, there's going to be rather less cost on the fuel size but the depreciation/repair accounting is going to be rather higher (Teslas are infamous for expensive repairs, for example). I'd actually argue that the IRS calculation is probably somewhat above average to reduce the number of times that folks are going to have to fight through depreciation tables and so on (not to mention that letting folks pocket the difference provides a mild incentive towards using more efficient cars).
AAA does their own estimate, based on the 5 most popular cars in the US, and comes up with a $/mile of $0.82 if you drive ~10k miles per year or less: https://newsroom.aaa.com/wp-content...ing-Costs-2020-Fact-Sheet-FINAL-12-9-20-2.pdf

$0.62/mile is low for many drivers, high for others. Certainly if you already own a car, the marginal cost of those miles is well less than the above.
 
Coming back to the issue of Amtrak, I think a cogent argument can be made that short to medium itinerary's fares should be removed from the yield management schemes and fix them at level where the service is very competitive with travel by road. Some state supported routes already do this. It would make sense to do more of it since road travelers get quite a bunch of subsidy in various forms anyway.
The Downeaster is one of the corridors where fares are set at a fixed price, and that definitely has made it easy for my family to rely on it as our regular mode of transport to and from our summer home in Maine. If we book at least three days in advance, the fare from Boston to Freeport is reliably $25 on all trains, which seems competitive with driving even for a family of three or four. Not to mention the train is way more fun and relaxing than driving on I-95. Clearly the state of Maine has decided to make this service accessible to as many people as possible, and it has built loyalty with customers like us.

The big challenge for us is how to get to and from Boston to reach the Downeaster. From our home base in upstate New York, the eastbound Lake Shore's late (often, very late) arrival into Boston makes it useless as a connection. Usually we just drive 100-plus miles to the commuter rail terminal in Wachusett, which takes us into North Station, avoiding the dreaded transfer from South Station. If we start out from PA, where we also have family, we can use the NEC, but getting a decent fare requires booking weeks in advance. At that point one can sometimes get an Acela seat from NYP to BOS for $58 per person, which is still probably more than driving for a party of three or four. But when the fare bumps up to $100-plus per person even for the slower regional trains, we wind up driving.
 
Coming back to the issue of Amtrak, I think a cogent argument can be made that short to medium itinerary's fares should be removed from the yield management schemes and fix them at level where the service is very competitive with travel by road. Some state supported routes already do this. It would make sense to do more of it since road travelers get quite a bunch of subsidy in various forms anyway.
I'd probably quasi-fix the fares - basically, fix them most of the time (and allow the purchase of multi-ride tickets at a [more sane] fixed rate), but allow for a second "bucket" to exist for peak times/trains. Basically, peak/off-peak faring makes sense, if just to try and nudge at least some ridership out of more in-demand trains.
 
I'd probably quasi-fix the fares - basically, fix them most of the time (and allow the purchase of multi-ride tickets at a [more sane] fixed rate), but allow for a second "bucket" to exist for peak times/trains. Basically, peak/off-peak faring makes sense, if just to try and nudge at least some ridership out of more in-demand trains.
This is kind of like the system Amtrak had 30 years ago before they adopted the yield management approach. In the Northeast and Empire corridors, there were set fares between each city pair, and each pair had a discounted fare that was available except at peak times. So if your scheduled departure was between 11a and 7p on Fridays or Sundays, you had to pay full fare, and at other times, you could get the discounted fare. Of course, most of the trains were unreserved, and they sometimes had people standing at busy times.
 
Last edited:
Just a note, this is at best a ballpark in terms of the "actual" cost. Fuel obviously varies (at $3.50/gallon and getting 20 MPG, that's $0.175/mile; at 40 MPG, that's $0.0875/mile; at 10 MPG, that's $0.35/mile). Insurance, depreciation, and so on are a lot more variable (it depends on the cost/quality/age of your car and so on - a 2008 Crown Vic is going to show up as a lot less than a 2022 Mercedes). For an electric car, there's going to be rather less cost on the fuel size but the depreciation/repair accounting is going to be rather higher (Teslas are infamous for expensive repairs, for example). I'd actually argue that the IRS calculation is probably somewhat above average to reduce the number of times that folks are going to have to fight through depreciation tables and so on (not to mention that letting folks pocket the difference provides a mild incentive towards using more efficient cars).

If I owned a Tesla or 2022 Mercedes then dropping $200 on a 160 mile train trip probably wouldn't be an issue. Amtrak definitely provides a service in the Northeast as I use it frequently. One area where it lacks is last minute travel or anything booked within 1 week of departure. It's too costly for most people including myself. Whether this is a permanent status or just a pandemic phenomenon I don't know. Amtrak has had dynamic pricing for sometime but they have taken it to a new level. So in keeping with the topic, taking an Amtrak train in the Northeast can be more expensive than driving a car. I recently attended a birthday party in New York where many of the guests opted to drive as the cost of Amtrak tickets from Boston to New York were too expensive. Granted they waited too long to book their train tickets.
 
I just looked up two trips I want to take: ATL to LAX, 4 nights in a roomette, roughly 3,600 miles, $2,175.

Or, fly to Ft. Lauderdale and take a cruise ship to London ,13 nights in a real room with private bath, 5,000+ miles: $1,500.

Interesting comparison.
 
I just looked up two trips I want to take: ATL to LAX, 4 nights in a roomette, roughly 3,600 miles, $2,175.

Or, fly to Ft. Lauderdale and take a cruise ship to London ,13 nights in a real room with private bath, 5,000+ miles: $1,500.

Interesting comparison.
It's almost as if train travel is the most expensive mode of travel in the United States.
 
From my perspective, Amtrak's dynamic pricing functions as a way to ration scarce capacity. I don't fault management for trying to maximize revenue from the trains they operate. But clearly on both the corridor and long-distance trains, there are many people who'd happily travel by train at Amtrak's lower-bucket fares who are discouraged from doing so by the higher prices that kick in as more space as sold. Many of those people wind up traveling by other modes or simply stay home.

Which does underscore that lack of equipment and capacity is one factor that most limits Amtrak's ability to capture a larger share of the traveling public. The other is the many travel markets that Amtrak just doesn't serve at all -- or serves very poorly. Putting the latter issue aside, it seems clear that even on the existing route network, Amtrak could find many more willing riders if it had enough equipment to accommodate them, which in turn would likely result in better fares for reservations made days rather than weeks in advance.
 
it seems clear that even on the existing route network, Amtrak could find many more willing riders if it had enough equipment to accommodate them, which in turn would likely result in better fares for reservations made days rather than weeks in advance.
What is the current availability of bedrooms on long-distance trains? Are long distance trains still limited to one “base” sleeper per run or have extra sleepers been added since last summer? Are bedrooms still selling out months in advance? Are the bedroom reservations that were made for this year’s holiday travel season being honored or are people being downgraded to coach, as was the case last summer?

A different thread offered an article written by someone who traveled in coach from Los Angeles to Seattle on the Coast Starlight just to see what it was like. (The descriptions of the bathroom facilities after a day or so, the "travel grunge" due to the lack of shower facilities for coach passengers, and the quality of the food served in the lounge car were enough to discourage all but the most hardy from traveling long distance in coach.)

If people are willing to pay inflated prices for bedrooms to be sure of having their own showers and toilets on a long-distance train trip, then Amtrak should make a best effort to provide as many bedrooms as there are requests for, even if this means adding extra sleepers to the consist to meet the demand. The secret to success in any business is to identify a need and then fill it.
 
Last edited:
What is the current availability of bedrooms on long-distance trains? Are long distance trains still limited to one “base” sleeper per run or have extra sleepers been added since last summer? Are bedrooms still selling out months in advance? Are the bedroom reservations that were made for this year’s holiday travel season being honored or are people being downgraded to coach, as was the case last summer?

A different thread offered an article written by someone who traveled in coach from Los Angeles to Seattle on the Coast Starlight just to see what it was like. (The descriptions of the bathroom facilities after a day or so, the "travel grunge" due to the lack of shower facilities for coach passengers, and the quality of the food served in the lounge car were enough to discourage all but the most hardy from traveling long distance in coach.)

If people are willing to pay inflated prices for bedrooms to be sure of having their own showers and toilets on a long-distance train trip, then Amtrak should make a best effort to provide as many bedrooms as there are requests for, even if this means adding extra sleepers to the consist to meet the demand. The secret to success in any business is to identify a need and then fill it.
Well, I do think the law of supply and demand still applies, though it applies within the framework of a transportation structure that's heavily shaped by government policy and priorities (and, yes, politics). Some of Amtrak's current problems -- e.g., the sleepers pulled from service with travelers downgraded to coach with little or no notice -- are self-inflicted or, at best, tied to broader economic issues (e.g., labor shortages) in the aftermath of the pandemic. But even if Amtrak were able to press every piece of sidelined/mothballed equipment into service, it wouldn't come close to having the capacity to carry everyone who might like to go by train, especially at busy travel times.

Now, if Congress were to decide that trains are an environmentally beneficial way for people to travel and therefore invest in enough new equipment for Amtrak to double its carrying capacity, and if long-distance services were a part of that investment, then sleeper prices would become somewhat less inflated and fares across the board could be lower, which would encourage a lot more people to ride.
 
Unfortunately all the capital in the world will not cause more trains to run if there is insufficient operating budget. That has been the perennial problem for Amtrak and many commuter agencies. They are flush with capital and insufficient operating budget, so much so that commuter agencies like NJT forever convert capital budget money to keep the operations lights on. No one seems to be interested in fixing this imbalance because it goes against the accepted wisdom that given enough capital there will be adequate operations money from farebox. This has not been the case for most passenger services for a long long time, but who can argue with ideology?
 
If the revenues from adding a car to a train exceed the cost of acquiring and operating the car, which should be the case for sleeping cars, then Amtrak shouldn’t need government funding to do so; it should be able to borrow the money from a lender (a bank, a speciality finance company, etc.) to acquire the car, and the lender would have a lien on the car. That’s how non-Amtrak railroads, and airlines, acquire equipment.
 
No one seems to be interested in fixing this imbalance because "it goes against the accepted wisdom that given enough capital there will be adequate operations money from farebox".

That's because the mentality is: "If you keep saying it - it must be true or will happen" ... even though time has proven otherwise.

Use that same ideology for the Interstate Highway System and see how many roads get widened. No one seem to care that the highways do not pay for themselves or make a profit ... but they think Amtrak should and that, if they say so, it will happen.
 
Unfortunately all the capital in the world will not cause more trains to run if there is insufficient operating budget. That has been the perennial problem for Amtrak and many commuter agencies. They are flush with capital and insufficient operating budget, so much so that commuter agencies like NJT forever convert capital budget money to keep the operations lights on. No one seems to be interested in fixing this imbalance because it goes against the accepted wisdom that given enough capital there will be adequate operations money from farebox. This has not been the case for most passenger services for a long long time, but who can argue with ideology?
No doubt this is true, and obviously adding more trains -- such as converting some once-a-day segments to twice-daily service, as discussed in other threads -- would mean additional operating costs that would have to be funded if cost recovery from fares isn't 100 percent, which it never will be. Even so, given that the marginal cost is low to add cars to existing departures, a larger pool of rolling stock could boost cost recovery on existing departures -- and would enable Amtrak to serve more riders at busy times, which would have value in broadening public support for the service. Having trains that are sold out or only have seats/rooms at sky-high fares does the opposite: It discourages would-be riders and leaves some of them feeling that train service isn't accessible or affordable.
 
If the revenues from adding a car to a train exceed the cost of acquiring and operating the car, which should be the case for sleeping cars, then Amtrak shouldn’t need government funding to do so; it should be able to borrow the money from a lender (a bank, a speciality finance company, etc.) to acquire the car, and the lender would have a lien on the car. That’s how non-Amtrak railroads, and airlines, acquire equipment.
Amtrak is the sole customer for long-distance (sleeper) passenger equipment on US rail and under US regulations. There is no rolling stock manufacturer with US-legal passenger coaches "in stock" for order or delivery. Amtrak has to order unique cars, and cannot simply "acquire a car". They have to find a manufacturer with capacity available in their order book, and must make a large enough order to cover the engineering and design costs of those cars. This probably means an order of in excess of 40-50 cars and delivery in probably 3-5 years.

Even assuming they could do so, it would also be hard to find a lender looking to back a 30+ year investment on said cars, given the weak nature of the market, and the single customer involved. Airline finance is fairly robust, but most countries follow the lead of US or EU regulators for planes. A plane built for United in the US can be sold to an airline in Asia and used there profitably. US rolling stock is not legal for use in any other market, and vice versa.

Like most things in life, it's complicated.
 
Like most things in life, it's complicated.
I'm sympathetic to this, really. Especially in the US, the question of train manufacturing, rolling stock (or lack thereof), is a big one, and poses a massive hurdle for Amtrak to overcome, however....

There comes a point where complicated things need to simply be figured out. Humans, companies, and other groups have figured out complex problems for centuries, oftentimes confounding onlookers as to how it was possible. As problems go, this isn't exactly unsolvable.

Even assuming they could do so, it would also be hard to find a lender looking to back a 30+ year investment on said cars, given the weak nature of the market, and the single customer involved. Airline finance is fairly robust, but most countries follow the lead of US or EU regulators for planes. A plane built for United in the US can be sold to an airline in Asia and used there profitably. US rolling stock is not legal for use in any other market, and vice versa.
Given they have a massive influx of cash, perhaps Amtrak could think smart, and order more trains than needed... Ordering only 25 VII sleepers was a short-sighted move at best.
 
I'm sympathetic to this, really. Especially in the US, the question of train manufacturing, rolling stock (or lack thereof), is a big one, and poses a massive hurdle for Amtrak to overcome, however....

There comes a point where complicated things need to simply be figured out. Humans, companies, and other groups have figured out complex problems for centuries, oftentimes confounding onlookers as to how it was possible. As problems go, this isn't exactly unsolvable.


Given they have a massive influx of cash, perhaps Amtrak could think smart, and order more trains than needed... Ordering only 25 VII sleepers was a short-sighted move at best.
The order for Superliner Sightseer Lounges (25 if I recall correctly) was a few cars too small in the first place, thereby setting up scrabbling between routes ever since.

With a handful of extra V sleepers the New Orleans barrier to transcon travel could be solved by telling the reservation computer that there is a Sleeping Car Only Train 419 that arrives in New Orleans in the morning after Train 19 arrives and a counterpart Train 420 that departs the night before Train 20 departs. There are amazing things that can be done without adding train miles -- if equipment is available.
 
Amtrak is the sole customer for long-distance (sleeper) passenger equipment on US rail and under US regulations. There is no rolling stock manufacturer with US-legal passenger coaches "in stock" for order or delivery. Amtrak has to order unique cars, and cannot simply "acquire a car". They have to find a manufacturer with capacity available in their order book, and must make a large enough order to cover the engineering and design costs of those cars. This probably means an order of in excess of 40-50 cars and delivery in probably 3-5 years.

Even assuming they could do so, it would also be hard to find a lender looking to back a 30+ year investment on said cars, given the weak nature of the market, and the single customer involved. Airline finance is fairly robust, but most countries follow the lead of US or EU regulators for planes. A plane built for United in the US can be sold to an airline in Asia and used there profitably. US rolling stock is not legal for use in any other market, and vice versa.

Like most things in life, it's complicated.
It’s complicated but Amtrak has done plenty of financings with private sector finance companies before in connection with equipment acquisitions. What I propose is nothing new for Amtrak.
 
As recently as 2018, almost the entire Acela fleet was leased as well as a considerable number of Superliners. The most recent fleet report says they are down to 47 leased Superliners and two Acela sets.
 
Back
Top