Trains.com

Amtrak Sleeping Car Pricing....seems like it is still broken.

5232 views
18 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    June 2009
  • From: Dallas, TX
  • 6,830 posts
Amtrak Sleeping Car Pricing....seems like it is still broken.
Posted by CMStPnP on Sunday, July 6, 2014 2:27 PM

Why is a Sleeper always cheaper Chicago to DC on the Capitol Limited vs the Cardinal?     Some would argue that is because of yield management and supply of sleepers.    I realize the Cardinal is probably restricted to single level sleepers.     However this raises a question.    If sleeper revenue generated per train is higher on one train, why doesn't the sleeper supply shift to that train and equalize the sleeping car price?     Is the shortage of single level sleepers that tight that the Cardinal cannot couple onto one or two more?     Curious what the Sleeping Car utiliization is on each Amtrak train and if Amtrak is running trains with all the sleeper space full while at the sametime running a train with most of the sleeping car space empty.........which I suspect could be happening at times.

  • Member since
    December 2007
  • From: Georgia USA SW of Atlanta
  • 11,827 posts
Posted by blue streak 1 on Sunday, July 6, 2014 3:11 PM

Capitol  -  max beds per car  =  40 ?  2 cars  =  80

Cardinal  max beds per car  =  32 ? normally 1 car Reported now sometimes 2 cars  = 64

Supply & demand .  Further Cardinal scheduled 6-1//2 longer WASH - CHI so at least one more meal service ? 

  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Sunday, July 6, 2014 4:09 PM

According to Amtrak's car diagrams, which includes recommendations for the number of adults in each room, the Viewliner sleeping car has 11 roomettes, two bedrooms, and one ADA compliant room for sale.  One of the roomettes is held for the sleeping car attendant. If Amtrak's recommendations for occupancy are followed, there are 28 available spaces in the car.

There are three cars a week westbound and three cars a week eastbound on the Cardinal for a total of 6 cars per week. This assumes one car per train, which I understand is the normal consist.  This means that there are 8,736 rooms for sale per year.  In FY13 the Cardinal carried 6,882 sleeping car passengers over at least one leg of the journey.  Thus, the average occupancy rate would have been 78.8 per cent, assuming an average of one passenger per room, it would have been 39.4 per cent assuming an average of two passengers per room, and it would have been 59.1 per cent assuming an average of 1.5 passengers per room.  

If the Cardinal carried two sleepers during FY13, the number of rooms would have been 17,742 and the averages shown above would be halved. If it had more than one sleeper for a portion of the year, without knowing the dates that the second sleeper ran, it would be impossible to calculate the occupancy rate. However, it would fall somewhere between the average rates, depending on the assumptions, for two cars and one car.

Also, if people stuffed more than two people in any room, i.e. children, then the actual car capacity is greater than the recommended capacity published by Amtrak.  I prefer a conservative calculation.

Interestingly, on my most recent trip on the Texas Eagle from LAX, the car attendant occupied Roomette 1 in the 2230 car, which is a Superliner, as opposed to a room in the transition sleeper, which at one time was the practice.  Moreover, whereas the LAX car attendants used to get off the car in San Antonio, she rode through to Austin.  She told me that Amtrak prefers to have the LAX attendant stay on the train until it clears San Antonio.  Austin is the next location with good hotel accommodations.

  • Member since
    November 2009
  • 422 posts
Posted by Dragoman on Sunday, July 6, 2014 5:15 PM

Sam1

...

If the Cardinal carried two sleepers during FY13, the number of rooms would have been 17,742 and the averages shown above would be halved. If it had more than one sleeper for a portion of the year, without knowing the dates that the second sleeper ran, it would be impossible to calculate the occupancy rate. However, it would fall somewhere between the average rates, depending on the assumptions, for two cars and one car.

...

Without knowing how many people did not ride because of "sold-out" status, trying to predict potential occupancy rates would seem to be an futile exercise.  The best one could do is try runnibg an extra sleeper on dates that are sold out well in advance, and see what happens.

I suspect Amtrak would be unlikely to take that approach.

  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Sunday, July 6, 2014 6:00 PM

Dragoman

Sam1

...

If the Cardinal carried two sleepers during FY13, the number of rooms would have been 17,742 and the averages shown above would be halved. If it had more than one sleeper for a portion of the year, without knowing the dates that the second sleeper ran, it would be impossible to calculate the occupancy rate. However, it would fall somewhere between the average rates, depending on the assumptions, for two cars and one car.

...

Without knowing how many people did not ride because of "sold-out" status, trying to predict potential occupancy rates would seem to be an futile exercise.  The best one could do is try runnibg an extra sleeper on dates that are sold out well in advance, and see what happens.

I suspect Amtrak would be unlikely to take that approach.

You are correct.  I was not attempting to predict how many people did not ride.

If a person called to book a room, and the rooms were sold out, Amtrak could tally this number of attempted phone reservations that were turned down because there were no rooms available on the requested travel date. Moreover, it could, I think, determine how many people attempted to book-on line or through a travel agency, and were turned down.  I am assuming Amtrak's on-line reservation system can do this, but I am not sure. However,  if a person goes on-line to check on a room, and he or she does not book a room because the price is too high, Amtrak would be hard pressed to know it.   

I presented several scenarios of what might have been the occupancy rates, depending on assumptions regarding the number of people occupying a room. It would not be more than two people per most rooms, if Amtrak's occupancy guidelines were followed, or less than one person per room. 

  • Member since
    December 2001
  • 8,156 posts
Posted by henry6 on Sunday, July 6, 2014 6:30 PM
On the contrary, even if sold out there is an indication if not a whole number of how many might have sought the space either by number of inquiries via on line requests or phone requests or live request. That is if that information is gathered. I'm just saying the capability to track some of the requests is available.

RIDEWITHMEHENRY is the name for our almost monthly day of riding trains and transit in either the NYCity or Philadelphia areas including all commuter lines, Amtrak, subways, light rail and trolleys, bus and ferries when warranted. No fees, just let us know you want to join the ride and pay your fares. Ask to be on our email list or find us on FB as RIDEWITHMEHENRY (all caps) to get descriptions of each outing.

  • Member since
    September 2007
  • From: Charlotte, NC
  • 6,099 posts
Posted by Phoebe Vet on Monday, July 7, 2014 7:41 AM

It is my understanding that Amtrak charges by the room, not the bed since other occupants just pay regular fare.  Doesn't that make the bed occupancy rate pointless?  What percentage of the ROOMS are sold?

Dave

Lackawanna Route of the Phoebe Snow

  • Member since
    December 2007
  • From: Georgia USA SW of Atlanta
  • 11,827 posts
Posted by blue streak 1 on Monday, July 7, 2014 10:22 AM

Phoebe Vet

It is my understanding that Amtrak charges by the room, not the bed since other occupants just pay regular fare.  Doesn't that make the bed occupancy rate pointless?  What percentage of the ROOMS are sold?

 
Good point had not considered.  However how is the factor of additional occupants in a room resolved?  A coach fare that does not take a seat in coach minus the meals eaten ?  Amtrak has the analysis but will never release unless ordered by congress.  That just as no RPMs broken down per train.
 
  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Monday, July 7, 2014 12:51 PM

Using the same methodology that I used to book end the occupancy rates for the Cardinal's sleepers, the average occupancy rate in FY13 for the Capitol's sleepers would have been 74.5 per cent if the car attendants occupied a room in the car, as opposed to camping out in the transition sleeper, and each room was occupied on average by one person.  If the attendants slept in the transition sleeper, the occupancy rate would have been 71 per cent.  If two persons occupied each room on the Capitol's sleepers, the average occupancy rate would have been 37.3 per cent and 35.5 per cent.

The actual occupancy rate would have been somewhere in between.  Moreover, I used Amtrak's recommended occupancy rates.  The number of passengers can be increased by allowing three people in the ADA room or three people in a bedroom.  

Amtrak's sells rooms with spaces in them.  It publishes the number of passengers traveling sleeper class. And the number of rooms in a car can be counted.

In FY13 the Capitol Limited recorded 45,626 sleeping car passengers.  How far they traveled or how many were in a room is unknown. However, the number of rooms, assuming the consist remained the same throughout the year, can be derived within reason.  The Capitol's sleepers had 30,660 rooms in FY13, assuming that the attendant slept in the transition car, or 29,200 if the attendants slept in the cars.  

Assuming an average of one person per room, the load factor would have been approximately 149 per cent if all the rooms were available and 156.3 per cent if the attendants occupied a room in each car.  If an average of two people occupied each room, the load factors would have been 74.4 and 78.1 per cent.  If an average of 1.5 persons occupied each room, the load factors would have been 111.6 and 117.2 per cent.

Capitol Limited's somewhat lower occupancy rate may account for some of the price difference between its sleepers and the Cardinal's sleeper.  But it probably is not the major driver of the difference.  If the Cardinal serves one more meal than the Capitol, it would drive some of the difference. The biggest difference driver in the sleeper class fares between the two trains probably is due to the fact that the fixed costs associated with the Capitols can be spread over a much larger number of units, i.e. passengers, number of rooms, etc. 

  • Member since
    July 2006
  • 9,610 posts
Posted by schlimm on Monday, July 7, 2014 1:26 PM

Looking at the Amtrak site for August 12, the fare for 2 people using a Superliner roomette on the Capital Ltd is $464.00, with 4 rooms left; $370.00 for just one person, 17 hr, 30 min. [Today, it arrived 1hr, 10 min. late].  On the Cardinal, it is only $391.00 with only 2 Viewliner rooms left, for a train ride almost 6 hours longer; $297.00 for only one passenger, 23 hr, 21 min.   

The pricing makes little sense to me, especially when a person could fly on SWA, Midway to Reagan for example, for a fare as low as $99, 1hr, 45 min. nonstop on August 12.  But currently it requires a change of plane.

C&NW, CA&E, MILW, CGW and IC fan

  • Member since
    June 2009
  • From: Dallas, TX
  • 6,830 posts
Posted by CMStPnP on Monday, July 7, 2014 2:48 PM

schlimm

Looking at the Amtrak site for August 12, the fare for 2 people using a Superliner roomette on the Capital Ltd is $464.00, with 4 rooms left; $370.00 for just one person, 17 hr, 30 min. [Today, it arrived 1hr, 10 min. late].  On the Cardinal, it is only $391.00 with only 2 Viewliner rooms left, for a train ride almost 6 hours longer; $297.00 for only one passenger, 23 hr, 21 min.   

The pricing makes little sense to me, especially when a person could fly on SWA, Midway to Reagan for example, for a fare as low as $99, 1hr, 45 min. nonstop on August 12.  But currently it requires a change of plane.

I agree with you on the price difference not making sense (but I am not sure I agree on it being more expensive than air.....it should be, your getting more than an airline would offer).    Back when I was looking for Sleeper accomodations last fall the price difference was $200 or more each way.     I also could not rationalize why so much difference in price........thats $400+ RT.      I don't think Amtrak cares how long the train runs with pricing unless it runs an extra overnight.

I suspect this is their yield management gone haywire.     If it was me I would ask what the floor price was for sleeping car compartments between city points and I am going to bet that Amtrak does not have a floor.........they just mindlessly apply their yield management system.      Scary but I think that is the market pricing intelligence we are dealing with here.

  • Member since
    December 2007
  • From: Georgia USA SW of Atlanta
  • 11,827 posts
Posted by blue streak 1 on Friday, July 11, 2014 2:40 AM

Sam1

Assuming an average of one person per room, the load factor would have been approximately 149 per cent if all the rooms were available and 156.3 per cent if the attendants occupied a room in each car.  If an average of two people occupied each room, the load factors would have been 74.4 and 78.1 per cent.  If an average of 1.5 persons occupied each room, the load factors would have been 111.6 and 117.2 per cent.

 
This poster completely forgot that certain sleeper space is taken up for crew rooms.  That certainly is  the Cardinal but the Capitol supposedly has a transition car most days ?  That makes your calculations much more difficult.
  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Friday, July 11, 2014 9:24 AM

Based on the information that is available to the public, all one can do with respect to determining sleeper occupancy levels is bracket the possible high and low occupancy rates.  Factoring in or out the room(s) occupied by the car attendant does not make the calculation any more difficult; it just means that one has to remember to account for it.

If one wants to know how the pricing for a room on the Cardinal compares with the Capitol Limited or any other train, one would ideally have the data points for every room for every day for 365 days.Just taking a snap shot of a day or two or even a week would not reveal a pattern for the year.

A statistical sample could be used to determine the distributed load factors. However, the results of a statistical sample are subject to variation, depending on the tightness of the sampling construct, and must be projected to the population as a range rather than a single number.

There is no evidence that Amtrak's Yield Management Program is not robust. I believe that it bought the algorithms from an outside vendor.  Yield Management algorithms are complex but not beyond a person with a reasonable background in mathematics to understand.  Information regarding generic yield management programs is available on the web. Information regarding Amtrak's Yield Management Program is not available, primarily because it is proprietary.

Whenever I have looked at booking a room from central Texas to the east coast, the price of the rooms on the Cardinal have almost always been higher than those for the Capitol Limited.  I suspect car capacity has something to do with it.  However, the biggest factor probably is the Cardinal's operating profile. It only runs three days a week in each direction and, therefore, wears more capital charges per room than the Capitol Limited, which drives the pricing, or at least it should.

  • Member since
    July 2006
  • 9,610 posts
Posted by schlimm on Friday, July 11, 2014 6:05 PM

Sam1
Whenever I have looked at booking a room from central Texas to the east coast, the price of the rooms on the Cardinal have almost always been higher than those for the Capitol Limited.

Given the difference I saw in the opposite direction with similar load factors, I think we could conclude with more random samples that the Capital charges more.  You really do not need to have a year's figures to see that.   Perhaps you saw the opposite because of your service from TX?

C&NW, CA&E, MILW, CGW and IC fan

  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Monday, July 14, 2014 11:13 PM

schlimm

Sam1
Whenever I have looked at booking a room from central Texas to the east coast, the price of the rooms on the Cardinal have almost always been higher than those for the Capitol Limited.

Given the difference I saw in the opposite direction with similar load factors, I think we could conclude with more random samples that the Capital charges more.  You really do not need to have a year's figures to see that.   Perhaps you saw the opposite because of your service from TX?

Unless the sample is drawn from at least one year of data points, with even more being desirable, it would not account for seasonal variations. 

Had you only analyzed the December through March financial statements for the company that I worked for, you would have been struck by the red ink (losses). But over a year we made money. What was lost during the winter was more than made up in the late spring, summer, and early fall.   

  • Member since
    June 2009
  • From: Dallas, TX
  • 6,830 posts
Posted by CMStPnP on Tuesday, July 15, 2014 7:31 AM

Sam1
Had you only analyzed the December through March financial statements for the company that I worked for, you would have been struck by the red ink (losses). But over a year we made money. What was lost during the winter was more than made up in the late spring, summer, and early fall.   

Most manufacturing companies handle that via a seasonal adjustment factor vs looking at the entire year.   The adjustment factor would smooth out the peaks and valleys if indeed there were any that regularly occured.     So for example since car sales are seasonal in this country most car companies use a seasonal adjustment when looking at car sales for a period within a year.  

In Texas natural gas is used primarily in winter for heating and electric is used primarily in summer for cooling.    It's possible electric utilities such as TXU, don't do that because they are in in business no matter what they do.....all they need to do is raise rates if they run into any financial trouble or missed forecast.     I would not compare utilities Accounting to more precise models out there and the last thing Amtrak needs is to emulate Utilities accounting.    I personally would prefer Amtrak be more vs less precise in it's Accounting method.    I worked as a contractor  for a large energy company in Texas (Enron) as a consultant and my most vivid memories of that experience:

1. Anderson Consulting advising their business end and Andersen Accounting auditing their books.   Inherent conflict of interest that led to Auditors overlooking criminal activity.

2. Massive charges in the footnotes of the financial statements on a regular basis.

3.  24-27 year old "Executives" imported and hired direct via Andersen Consulting with no real business operational experience  attempting to make multi-million dollar decisions.

What a E Ticket train ride that was.     Would hate to see Amtrak operate like that.

  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Tuesday, July 15, 2014 9:11 AM

CMStPnP

Sam1
Had you only analyzed the December through March financial statements for the company that I worked for, you would have been struck by the red ink (losses). But over a year we made money. What was lost during the winter was more than made up in the late spring, summer, and early fall.   

Most manufacturing companies handle that via a seasonal adjustment factor vs looking at the entire year.   The adjustment factor would smooth out the peaks and valleys if indeed there were any that regularly occured.     So for example since car sales are seasonal in this country most car companies use a seasonal adjustment when looking at car sales for a period within a year.  

In Texas natural gas is used primarily in winter for heating and electric is used primarily in summer for cooling.    It's possible electric utilities such as TXU, don't do that because they are in in business no matter what they do.....all they need to do is raise rates if they run into any financial trouble or missed forecast.     I would not compare utilities Accounting to more precise models out there and the last thing Amtrak needs is to emulate Utilities accounting.    I personally would prefer Amtrak be more vs less precise in it's Accounting method.    I worked as a contractor  for a large energy company in Texas (Enron) as a consultant and my most vivid memories of that experience:

1. Anderson Consulting advising their business end and Andersen Accounting auditing their books.   Inherent conflict of interest that led to Auditors overlooking criminal activity.

2. Massive charges in the footnotes of the financial statements on a regular basis.

3.  24-27 year old "Executives" imported and hired direct via Andersen Consulting with no real business operational experience  attempting to make multi-million dollar decisions.

What a E Ticket train ride that was.     Would hate to see Amtrak operate like that.

To adjust interim period results for seasonal factors, you have to know what they are. One way to get them is to use regression analysis to develop a trend line so as to be able to forecast the seasonal impact on the interim period results.  One way or the other to get a complete picture for any entity, you have to know the the seasonal impacts on the financials.  

Most businesses cycle through at least a year.  That is why, if they are a public company, they are required to issue annual financial statements.  As a matter of fact, most private entities, partnerships, etc. generate annual financial results.  

Your misunderstanding of the electric utility business in Texas is palpable. Beginning in 1988 Texas' investor owned electric utilities began the deregulation processes. The first step occurred in 1992 with the introduction of competition in the wholesale power market. By 2002 the entire generation and retail market had been deregulated.  

Today the electric utility business in North Texas, as an example, is highly competitive. In the Metroplex residents can choose from more than 200 electric utility plans offered by more than 35 retailers. Moreover, power generators, if they want to sell their power, must bid into the power pool every 15 minutes. Outside of stock, bond, and foreign exchange traders, I cannot think of anyone that operates in a more competitive market.

The demand curve for electricity in North Texas has two modes.  The largest, of course, is the summer load, which is driven largely by the demand for air conditioning.  The second mode, which occurs in December and January, reflects the increase use of electric heat in the winter.  

What do you know about utility accounting?  It conforms to GAAP.  In addition, the regulators impose greater requirements on utilities because of the rate making processes for the portion of the business that is still heavily regulated, i.e. the poles and wires segment of the business.

Amtrak is a stock company.  Its accounting practices conform to GAAP.  Like investor utilities, its financial statements are audited by outside auditors.  Its requirements are not significantly different than any other company that complies with GAAP.

How about giving us several specific examples of of how Amtrak's accounting practices differ from those of the electric utility business. What makes you believe that Amtrak's accounting practices are more precise or less precise than those in the utility business.

What is the basis for your views on accounting and finance?  Are you an accountant?  What certifications do you have?  What is your educational?  What type of consulting did you do?  You can tell us the type of consulting you did without disclosing who you worked for, although you have already given us the name of Enron.

  • Member since
    June 2009
  • From: Dallas, TX
  • 6,830 posts
Posted by CMStPnP on Tuesday, July 15, 2014 4:41 PM

Yeah, yeah, yeah, I didn't drink the Kool Aide.     There are very few Electrical Distribution and Transmission companies in Texas of which OnCor (formerly TXU) is still the largest and it is still regulated.     The mirage of deregulation of all these end user companies that charge different rates but buy their electricity from basically the same sources is a joke.    OnCor is basically in charge still and everyone calls OnCor if there is a power outage so really.........the power company is OnCor not some consumer facing shell company with a nice suburban office somewhere. .

Utility Accounting is recognized as pretty atrocious.....always has been.

  • Member since
    April 2003
  • 305,205 posts
Posted by Anonymous on Wednesday, July 16, 2014 12:54 PM

CMStPnP

Utility Accounting is recognized as pretty atrocious.....always has been.

Investor owned utility accounting complies with Generally Accepted Accounting Principles in all respects. Public owned utility accounting complies with Generally Accepted Government Accounting Principles. The same is generally true for co-ops.

The financial statements generated by most utilities - electric, gas, telecommunication, etc. - are audited by independent auditors. In the case of the large investor owned electric utilities, because of the complexity of their accounting systems, the external auditors are usually drawn from one of the large accounting firms.

In addition to complying with all of the requirements of GAAP or government accounting principles, electric utilities in the U.S. must comply with the requirements of the Federal Energy Regulatory Commission and, in most states, the Public Utility Commission. For those companies with a nuclear plant, they must comply with the stringent accounting and reporting requirements of the Nuclear Energy Regulatory Commission.

Enron, which appears to be the extent of your involvement with the energy industry in Texas, although in what capacity we don't know, was not a regulated electric utility. It bought and sold gas and electricity contracts in California, where it engaged in fraudulent practices designed to distort the wholesale power market, as well as some other states, but it did not operate an electric utility in Texas. It was not regulated by the PUC.  

Enron perpetrated an accounting fraud, which any corporation could do. Long story short it was discovered by the internal audit manager - not Anderson - that blew the whistle when the fraud was discovered.

How does the Enron scandal make all utility accounting systems atrocious? The flaw in your logic, which is certainly not the first time that it has occurred, is to generalize a single incident to the population. It is akin to saying that since A is corrupt, it necessarily follows that B through Z are also corrupt, even though there is no evidence to support the claim.  It is a logical fallacy.

What is the basis of your statement that utility accounting is atrocious?  By what standards?  

Again, are you an accountant?  What are your certifications?  What is your educational background? What experience have you had with electric utility accounting?  What was your role at Enron?  When were you an Enron contractor?  

Join our Community!

Our community is FREE to join. To participate you must either login or register for an account.

Search the Community

Newsletter Sign-Up

By signing up you may also receive occasional reader surveys and special offers from Trains magazine.Please view our privacy policy