Amtrak Diner and Food Service Discussion

Arlington wrote:The Texas Eagle did$8.6m in Sleeper Class in the year ending Sept 2015 (Table A-3.4, page 21 of PDF) . Back in 2012, its food costs were $7.8m, which at even just 3% per year of escalation puts them at $8.4m. THis on a train where costs were $56m and revenues were $27m for 2015.

If anything, the story on the Eagle has to be like on the Silver Star: better to eliminate the diner, lower fares, and boost sleeper-class ridership. Next step for the Texas Eagle has got to be figuring out how the cafe can serve airline food, not reworking the diner menu.

The Eagle is the only single overnight long distance train that serves two dinners (towards San Antonio), and three dinners on a single round trip. It's food costs has to be higher than all the other overnight long distance train, because of the extra meal being served per direction, or extra two meals being served on a roundtrip. That's the reason why it was amongst the first long distance train to downgrade from a full diner to a diner-lounge car (cross country club car). When comparing it to the other overnight long distance trains, try to remember that major difference.

electricron wrote:

Arlington wrote:The Texas Eagle did$8.6m in Sleeper Class in the year ending Sept 2015 (Table A-3.4, page 21 of PDF) . Back in 2012, its food costs were $7.8m, which at even just 3% per year of escalation puts them at $8.4m. THis on a train where costs were $56m and revenues were $27m for 2015.

If anything, the story on the Eagle has to be like on the Silver Star: better to eliminate the diner, lower fares, and boost sleeper-class ridership. Next step for the Texas Eagle has got to be figuring out how the cafe can serve airline food, not reworking the diner menu.

The Eagle is the only single overnight long distance train that serves two dinners (towards San Antonio), and three dinners on a single round trip. It's food costs has to be higher than all the other overnight long distance train, because of the extra meal being served per direction, or extra two meals being served on a roundtrip. That's the reason why it was amongst the first long distance train to downgrade from a full diner to a diner-lounge car (cross country club car). When comparing it to the other overnight long distance trains, try to remember that major difference.

If we "must" lay out two dinners before people, that makes airline food all the more urgent, since the Sleeper customers appear unwilling to pay a fare that actually covers all that food (which also suggests that many travelers chose other modes because Amtrak's fares are too high, no matter how many dinners dangle in front of them). Chicago-Texas is an extremely price-competitive air travel market (with American, Southwest hubbed directly on-route and United also flying Texas-ORD)

I agree. It makes no sense at all to lose passengers simply because of high food service costs which are being passed through in fares.

gokeefe wrote:I agree. It makes no sense at all to lose passengers simply because of high food service costs which are being passed through in fares.

But the opposite is happening. Look at the last pdf link again, on pages 20 and 21, charts 3.3 and 3.4.
Reducing food service losses might make the train more profitable, or lose less money, but the lack of the diner is turning coach passengers away in larger numbers than it is attracting more sleeper passengers.

Silver Star ridership FY15 = 383,347; FY14 = 405,695; Loss over year = 22,348; (-5.5%)
Silver Meteor ridership FY15 =346,097; FY14 = 348,581; Loss over year = 2,484 ;(-0.7%)
On average, long distance trains ridership lost less than 1%. The Star lost 5%, the Meteor met the average.

Silver Star revenues FY15 = 33,108,142; FY14 = 34,557,675; Loss over year = 1,448,741; (-4.2%)
Silver Meteor revenues FY15 = 38,455,934;, FY14 = 38,499,563; Loss over year = 43,629; (-0.1%)
On average, long distance trains revenues loss 5%. The Star just beat the average, the Meteor almost broke even.

One would think dropping food expenses from sleeper fares on the Silver Star would cause an increase in sleeper ridership, which it did!
Silver Star sleeper ridership FY15 = 32,703 ; FY14 = 30,852; Gain over year = 1851; (+6.0%)
Silver Meteor sleeper ridership FY15 = 43,434 ; FY14 = 42,922 ; Gain over year = 512; (+1.2%)

Was the gain of 1,851 sleepers worth the loss of 22,348 other riders?

Silver Star sleeper revenues FY15 = 8,089,017; FY14 = 7,783,079; Gain = 305,938;(+3.9%)
Silver Meteor sleeper revenues FY15 = 12,057,773; FY14 = 11,807,364; Gain = 250,409;(+2.1%)
On average, long distance sleeper ticket revenues lost 5%. Both the Star and Meteor revenues were up.

We can also calculate the average sleeper revenue per rider using the data above:
For FY 2015
Silver Star = 8,089,017 / 32,703 = 247.35
Silver Meteor = 12,057,773 / 43,434 = 277.61

For FY 2014
Silver Star = 7,783,079 / 30,852 = 252.27
Silver Meteor = 11,807,364 / 42,922 = 275.09

So, in summary, the average Silver Star sleeper fares dropped $4.92 form FY 14 to FY15,
the average Silver Meteor sleeper fares dropped $2.52.

Was a savings of $4.92 on the average sleeper fares worth loosing the dining car?

Sometimes we miss seeing the entire forest while concentrating on looking at just one tree.
Is food service losses that one tree?

Maybe we should wait to see the data over an entire FY after the dining car change was made to the Silver Star to look at the data so closely, but I'm not seeing any significant savings to passengers with the data published so far.

electricron wrote:Maybe we should wait to see the data over an entire FY after the dining car change was made to the Silver Star to look at the data so closely, but I'm not seeing any significant savings to passengers with the data published so far.

Sorry to say: You did all that work on the wrong time periods. The Star only lost its Diner on July 1 (so 9/12 of FY 2015 were "with diner", and only 3/12 (Jul, Aug, Sep) were without, and any changes in coach ridership are likely dominated by other factors (falling cost of driving, OTP, weather, who knows?). I used that PDF because it is the last full-year of data against which we could compare the OIG's 2012 detailed breakout of diner costs, not because it was good at diner before-and-after on the Star.

IN the Silver Star thread we might be able to piece together better numbers.

But looking at Feb MPR, you see that Star Sleeper service is up by 5.6% this year (but that's only 700 passengers), while the whole Star is down 19,000 passengers (or 19,700 coach passengers).

Aint no way that 19,700 coach passengers were scared off by the lack of a diner. This is much much more likely to be due to some macro factor, and what you really should be saying is that it is AMAZING that sleeper traffic is growing by 5.6% on a train where ridership is otherwise down 11% (the Auto Train is down 11% too)

Actually, I'd rather see explained what's happened on the Palmetto, whose traffic YTD is up 65% (and 110%+ in Feb). I suspect that's re-accommodation of the night LD passengers on the day train during track outages/repair. But the big coach swings are 99.9% unrelated to the Star losing its diner.

Arlington wrote: Actually, I'd rather see explained what's happened on the Palmetto, whose traffic YTD is up 65% (and 110%+ in Feb). I suspect that's re-accommodation of the night LD passengers on the day train during track outages/repair. But the big coach swings are 99.9% unrelated to the Star losing its diner.

Didn't the Palmetto begin taking on intra-NEC traffic?

Arlington wrote: Sorry to say: You did all that work on the wrong time periods. The Star only lost its Diner on July 1 (so 9/12 of FY 2015 were "with diner", and only 3/12 (Jul, Aug, Sep) were without, and any changes in coach ridership are likely dominated by other factors (falling cost of driving, OTP, weather, who knows?). I used that PDF because it is the last full-year of data against which we could compare the OIG's 2012 detailed breakout of diner costs, not because it was good at diner before-and-after on the Star.

IN the Silver Star thread we might be able to piece together better numbers.

But looking at Feb MPR, you see that Star Sleeper service is up by 5.6% this year (but that's only 700 passengers), while the whole Star is down 19,000 passengers (or 19,700 coach passengers).
This is much much more likely to be due to some macro factor, and what you really should be saying is that it is AMAZING that sleeper traffic is growing by 5.6% on a train where ridership is otherwise down 11% (the Auto Train is down 11% too)

Actually, I'd rather see explained what's happened on the Palmetto, whose traffic YTD is up 65% (and 110%+ in Feb). I suspect that's re-accommodation of the night LD passengers on the day train during track outages/repair. But the big coach swings are 99.9% unrelated to the Star losing its diner.

I have no idea why the Star lost so many coach passengers compared to all the other trains running parallel to it. But the 5% increase in sleeper passengers on the Star did not come as a result of a loss of sleeper passengers on the Meteor. So, I don't agree the increase has anything to do with the loss of the diner either.

While 5% is a significant increase data wise, it really isn''t in the real world. A single Viewliner sleeper has 3 rooms and 12 roomettes, all with double occupancy. A completely full Viewliner car has a total capacity of 30. 5% of 30 is 4.5. That's the maximum number of more passengers 5% can be in a Viewliner. That is equivalent to selling 2.5 rooms or roomettes more. Since the cars aren't completely full, let's suggest 2 more cabins of the 15 are occupied. A non-wowing event to be celebrating.

electricron wrote: I have no idea why the Star lost so many coach passengers compared to all the other trains running parallel to it. But the 5% increase in sleeper passengers on the Star did not come as a result of a loss of sleeper passengers on the Meteor. So, I don't agree the increase has anything to do with the loss of the diner either.

That "reasoning" (that the only way losing the diner affected sleeper ridership was by siphoning sleeper passengers from the Meteor) makes no sense. The sleeper passengers gained are probably coach passengers who decided to upgrade, now that it costs less to upgrade (those passengers for whom C+S is worth paying for but not C+S+D). Of course, a coach passenger upgrading to sleeper doesn't affect overall ridership (beyond possibly having another rider take the coach seat), so if your focus is on ridership, it's not a win. Then again, if the focus is on ridership, sleepers shouldn't exist...

leviramsey wrote:

Arlington wrote: Actually, I'd rather see explained what's happened on the Palmetto, whose traffic YTD is up 65% (and 110%+ in Feb). I suspect that's re-accommodation of the night LD passengers on the day train during track outages/repair. But the big coach swings are 99.9% unrelated to the Star losing its diner.

Didn't the Palmetto begin taking on intra-NEC traffic?

yes (much to my chagrin having to wait hours for a northbound Palmetto running late because there is no longer an NEC only train).

So while ridership and revenue may be a hopeless muddle* the one, clear thing, 100% caused by diner removal is that diner elimination is on track to save ~$7m/yr...benefits far exceeding the value of any downside you can reasonably tie to it.

*The loss of coach riders is so large that it almost entirely consists of people who'd never have any interaction with the diner for good or bad on any trip, and therefore must have a cause other than diner removal, and that Sleeper class swam against this tide must mean that lower sleeper class prices (and not diner/no-diner) enabled this.

It would be interesting to know how Ringling Bros. Barnum & Bailey Circus runs their Pie Car. They must feed everyone aboard when en route. How much choice do they offer? What is their cost per meal? What is the level of preparation done on board?

The Crescent arrives New Orleans after 7:30 PM. It seems there should be two dinners served on it, no?

Noooooooooo wayyyyyyyyyy !!!!!!!!!!
The train schedule I'm reading states the CONO departs Chicago at 8:05 pm and arrives in New Orleans at 3:32 pm 19.5 hours later. Does it even serve dinner heading south? I would eat just breakfast and lunch.

In the opposite direction the train schedule I'm reading states the CONO departs New Orleans at 1:45 pm and arrives in Chicago at 9:00 am. I would eat just dinner and breakfast.

Therefore on the CONO, Amtrak probably serves just two meals per direction, while on the TE Amtrak serves four meals per direction. It takes the TE over 32 hours to travel between Chicago and San Antonio, 12.5 hours more than it takes to travel between Chicago and New Orleans.

My question was about the Crescent, but thanks anyway!

electricron wrote:Therefore on the CONO, Amtrak probably serves just two meals per direction, while on the TE Amtrak serves four meals per direction. It takes the TE over 32 hours to travel between Chicago and San Antonio, 12.5 hours more than it takes to travel between Chicago and New Orleans.

Talk about an opportunity for cost savings. Speed up the train and reduce labor and meal costs while still receiving the same or better revenues?!? Wow. I never thought of the losses on the Long Distance services as being proportional to the average speed of the train. I wonder what the coefficient/correlation looks like.