December 2015 Performance Report

[lo2e]

The December 2015 Performance Report is now available - https://www.amtrak.com/ccurl/747/108/Amtrak-Monthly-Performance-Report-December-2015.pdf

First of all, someone needs to work on their cut-and-paste skills, as page A-3.4 has FY15 and FY14 in the ridership and revenue charts, likewise on the bottom chart of page A-3.5.

Not a great month for ridership across the board, with very few notable exceptions. Surfliner and Capitol Corridor did amazingly well, Downeaster is bouncing back nicely from a disastrous FY15. On the long distance side, EB, LSL, Crescent, CZ, SWC, and the Palmetto (thanks to allowing NER customers) all did well. They're blaming a small decline on the NEC on gas prices and moving some of the NER passenger traffic to the Palmetto, so I don't think the sky is falling there.

There's just no better way to say it, the midwest is uglier than a bowling shoe on the corridor routes. Nearly all of the Chicago-based corridors were pretty bad, HF and MRR were awful, and Michigan services weren't all that great (especially Pere Marquette). This is getting to be more than anomaly, it is becoming a trend.

I don't know what's going on with the Auto Train, but the Amtrak juggernaut appears to be losing its superpowers lately. Gas prices?

Since there has been a lot of discussion around the Silver Star lately, I've found something rather interesting, tell me if I'm interpreting this wrong:

• Total Ridership on the Star was unfavorable year-over-year but favorable to budget, so they were expecting a drop and luckily it wasn't as big a drop as they were expecting. Year-to-date ridership is unfavorable year-over-year and unfavorable to budget. If we have more months like December, the ridership might bounce back enough to become favorable to budget for the year, we'll wait and see.
  
• In December, total ridership on the Star was 36,541, while it indicates on page A-3.5 that sleeper ridership was 3,207. So the difference, 33,334, must have been coach passengers, correct? Sleeper ridership on the Star was up 2+% in December and year-to-date is up a whopping 8%. Pretty impressive, all things considered.
  
• But here's where things start to get funky... If what I put is correct, because of this, it appears the Star is losing coach passengers in a huge way. December FY2016 had 33,334 coach passengers compared with 36,083 in December FY15, a drop of 2,749 (7.6%), or about 88 coach passengers per day on average. YTD looks even worse, with FY16 82,717 vs. FY15 97,648 - a drop of 14931 (15.3%), or a 3-month average of 4,977, or 162 per day (92 days in Oct-Dec). Ouch.

Are these numbers accurate? I'll admit I haven't done this analysis on any of the other LD routes (yet), so I don't know if this is a pattern system-wide or something special with the Star. Fewer short-distance passengers and more endpoint to endpoint ones? Gas prices? Competition? Moving to sleepers? I'm really not sure what to make of this yet, but concerning would be an understatement if my analysis is correct.

 

[OBS]

The Star is currently operating with 3 coaches instead of 4, Supposedly to reduce strain on single food service car...

 

[lo2e]

The Star is currently operating with 3 coaches instead of 4, Supposedly to reduce strain on single food service car...

Thanks, I didn't catch that - any idea how long that's been the case?

 

[jis]

Well since the food service is not changing anytime soon, I suppose the 3 Coach is probably pretty permanent, unless they figure that the food service car can handle a fourth Coach regularly.

 

[chrsjrcj]

The Star is currently operating with 3 coaches instead of 4, Supposedly to reduce strain on single food service car...

It's absurd that this is Amtrak's strategy. Only in Amtrak World is too many riders a bad thing.

 

[AmtrakLKL]

The Star is currently operating with 3 coaches instead of 4, Supposedly to reduce strain on single food service car...

Every eastern long distance train is currently operating with fewer coaches due to the normal low season in January and February. The Star, Meteor, Crescent and Lake Shore are all running three coaches. The Cardinal has running two coaches. All the cars that are off trains are now going through PM cycles that were re-arranged to permit all the extra holiday capacity.

BTW, the Star ran with five full coaches and two full sleepers during the Christmas/New Years peak. Lines were long in the cafe but not as bad as I expected, so I don't think capacity is being reduced to alleviate overcrowding in the lounge.

 

[spidersfan351]

I watch the Star go by most days here from my office, you guys are right, it's been three coaches since the holidays. Just wanted to add that they do run a third VL sleeper during peak seasons "occasionally". I noticed this during November but not December. I did spot the Meteor with Four VL sleepers, plus a baggage (five total) one of the Christmas holidays.

I would add that while small consists can seem frustrating, it is good that Amtrak is doing this as a way to improve peak service.

 

[AmtrakLKL]

I watch the Star go by most days here from my office, you guys are right, it's been three coaches since the holidays. Just wanted to add that they do run a third VL sleeper during peak seasons "occasionally". I noticed this during November but not December. I did spot the Meteor with Four VL sleepers, plus a baggage (five total) one of the Christmas holidays.

I would add that while small consists can seem frustrating, it is good that Amtrak is doing this as a way to improve peak service.

The Star did not get the extra sleeper this past holiday season. For 2015's Thanksgiving and Christmas peaks, the extra sleeper was put on the Meteor as a crew car to free up revenue space in the other three sleepers. Makes sense - put the capacity where you can sell the rooms at a higher fare.

 

[afigg]

Some comments on and stats from the December 2015 monthly report.

System metrics observations: somewhere between mixed and not good overall. Ticket revenue for December ($182.4 million) was down -$7.7M from 2014 and below the budget. The flat ticket and food & beverage revenue continues to be offset by expenses running less than the budget (cheap oil helps and hurts), but the adjusted system operating loss for December was $12.8M and Year-To-Date was $19.6M. For the first 3 months of FY2015, Amtrak actually had an adjusted surplus of $5.3M.

One of the metrics is total passenger miles which was 587,602,000 for Dec 2015 compared to 604,643,000 in December 2014. Meanwhile seat and train miles were up, so there are more seats available. December was a down month, as the YTD passenger miles is up at little at 1,688,314,000 in 2015 over same period in 2014.

Ridership and Ticket Revenue
As noted, the Midwest corridor services are off to a rather poor start for the FY along with the VA Regionals and 2 Vermont trains. Cheap gasoline prices are hurting and probably discount buses are a factor as well. However, for the Midwest corridors, I suspect the poor On-Time Performance of 2014 and the summer of 2015 also had a corrosive effect. Takes time for ridership to rebound from long service interruptions and repeated late trains. Unfortunately, there is probably another Spring to Fall period of track work delays coming for the CHI-STL and CHI-DET corridors.

For the NEC, I continue to think that Amtrak has pushed the prices up a bit too much, especially for a softening economy. Meanwhile, for the busier corridors, the Keystone, Surfliner and Capitol Corridor continue to show ridership growth and the Downeaster is rebounding from the prolonged period of track work delays.

Ridership and Revenue summary for the month of December compared to 2014:
System: ridership -2.8%, revenue: -3.9%
Acela: ridership -5.7%, revenue: -3.6%
NE Regional: ridership -2.3%, revenue -4.3%
State supported corridors: ridership -3.8%, revenue -3.7%
LD trains: ridership +2.9%, revenue -3.7%

Ridership and Revenue summary for the YTD from October to December:

System: ridership -2.0%, revenue: -2.2%
Acela: ridership -1.6%, revenue: -0.9%
NE Regional: ridership -2.5%, revenue -1.76%
State supported corridors: ridership -2.5%, revenue -1.3%
LD trains: ridership +0.6%, revenue -4.9%

On-Time Performance (OTP)

The good news in the continuing improvement in OTP for the month and YTD over the same periods in 2014. The decline in freight traffic is helping the LD trains and likely many of the corridor services as well. The NEC OTP was improved with the Acela jumping to 88.2% OTP in December.

The Michigan trains had a big jump in OTP at 82.3% in December compared to 41.6% in Dec 2014. The Capitol Limited OTP was 90.3% in December which is a huge improvement.

On-Time Endpoint Performance for the month of December compared to 2014:

System: 83.2%, 2014: 76.3%
NE Corridor: 87.2%, 2014 81.6%
State supported corridors: 84.4%, 2014 76.0%
LD trains: 68.0%, 2014 65.9%

 

[neroden]

But here's where things start to get funky... If what I put is correct, because of this, it appears the Star is losing coach passengers in a huge way.

Well, that's what I *thought* would happen when the dining car was removed. I said so. You can go back and look. ;-)

 

[lo2e]

But here's where things start to get funky... If what I put is correct, because of this, it appears the Star is losing coach passengers in a huge way.

Well, that's what I *thought* would happen when the dining car was removed. I said so. You can go back and look. ;-)

I believe it! I thought the hurt would be much worse to the sleeper end of things, I wasn't thinking coach would take a big hit.

 

[cirdan]

But here's where things start to get funky... If what I put is correct, because of this, it appears the Star is losing coach passengers in a huge way.

Well, that's what I *thought* would happen when the dining car was removed. I said so. You can go back and look. ;-)

I believe it! I thought the hurt would be much worse to the sleeper end of things, I wasn't thinking coach would take a big hit.

Maybe the reduction in sleeper price engendered by the removal of the diner means that sleeper fares are now more accessible to passengers who would otherwise be in coach?

Sometimes unitentional consequences can be fascinating.

 

[lo2e]

Indeed, although the drop in coach is still quite a lot, even if you assume that the entire growth in sleeper ridership were those who were previously in coach.

YTD Coach Ridership down 14,931 from FY15

YTD Sleeper Ridership "only" up 661 from FY15

 

[jis]

But here's where things start to get funky... If what I put is correct, because of this, it appears the Star is losing coach passengers in a huge way.

Well, that's what I *thought* would happen when the dining car was removed. I said so. You can go back and look. ;-)

I think removing one Coach may have more to do with it directly that Dining Car existence or lack thereof. They reduced the length of the train by one Coach. Seats that are not there cannot possibly carry riders or earn revenue.

 

[chrsjrcj]

I don't buy the slow season argument. Amtrak sold those seats last year, but can't this year? The only thing I can think of would be lower gas prices.

 

[lo2e]

But here's where things start to get funky... If what I put is correct, because of this, it appears the Star is losing coach passengers in a huge way.

Well, that's what I *thought* would happen when the dining car was removed. I said so. You can go back and look. ;-)

I think removing one Coach may have more to do with it directly that Dining Car existence or lack thereof. They reduced the length of the train by one Coach. Seats that are not there cannot possibly carry riders or earn revenue.

When did they remove that coach? I thought January was the first time with one less coach, but there's probably someone that knows much better than I do.

 

[cirdan]

But here's where things start to get funky... If what I put is correct, because of this, it appears the Star is losing coach passengers in a huge way.

Well, that's what I *thought* would happen when the dining car was removed. I said so. You can go back and look. ;-)

I think removing one Coach may have more to do with it directly that Dining Car existence or lack thereof. They reduced the length of the train by one Coach. Seats that are not there cannot possibly carry riders or earn revenue.

It would be interesting if somebody who understands the figures can do a quick calculation to see if coach passengers are indeed constrained by reduced capacity.

If so, it would indeed have been a short-sighted move on Amtrak's part.

 

[jis]

But here's where things start to get funky... If what I put is correct, because of this, it appears the Star is losing coach passengers in a huge way.

Well, that's what I *thought* would happen when the dining car was removed. I said so. You can go back and look. ;-)

I think removing one Coach may have more to do with it directly that Dining Car existence or lack thereof. They reduced the length of the train by one Coach. Seats that are not there cannot possibly carry riders or earn revenue.

When did they remove that coach? I thought January was the first time with one less coach, but there's probably someone that knows much better than I do.

It happened back in September or so as I recall.

 

[lo2e]

It happened back in September or so as I recall.

Ok, I'll go with that. So here's how coach ridership looked for Sep-Nov:

Sept 2015 Total Ridership 26,525, Sleeper Ridership 2,630, Coach Ridership 23895 - Sept 2014 Total 28,038, Sleeper 2,062, Coach 25,976 - Coach decrease 2081 (8%), Average 69 fewer coach passengers per day

Oct 2015 Total 23,871, Sleeper 2,432, Coach 21,439 - Oct 2014 Total 31,067, Sleeper 2,204, Coach 28,863 - Coach decrease 7,424 (25.7%), Average 239 fewer coach passengers per day

Nov 2015 Total 31,136, Sleeper 3,192, Coach 27,944 - Nov 2014 Total 35,542, Sleeper 2,840, Coach 32,702 - Coach decrease 4,758 (14.5%), Average 158 fewer coach passengers per day

December is above.

I think the results speak for themselves. September seems to be in line with using one less coach, October and November go pretty far beyond that. Yikes...

Edit to add: Although to be fair, the southeast had major flooding in October, so that could partially account for those numbers, but sleeper ridership actually went up in October, so I'm not convinced the flooding was the only culprit.

 

[neroden]

I think the data is too weak to prove that I was right, but it does *suggest* that I *might* have been right.

I've actually encountered people who took Amtrak, in coach, only after they found out that the train had a dining car -- *and didn't even use the dining car*. Its mere existence was somehow reassuring to them.

 

[jis]

Leaving aside just the Star for the moment, do we have an estimate of what the net passenger loss due to lower gas prices have been? Is there any way to eke out that estimate?

Specific to the Star, since it tended to be heavy on short turn Coach passengers I would expect it to get hit harder with a gas price related downturn in loading than other trains that generally tend to be longer turn Coach passengers. Anecdotally there has been significantly lower short turn loading at least in Florida due to the precipitous fall in gas prices.

I agree that it is almost impossible to separate out eh effect of lack of Diner from the effect of drop in gas price.

 

[neroden]

The gas price effect is extremely hard to sort out. There seems to be pretty strong evidence that low gas prices have a major effect on ridership in the Midwest, and equally strong evidence that they have little or no effect in the Northeast and California markets.

I suppose this makes sense: first of all, gas prices are still higher in the Northeast and California than they are in the Midwest, and second, congestion was the main reason for ridership in the Northeast and California anyway, and congestion is just as bad as ever.

I have absolutely no idea about the effects of gas prices on the market in the Southeast, however.

 

[WoodyinNYC]

I've actually encountered people who took Amtrak, in coach, only after they found out that the train had a dining car -- *and didn't even use the dining car*. Its mere existence was somehow reassuring to them.

Couple of things here.

One. Many potential riders have only a vague idea of the accommodations on board. Many may be thinking of their rare experiences with commuter train sardine packed rides, or at best the corridor trains. But they have a hazy idea that the trains with dining cars are a step up, and should have the wider seats, greater leg space, and invitation to get up and walk a bit that they've seen in ads or movies etc.

Two. When the rest of the Viewliner IIs arrive on some sweet summer's day (2017 if you ask me), then the Eastern LD trains will have 3 or 4 NEW cars in each consist -- one bag car, one diner, and one V II sleeper. Those cars will replace the oldest, most bedraggled looking cars in the active fleet. The LD trains then will look pretty good -- or at least half good -- and greatly help the public's mental image of riding Amtrak. (I expect that the new bi-levels in the Midwest will greatly help that mental image of Amtrak overall as well.)

 

[Eric S]

A CityLab article looked at the effect gas prices (and on-time performance) had on Amtrak ridership. I don't know that it gives you anywhere enough information, though, to help with this analysis here.

 

[west point]

Until Amtrak is required to publish each train's & route's revenue passenger miles instead of just overall RPMs one cannot tell how each route is doing. For example the Downeasters appear to carry mostly passengers from Maine to Boston north so its RPMs per train mile might be higher than another train's RPMs. An extreme example is the Crescent's RPM ATL <> New Orleans which by all measures is rather low.

IMO the number of passengers on a train or service has no meaning.