Ridership on LD vs. State/NEC Trains

[Skim]

I notice LD trains tend to be the preference of train buffs on this board. Among the general public, however, do LD or the shorter (mostly state) routes tend to have more ridership?

Being near St. Louis, we have mostly state routes nearby and most of the ridership goes there as opposed to the Texas Eagle, but I'm curious as to the overall system.

 

[Anderson]

The breakout last year was:

11.4m NEC (3.4m Acela, 8.0m NER)

15.0m short corridors (remember, 27 routes ranging from 1x daily to over 10x daily)

4.7m LD (13 1x daily, 2 3x weekly)

Basically, the short corridors have a lot of ridership...but that's because there are a lot of trains in those categories.

 

[zephyr17]

Is that raw passenger counts (boardings) or passenger miles?

 

[BCL]

I think there's a lot of dicussion of LD trains here because passengers are looking to prepare for something that's unknown. The people taking the regional train routes settle into a routine and don't really need to ask anyone after a few trips.

 

[Skim]

I think there's a lot of dicussion of LD trains here because passengers are looking to prepare for something that's unknown. The people taking the regional train routes settle into a routine and don't really need to ask anyone after a few trips.

The latter sounds like most of the Lincoln Service and MRR passengers, myself included.

Thanks for the numbers and other info.

 

[Paulus]

Is that raw passenger counts (boardings) or passenger miles?

Raw passenger counts. By passenger mile:

NEC: 1,892 million

Corridor: 1,976 million

Long distance: 2,931 million

 

[George Harris]

Is that raw passenger counts (boardings) or passenger miles?

Raw passenger counts. By passenger mile:

NEC: 1,892 million

Corridor: 1,976 million

Long distance: 2,931 million

Now let's put the passenger count in the same post:

The breakout last year was:11.4m NEC (3.4m Acela, 8.0m NER)

15.0m short corridors (remember, 27 routes ranging from 1x daily to over 10x daily)

4.7m LD (13 1x daily, 2 3x weekly)

And you see how the constant emphasis on passenger count versus miles skews the results.

A little simple arithmetic gives you the average miles per passenger on each of these and that comes out to:

166 miles each on the NEC

132 miles each on the short corridors

624 miles each on the long distance trains.

Therefore for miles traveled, each long distance passenger is equal to 3.76 northeast corridor passengers and 4.74 short corridor passengers.

I regard this constant emphasis on raw passenger numbers as an attempt to make the long distance trains look worse than their reality and make the northeast corridor in particular look better than its reality.

 

[railiner]

Is that raw passenger counts (boardings) or passenger miles?

Raw passenger counts. By passenger mile:

NEC: 1,892 million

Corridor: 1,976 million

Long distance: 2,931 million

Now let's put the passenger count in the same post:

The breakout last year was:

11.4m NEC (3.4m Acela, 8.0m NER)

15.0m short corridors (remember, 27 routes ranging from 1x daily to over 10x daily)

4.7m LD (13 1x daily, 2 3x weekly)

And you see how the constant emphasis on passenger count versus miles skews the results.

A little simple arithmetic gives you the average miles per passenger on each of these and that comes out to:

166 miles each on the NEC

132 miles each on the short corridors

624 miles each on the long distance trains.

Therefore for miles traveled, each long distance passenger is equal to 3.76 northeast corridor passengers and 4.74 short corridor passengers.

I regard this constant emphasis on raw passenger numbers as an attempt to make the long distance trains look worse than their reality and make the northeast corridor in particular look better than its reality.

Like anything, statistics can be skewed to favor whatever point is trying to be made.

What would be the results if such things as revenue per train mile were compared? There are so many ways to measure things......

 

[Anderson]

George,
You make a decent point, but I feel compelled to point out that yield-per-passenger-mile is something else that needs to be looked at. It's not hard (and not rare) for a coach ticket RVR-NYP to cost as much as a coach ticket RVR-ORL. Using the raw mile counts and the revenue figures, you get:

NEC: 1,892m passenger miles on $1,043.8m in ticket revenue, giving about $0.552/mile in revenue.
Short corridors: 1,976m passenger miles on $447.5m in ticket revenue, giving about $.226/mile in ticket revenue.
LD: 2,931m passenger miles on $516.0m in ticket revenue, giving about $0.176/mile in revenue.

The result of this is that your average NEC passenger is forking over around $91 for their ticket while your average LD passenger is forking over $109 for their ticket...but traveling several times the distance for that extra $18. If you kick the Auto Train out of the LD equation, half of that difference vanishes ($99/ticket instead of $109/ticket).

Mind you, the state situation is more complicated (especially with PRIIA going into effect), as some states are "sitting" on some fares (such as NY and the Adirondack) for political reasons, and Amtrak is getting additional non-ticket subsidy revenue on some routes.

Passenger counts are important, yes, if only for political reasons. But as I've noted, I find revenue and per-passenger-revenue to be the more important figures (and more informative, even within a given route) since they tell a lot of stories on their own. Adding in passenger miles for various routes (as well as, quite frankly, actual load factors and information on known capacity bottlenecks) would be nice as well, but I don't have ready access to that information.

One observation for chewing over: The NEC showed a pre-OPEB "fully allocated" gain of $298.0m in FY12, which comes to $.158/mile. Removing that from revenue comes to costs of $.394/passenger mile, on about a 51% load factor. LD trains showed a pre-OPEB "fully allocated" loss of $575.1m, which comes to $.196/mile. Adding that to revenue comes to $.372/passenger mile, on about a 62% load factor. I'm a bit burned out right now to try and calculate what that means in seat-mile terms, but...well, chew over at will.

 

[jis]

But one big point missed is that when you shut down the NEC specially around New York, it starts affecting everyone's life by gumming up the highways in the area. When you shut down an LD train the effect is not as noticeable in general.

 

[cirdan]

But one big point missed is that when you shut down the NEC specially around New York, it starts affecting everyone's life by gumming up the highways in the area. When you shut down an LD train the effect is not as noticeable in general.

If you look at the bigger economic implications of that, and propose a system where everybody pays the full costs of the infrastructure they need (roads, trains, anything else) then there would probably continue to be both roads and trains around NYC (even though they would be more expensiev to use), but the LD trains would vanish and all but maybe a handful of interstate highways would crumble and become dirt roads as nobody would be paying enough to maintain them. This would affect the economic prosperity of smaller towns and rural areas and more businesses would relocate to the big citiies where the infrastructure would be better. That wouldn't necessariyl be a good thing as it would further drive up propoerty prices and rents in those areas and add to pollution, while rural property prices would drop further. This is already observable in many third world countries, where the capital city typically has more than half the country's total population and the imbalance continues to grow.

By actually putting money into rural areas, the government is therefore keeping the pressure off cities.

 

[jis]

I am all for keeping the LD network healthy and even growing it. I just find the arguments about robbing NEC to pay for it somewhat misguided.

(null)

 

[Paulus]

And you see how the constant emphasis on passenger count versus miles skews the results.

A little simple arithmetic gives you the average miles per passenger on each of these and that comes out to:

166 miles each on the NEC

132 miles each on the short corridors

624 miles each on the long distance trains.

Therefore for miles traveled, each long distance passenger is equal to 3.76 northeast corridor passengers and 4.74 short corridor passengers.

I regard this constant emphasis on raw passenger numbers as an attempt to make the long distance trains look worse than their reality and make the northeast corridor in particular look better than its reality.

Emphasis on passenger-miles simply serves to skew in the other direction. Of course there are going to be higher passenger miles on the long distance trains; they have stops further spaced than some corridors are long. Furthermore, corridors are where cities are spaced closely enough that the trip can be made within a reasonable timeframe. But the overemphasis on passenger-miles leads to the ludicrous suggestion that a single trip to LAUS-CUS is worth 10 trips WAS-NYP or 20 LAUS-SAN.

One observation for chewing over: The NEC showed a pre-OPEB "fully allocated" gain of $298.0m in FY12, which comes to $.158/mile. Removing that from revenue comes to costs of $.394/passenger mile, on about a 51% load factor. LD trains showed a pre-OPEB "fully allocated" loss of $575.1m, which comes to $.196/mile. Adding that to revenue comes to $.372/passenger mile, on about a 62% load factor. I'm a bit burned out right now to try and calculate what that means in seat-mile terms, but...well, chew over at will.

Would this help? Note that it uses all costs including OPEBs.

 

[Anderson]

But one big point missed is that when you shut down the NEC specially around New York, it starts affecting everyone's life by gumming up the highways in the area. When you shut down an LD train the effect is not as noticeable in general.

If you look at the bigger economic implications of that, and propose a system where everybody pays the full costs of the infrastructure they need (roads, trains, anything else) then there would probably continue to be both roads and trains around NYC (even though they would be more expensiev to use), but the LD trains would vanish and all but maybe a handful of interstate highways would crumble and become dirt roads as nobody would be paying enough to maintain them. This would affect the economic prosperity of smaller towns and rural areas and more businesses would relocate to the big citiies where the infrastructure would be better. That wouldn't necessariyl be a good thing as it would further drive up propoerty prices and rents in those areas and add to pollution, while rural property prices would drop further. This is already observable in many third world countries, where the capital city typically has more than half the country's total population and the imbalance continues to grow.

By actually putting money into rural areas, the government is therefore keeping the pressure off cities.

Honestly, this hypothetical probably isn't 100% right...though it's worth thinking over if the Highway Trust Fund slams into a wall in a few years like it is projected to. What you'd probably see is a cutback in the interstate highway system combined with widespread tolling, particularly out west where you've got a non-trivial amount of pass-through trucking traffic (though some interstates like I-95 (VA-GA), I-81 (VA, PA), and I-80 (PA) in the east would also be susceptible to this). This would be to the benefit of the freight railroads at the expense of long-distance trucking.

As to passenger rail, that probably depends on a combination of Amtrak's position when that happens, what the tolling numbers end up at on sooner rather than later, and just how fast the wheels come off the interstates. To make a serious guess, I'm inclined to bet that Amtrak cross-subsidizes some of the LD trains amid a spike in load factors for short-to-medium distance passengers as well as on some key routes (especially if I-95 gets a big total toll...going off something akin to the the Pennsylvania Turnpike's roughly $.11/mile toll, driving from DC to Orlando would probably have something like a $95 toll each way, which would effectively make driving to FL alone uncompetitive with a coach fare...at $3/gallon and 40 MPG, you'd have close to $65 in gas in a very efficient car; that goes to $85 at 30 MPG, and the combined $160-180 each way in non-maintenance driving costs would just be a killer).

Beyond that, it'd come down to a waiting game: Airline service on a number of runs would either cease to exist or become uncompetitive as well, after all.

 

[jis]

Part of the problem with the argument about subsidizing rural life helps relieve pressure on the urban and suburban areas is that the numbers do not bear that out. It certainly hasn't significantly slowed the move of the population in the US from rural to urban areas. The general flow of population continues unabated towards urban and suburban areas. Partly this may be because of the inherent efficieciencies, energy and otherwise of such conglomerations over a population spread out far and wide requiring long low volume and inefficient transportation links. This work as long as energy and transportation costs remain low. But once that goes away the move towards higher density population centers is inevitable. And then the primary question becomes what is the most efficient way of providing transportation within those conurberations and between them. In the long run the smaller towns and rural areas will see inevitable decline.

Actually third world countries have more spread out population than say a country like France. Countries like India are way more rural than even the US already. See this spreadsheet for some real data and projections. This is most likely because subsistence livlihood is still an acceptable mode of life in those countries. It is not in the US and Europe, and there are progressively fewer high quality jobs available in the rural hinterland. The new generation is in general moving out to urban areas creating a renaissance of the urban core in many cities in the US. Unfortunately the planners in the US and specially Congress has romantic glasses on. They will wake up from their fantasy one of these days hopefully, before it is too late. But this is getting far afield.

Within this context, I believe that LD trains serve a purpose of connecting a large number of rural areas along a relatively high density communication corridor between urban conurberations most efficiently, more efficiently than road or air. There are other off the beaten path rural areas that require less efficient means of connection. For connecting large conurberations that are far removed from each other possibly some mix of rail and air makes the most sense. But a country with a well developed highway system is possibly at a economic disadvantage in its ability to develop a rational transportation policy matching the demographic trend most efficiently. However, the net adverse effect of such will possibly become visible way after my generation is gone.

 

[cirdan]

Part of the problem with the argument about subsidizing rural life helps relieve pressure on the urban and suburban areas is that the numbers do not bear that out. It certainly hasn't significantly slowed the move of the population in the US from rural to urban areas. The general flow of population continues unabated towards urban and suburban areas. Partly this may be because of the inherent efficieciencies, energy and otherwise of such conglomerations over a population spread out far and wide requiring long low volume and inefficient transportation links. This work as long as energy and transportation costs remain low. But once that goes away the move towards higher density population centers is inevitable. And then the primary question becomes what is the most efficient way of providing transportation within those conurberations and between them. In the long run the smaller towns and rural areas will see inevitable decline.
Actually third world countries have more spread out population than say a country like France. Countries like India are way more rural than even the US already. See this spreadsheet for some real data and projections. This is most likely because subsistence livlihood is still an acceptable mode of life in those countries. It is not in the US and Europe, and there are progressively fewer high quality jobs available in the rural hinterland. The new generation is in general moving out to urban areas creating a renaissance of the urban core in many cities in the US. Unfortunately the planners in the US and specially Congress has romantic glasses on. They will wake up from their fantasy one of these days hopefully, before it is too late. But this is getting far afield.

Within this context, I believe that LD trains serve a purpose of connecting a large number of rural areas along a relatively high density communication corridor between urban conurberations most efficiently, more efficiently than road or air. There are other off the beaten path rural areas that require less efficient means of connection. For connecting large conurberations that are far removed from each other possibly some mix of rail and air makes the most sense. But a country with a well developed highway system is possibly at a economic disadvantage in its ability to develop a rational transportation policy matching the demographic trend most efficiently. However, the net adverse effect of such will possibly become visible way after my generation is gone.

The US may be a special case in this respect. Settling the hinterland has historically been an important objective (I'm talking about 100 years ago and more) as it was only by having their own people there that the government could functionally claim sovereignty over that land (whereas if they left it unsettled, somebody else could claim it (or befriend the indigineous people who lived there), borders being much less clearly defined than they are today). This thinking may have grown into a reflex and so still pervades policy despite the world having changed. The Transcontinental Railroad was built over fears that the vast area of nothingness between east and west could lead the country to split apart (and the Civil War had proved that this was indeed a real possibility) and all sorts of laws were put in place making it easy for people to set out west and claim as much land as they could put a fence around.

India probably still has a major agricultural revolution ahead of it. How much longer can subsistence farmers hold out if farming becomes increasingly industrialized? I predict that the combined effects of higher standards and costs of living, pressure on land prices and pressure on the costs of food will lead to fewer but bigger farms employing fewer people, much as has happened in the US and Europe. It won't happen over night of course, but it's the course of development.

Having said that, I was talking to a John Deere sales representative (in Europe) not long and he was telling me the company is seeing increased sales in small and micro tractors, which he interpreted as reflecting a revival of small subsistence-style farms with mid-aged people typically dropping out of the rat race and stress of salary life, maybe after a burn-out or other transformational experience, and now doing some e-based consulting from home while running a small farm on the side.

 

[Anderson]

I would point out, however, several points:
(1) Many farms like that are likely being operated either at a loss or as relatively incidental income rather than as a way of supporting a family. They're really probably a bit more in the vein of either an expensive hobby (say, a substitute for a fancy sports car, and potentially less expensive) than a "real" business. A plethora of tax incentives may play a role here, too, depending on the country.

(2) At least in the US, were that to happen, my presumption would be that the land was the investment and that the farming was a good way to cover taxes on the land, get some benefit from a wholesome hobby, and potentially make some side income.

(3) Even presuming such a tendency, it's unlikely that these folks are going to make up a significant portion of the population. This is rarely going to be something cheap to "get into", so again...you're looking at folks in the top 5-10% of the population (at most) who could afford to play that "game".

(4) Finally, there is a possibility that land in Europe could "get cheap" because of demographic trends (i.e. declining/aging population) combined with farmer-friendly legislation that, again, piles on the tax preferences.

My point is that this isn't likely to be a world-shaking trend. A million burned-out white collar folks move to the countryside and get farms...while another 20-30 million end up in the cities/suburbs through population growth and migration.

India, China, and other countries are likely to keep urbanizing, with a limited dose of suburban/exurban growth on the parts of the wealthy classes (who end up desiring a bit of space)...though I suspect that you may get a return to "streetcar suburbs" if for no other reason than so servants can get to work. It won't be anything like the US saw in the last 50-60 years ("drive to qualify" suburbs), but you'll see a bit all the same.

At the same time, it will be very interesting to see how China's transit Hail Mary plays out and how far out the networks of suburban trains, metros, etc. end up going (i.e. how many ring roads out the lines penetrate from Beijing). The fact that China is running "suburban services" with OBS from Beijing should say something, however.

 

[cirdan]

It's as you say, largely a question of incentives through tax breaks. If the government wants more people to move to a certain area, they can find reasons to make more people want to live there.

I wouldn't underestimate the number of people though, who want to ot out of the fast-lane society.

I am nowhere near the top 10% you mention, but have just as an exercise looked at how much the equity in my house is worth, and how much of a farm that would buy me in somewhere like Argentina. I was pleasantly surprised.

 

[Nathanael]

I have met a fair number of family farmers, and they tend to be a little bit fanatical. You do not (currently) make a good living that way, but if it's the life you want to lead, you *really* want to lead it. I'm absolutely willing to subsidize them to some extent, because it gets us better food than subsidizing Monsanto.

 

[Nathanael]

The result of this is that your average NEC passenger is forking over around $91 for their ticket while your average LD passenger is forking over $109 for their ticket...

I'll chime in on the "statistics can be made to say whatever you want" front by pointing out that averages are misleading. The variance in the price of "long-distance" train tickets is MASSIVE. I have routinely paid more to go from Syracuse, NY to Chicago (techically "long distance") than from Chicago to Los Angeles. Which costs more for Amtrak to operate? I suspect the latter.

There is absolutely no point in mixing up the single-overnights in the East and Midwest with the double-overnights in the Mountain West -- they have very different characteristics, and should not be treated as a single "long-distance" lump.

Personally, I find some other metrics more interesting. Load factor (passenger miles / seat miles). Passengers/train. Given induced demand, and given that some corridors run more trains per day than others, this is one of the few ways of getting at whether there is "really" more demand on one route than another.

 

[cirdan]

The result of this is that your average NEC passenger is forking over around $91 for their ticket while your average LD passenger is forking over $109 for their ticket...

I'll chime in on the "statistics can be made to say whatever you want" front by pointing out that averages are misleading. The variance in the price of "long-distance" train tickets is MASSIVE. I have routinely paid more to go from Syracuse, NY to Chicago (techically "long distance") than from Chicago to Los Angeles. Which costs more for Amtrak to operate? I suspect the latter.

There is absolutely no point in mixing up the single-overnights in the East and Midwest with the double-overnights in the Mountain West -- they have very different characteristics, and should not be treated as a single "long-distance" lump.

Personally, I find some other metrics more interesting. Load factor (passenger miles / seat miles). Passengers/train. Given induced demand, and given that some corridors run more trains per day than others, this is one of the few ways of getting at whether there is "really" more demand on one route than another.

True, you can easily pay a 4- figure sum for a one-way ticket on some LD routes in sleeper.

So maybe median fares would be of more relevance than mean fares?

Costs, earnings and cost recovery are figures that mean frightfully little when looked at in a vacuum.

To put them into context, you have to define the purpose of the railroad.

Is the purpose of Amtrak to make money, or at least lose as little as possible?

Is the purpose of Amtrak to provide connectivity and transportation to as many people as possible?

Is the purpose of Amtrak to relieve pressure on roads and airports?

To some extent you can respond to all of the above questions with yes, and you can also respond with no. Its all a question of perspective. In reality Amtrak seeks to balance all these goals, and because trying to meet one goal better automatically means meeting another less well, its all a game of trade-offs.

 

[Anderson]

From what I can tell, part of the problem is that on the single-level LD trains in particular, the capacity pressure has just gotten insane over the last few years. There's probably a report somewhere saying in no uncertain terms that if Amtrak gets additional cars, they can fill them much of the time.

With the western trains (and the Cap), it's a bit more of a roll of the dice because the lower buckets tend to actually still exist at times. I once spent as much CHI-WAS-RVR as I did NFK-WAS-CHI-SEA...but one direction was in the middle of a random early spring week while the other was Easter Monday/Tuesday.

 

[Bus Nut]

Isn't it a little premature to say that NEC runs a profit on fully allocated costs when the massive costs of infrastructure repair on the NEC line are not being taken into account?

The LD lines must pay track access fees which may or may not cover the opportunity costs (operating) and share of maintenance costs (capital) to the line.

While above the rail operating costs for Amtrak are fairly easy to account for, capital maintenance costs are contingent and the funding is all over the place. The NEC is well overdue for capital investment and that is going to be very expensive.

It's nice that NEC can run operating surpluses, but that is precisely because of the massive capital investments made over the last century.

 

[Paulus]

Isn't it a little premature to say that NEC runs a profit on fully allocated costs when the massive costs of infrastructure repair on the NEC line are not being taken into account?

The NEC actually appears to be funding its own costs at the moment, though I don't think it's making enough money to fund the billions in deferred maintenance that is required.

The LD lines must pay track access fees which may or may not cover the opportunity costs (operating) and share of maintenance costs (capital) to the line.

They don't pay anywhere near the opportunity costs.

 

[Anderson]

Isn't it a little premature to say that NEC runs a profit on fully allocated costs when the massive costs of infrastructure repair on the NEC line are not being taken into account?

The NEC actually appears to be funding its own costs at the moment, though I don't think it's making enough money to fund the billions in deferred maintenance that is required.

The LD lines must pay track access fees which may or may not cover the opportunity costs (operating) and share of maintenance costs (capital) to the line.

They don't pay anywhere near the opportunity costs.

1) On the NEC, I think a fair question is this: If Amtrak were magically given a round of SOGR improvements and the replacement/renovation of basically all pre-WW2 infrastructure elements along with some limited capacity bottleneck fixes, would the current revenue plus increases due to speed and OTP improvements be enough to cover keeping the NEC in good shape thereafter? I see this question as valid in no small part due to the fact that Amtrak inherited a lot of deferred (or bungled) maintenance from the Penn Central.

2) It depends on the route. The opportunity cost is covered if the slot(s) taken by the train aren't otherwise in use; if they would be used otherwise, though, it's going to depend on how much revenue the railroad could get from that slot.