How much does it cost to run a train?

[paul1609]

This thread has spurred me on to attempt to do some Calculations to work out the Energy Cost per Passenger Mile on the Railways for different types of trains.

According to data I found from the ORR, Class 390s, on average, consume about 29KWh/Mile. Let's assume that the trains are at capacity, which for a 9-Car 390 I think is 469 seats. So that works out to about 0.0618 KWh/Passenger Mile. Apparently, and correct me if I'm wrong, but TOCs pay £0.21/KWh for traction electricity. Giving us a rough total of £0.013/Passenger Mile.

Using similar data for the Class 379s, we find they average at about 17KWh/Mile and have a capacity of 209 seats. Again working this out with cost of electricity it comes to a total of about £0.017/Passenger Mile.

Finally, I thought I would investigate the costs for DMUs. According to a post on this site a single car of a Class 156 will use about 6mpg. This converts to 0.76 Litres of Diesel/Mile. Assuming a single car has 75 seats this equates to about 0.01 Litres/Passenger Mile. It seems quite hard to work out how much the TOCs pay for Diesel, but I think an upper bound would be around 80 pence/Litre. All together then £0.008/Passenger Mile.

This came as a bit of a shock to me as I assumed Electric traction would be much cheaper, and is a bit concerning. Delving a bit deeper it seems to me that this is due to old DMUs having less performance than EMUs so are using less energy and the railways being able to get Diesel much cheaper than Electricity (which is also true for the public).

If it helps the latest delivery of diesel for my heritage railway cost 70.8p per litre including 5% VAT, Id imagine because of quantities concerned a TOC would be paying much less than that. Id imagine that 21 p/kwh is also very much on the high side, Im averaging less than that (obviously not for traction!).

 

[stuu]

Id imagine that 21 p/kwh is also very much on the high side, Im averaging less than that (obviously not for traction!).

Network Rail charged just over 22p per kWh in 2023/24

 

[Ducatist4]

This is interesting in relation to fuel costs from 2022-23:

 

[Pegpilot]

In answer to your question, Jan, ORR reported 53 billion pax km in 22/23, yielding a cost/pax km of 47p or so. But 22/23 was partially Covid impacted and we were up at about 67 Bn pax km in 2019/2020 before covid hit. Adjusting for that would give a figure more like 40p/pax km.

 

[Peter Mugridge]

If it's of any interest to the thread...

15 years or so ago, a request from a company based in the London area went in to Virgin Trains for a charter using the Pretendolino rake of Mk 3 stock Euston - Carlisle and back, out via Shap and back via the Settle and Carlisle; no special traction requested.

Now... this will include a profit margin so will be a bit above the actual cost, but the quote came back at £30,000...

Needless to say, a different staff entertainment option was immediately chosen...!!

 

[6Gman]

If it's of any interest to the thread...

15 years or so ago, a request from a company based in the London area went in to Virgin Trains for a charter using the Pretendolino rake of Mk 3 stock Euston - Carlisle and back, out via Shap and back via the Settle and Carlisle; no special traction requested.

Now... this will include a profit margin so will be a bit above the actual cost, but the quote came back at £30,000...

Needless to say, a different staff entertainment option was immediately chosen...!!

That sounds around a ballpark figure - so about £45k+ these days.

I used to work on charter train planning and the pricing was pretty basic to say the least! But it could mount up quickly. On the example above, for example, using Shap both ways would save traction and traincrew costs - once we started buying in from other operators costs rose quickly.

 

[Jan Mayen]

In answer to your question, Jan, ORR reported 53 billion pax km in 22/23, yielding a cost/pax km of 47p or so. But 22/23 was partially Covid impacted and we were up at about 67 Bn pax km in 2019/2020 before covid hit. Adjusting for that would give a figure more like 40p/pax km.

Good to know. Thank you!

 

[Bald Rick]

For context, the staff costs for that service will be Sub £1k by quite a way.

The staff cost in terms of costs to the employer will be well over £1k by quite a way.

 

[TreacleMiller]

The staff cost in terms of costs to the employer will be well over £1k by quite a way.

Nah, they aren't. Factor in the pension, NI all the rest of it you'll be closer but not over for an "Hourly" rate for that particular service.

If you're going to count training and all the rest... That's literally how you don't do a costing exercise.

Ofcourse, revenue services cover every cost such as ECS and all the rest of it but even if you increased staffing costs by a factor of 10X this service listed here is still massively profitable.

 

[Bald Rick]

Nah, they aren't. Factor in the pension, NI all the rest of it you'll be closer but not over for an "Hourly" rate for that particular service.

You have to include the cost to the company of the marginal staff costs. It won’t just be the staff’s hourly rate. It will be the cost of providing the additional staff for the shifts necessary to get the trains prepped, off the depot, to Newcastle, down to London, and in one case then to Bounds Green. Then the same in reverse.

Being extra services on days of diversions it is unlikely to be accommodated from within cover, especially on a Sunday with early starts and late finishes and a diversion via the Durham Coast. So that means Rest Day payments.

I don’t know how the crew diagrams are going to work, but I’d be very surprised if there were fewer than 2 full shifts needed of both Driver and TM for each return trip. Plus the catering staff. Multiply by the RDW rate for drivers and TMs, add in a few shifts of catering crew, add in employers NI contributions of 13.8% (no need for pension as it’s RDW) and it will be comfortably more than £1,000 per return trip. Probably three times that based on two driver shifts, two TM shifts and four catering crew shifts.

 

[TreacleMiller]

You have to include the cost to the company of the marginal staff costs. It won’t just be the staff’s hourly rate. It will be the cost of providing the additional staff for the shifts necessary to get the trains prepped, off the depot, to Newcastle, down to London, and in one case then to Bounds Green. Then the same in reverse.

Being extra services on days of diversions it is unlikely to be accommodated from within cover, especially on a Sunday with early starts and late finishes and a diversion via the Durham Coast. So that means Rest Day payments.

I don’t know how the crew diagrams are going to work, but I’d be very surprised if there were fewer than 2 full shifts needed of both Driver and TM for each return trip. Plus the catering staff. Multiply by the RDW rate for drivers and TMs, add in a few shifts of catering crew, add in employers NI contributions of 13.8% (no need for pension as it’s RDW) and it will be comfortably more than £1,000 per return trip. Probably three times that based on two driver shifts, two TM shifts and four catering crew shifts.

Which isn't how these are or would be diagrammed.

Even if it was the case, and the entire crew were on a 12 hour RDW deal you'd be at less than 5% of the operational revenue. Note the figures given by the OP are a single journey.

The reason I'm stressing this point is because it's absolutely tiring to hear / read complaints here and elsewhere that high costs are due to high salaries. It's just not the case.

 

[fishwomp]

According to https://news.sky.com/story/average-ryanair-fares-up-24-as-company-forecasts-record-profits-13001754, the average fare paid on Ryanair was £50.

£12bn (annual) revenue, and £1.7bn profit - or 15%. (2023)

With an aircraft fleet with 198 seats (33 rows, 6 wide) per plane: so £10k in revenue per flight from those £50 tickets - or £8,500 cost (_all_ costs).

Average flight duration must be ~90 mins, possibly longer, they have the quickest turnarounds in the industry, so for a 15 hour day of flying that makes around 10 flights at most..

A plane makes £100k per day revenue - and £15k profit.

Per Bald Rick's LNER's rolling stock cost of £15k return from Newcastle to London for LNER (diverted via Durham coast, let's say £10k for direct return) - does that make a single 500 seat (?) train cost £5k, occupying the asset for 3.5 hours - in _rolling stock_ only?

Sadly, something doesn't add up

If O'Leary were running ECML without the government procuring his rolling stock, it might be 387s to Newcastle.. but some London - Cardiffs do that..(or a use for now scrapped Royal Mail 325s..)

 

[Bald Rick]

Sadly, something doesn't add up

Yes, you are using Ryanair’s average fare, when their average revenue per passenger ticket sold is rather higher!

 

[weeklycommuter]

You have to include the cost to the company of the marginal staff costs. It won’t just be the staff’s hourly rate. It will be the cost of providing the additional staff for the shifts necessary to get the trains prepped, off the depot, to Newcastle, down to London, and in one case then to Bounds Green. Then the same in reverse.
Being extra services on days of diversions it is unlikely to be accommodated from within cover, especially on a Sunday with early starts and late finishes and a diversion via the Durham Coast. So that means Rest Day payments.
I don’t know how the crew diagrams are going to work, but I’d be very surprised if there were fewer than 2 full shifts needed of both Driver and TM for each return trip. Plus the catering staff. Multiply by the RDW rate for drivers and TMs, add in a few shifts of catering crew, add in employers NI contributions of 13.8% (no need for pension as it’s RDW) and it will be comfortably more than £1,000 per return trip. Probably three times that based on two driver shifts, two TM shifts and four catering crew shifts.

The easiest way to reduce the cost of running the train is to get rid of the catering as many train operators have done.

 

[Trainbike46]

This thread has spurred me on to attempt to do some Calculations to work out the Energy Cost per Passenger Mile on the Railways for different types of trains.

According to data I found from the ORR, Class 390s, on average, consume about 29KWh/Mile. Let's assume that the trains are at capacity, which for a 9-Car 390 I think is 469 seats. So that works out to about 0.0618 KWh/Passenger Mile. Apparently, and correct me if I'm wrong, but TOCs pay £0.21/KWh for traction electricity. Giving us a rough total of £0.013/Passenger Mile.

Using similar data for the Class 379s, we find they average at about 17KWh/Mile and have a capacity of 209 seats. Again working this out with cost of electricity it comes to a total of about £0.017/Passenger Mile.

Finally, I thought I would investigate the costs for DMUs. According to a post on this site a single car of a Class 156 will use about 6mpg. This converts to 0.76 Litres of Diesel/Mile. Assuming a single car has 75 seats this equates to about 0.01 Litres/Passenger Mile. It seems quite hard to work out how much the TOCs pay for Diesel, but I think an upper bound would be around 80 pence/Litre. All together then £0.008/Passenger Mile.

This came as a bit of a shock to me as I assumed Electric traction would be much cheaper, and is a bit concerning. Delving a bit deeper it seems to me that this is due to old DMUs having less performance than EMUs so are using less energy and the railways being able to get Diesel much cheaper than Electricity (which is also true for the public).

Cost of energy is not the only factor to consider to compare the costs of DMUs and EMUs. My understanding is that DMUs need, in general, more maintenance than EMUs, as there are more parts, especially the engine. This then means that you need more trains to fill the same number of diagrams, as there is more time needed for maintenance.

I would also try and check your original data is right, it seems unlikely that a newer, high-density EMU would be less energy-efficient per passenger than an older high-speed EMU with lower seating density.

Maybe the best way would be to select a few comparable-ish TOCs that only use one type of traction (e.g., c2c, SouthEastern, and Chiltern) and divide total electricity for traction or diesel consumption by the total number of passenger miles from the same TOC?

 

[fishwomp]

Yes, you are using Ryanair’s average fare, when their average revenue per passenger ticket dold is rather higher!

True. Time for LNER to start selling lottery tickets..

 

[357]

True. Time for LNER to start selling lottery tickets..

Don't give them ideas

 

[fishwomp]

Yes, you are using Ryanair’s average fare, when their average revenue per passenger ticket sold is rather higher!

FWIW, there is more and fresher detail in their 2024 annual report. [https://investor.ryanair.com/wp-content/uploads/2024/06/Ryanair-2024-Annual-Report.pdf] - I was using 2023 press releases earlier.

Using 1.20 eur per gbp..

Average fare - FY2024: - €50 [£42],
Ancilliary revenue aka coffee, bags, lottery tickets, spurious charges and penalties 23.40€ [£19.50] per passenger.
So total per passenger - £70.

Cost per passenger avg 62€ [£52].

Average flight, 780 miles, and 9.4 hours of flying per plane per day.

That's pretty shockingly good value - and low cost - compared to rail.

 

[Annetts key]

What about taxes paid? Or rather, various tax exemptions?

Aviation remains heavily subsidised

The result of subsidies in the aviation sector is artificially cheap ticket prices, which serves to drive up demand and reduce incentives for more sustainable aviation.

Airlines pay zero duty on their fuel
But perhaps the biggest subsidy is the aid enjoyed by airlines through tax exemptions on their fuel and tickets. Airlines, unlike motorists across Europe, pay zero duty on their fuel. Plane tickets sold are mostly exempt from VAT. These subsidies total tens of billions of euro each year, money which could be reinvested in greening the economy or lowering labour taxes.

Above quotes are extracts from this web page

Historically, the aviation industry has benefitted from many tax breaks. Even now, no fuel duty is paid on jet fuel, and no VAT is applied. This irregularity becomes harder to explain when we look at how tax is applied to other transport modes. When compared to car travel, for example, aviation’s exemption from fuel duty and VAT appears more like an indirect subsidy that allows airfares to be kept artificially low. The absence of tax has helped to fuel passenger growth and the sector’s CO2 emissions have increased 125% since 1990.

Currently, the only tax levied on air travel is Air Passenger Duty (APD), which is paid by passengers, not airlines. No duty or VAT is paid on aviation fuel. If aviation did pay taxes equivalent to motorists it would raise a further £7.5 billion for the Exchequer (on top of the £4 billion already collected through APD).

Above quotes are extracts from this web page

 

[Trainbike46]

In addition to the good points by Annetts Key above, part of the reason for the lower cost at airlines compared to rail is the higher speed - if you run at higher (average) speeds, you can run the same timetable, and therefore transport the same passenger-km, with less staff and trains/planes. I believe Ryanair has relatively low (one could say unfairly low in some cases) pay as well. This is also contributes to why HS2 is expected to do well on operational costs compared to existing rail.

If I was trying to improve revenue to cost in UK rail as a whole (including infrastructure), I would probably look to:
- reduce ECS running
- speed up journey times
- increase reliability (I know, easier said than done)

Of course, where this can be done without massive investment it often has been done already, or is currently in the works.

 

[The exile]

If you're going to count training and all the rest... That's literally how you don't do a costing exercise

Depends what you want the “exercise” to prove!

 

[Western Lord]

FWIW, there is more and fresher detail in their 2024 annual report. [https://investor.ryanair.com/wp-content/uploads/2024/06/Ryanair-2024-Annual-Report.pdf] - I was using 2023 press releases earlier.

Using 1.20 eur per gbp..

Average fare - FY2024: - €50 [£42],
Ancilliary revenue aka coffee, bags, lottery tickets, spurious charges and penalties 23.40€ [£19.50] per passenger.
So total per passenger - £70.

Cost per passenger avg 62€ [£52].

Average flight, 780 miles, and 9.4 hours of flying per plane per day.

That's pretty shockingly good value - and low cost - compared to rail.

if anyone thinks that running an airline is a piece of cake, just look at the number of failures there have been over the years. Like it or not, Ryanair is an exceptionally well run business.

 

[fishwomp]

What about taxes paid? Or rather, various tax exemptions?

err.. you go and talk about not paying tax on fuel ... but that's also the case for railways.
In fact, far from paying tax, the railway receives actual subsidy to run trains.
Are there significant actual subsidies for airlines?

if anyone thinks that running an airline is a piece of cake, just look at the number of failures there have been over the years. Like it or not, Ryanair is an exceptionally well run business.

You could also see it as a sign of an industry that is efficient and adapts: failures from those that didn't adapt, or those that cut costs too much.

Not too much profit to be lazy, but enough to be successful.

 

[Annetts key]

err.. you go and talk about not paying tax on fuel ... but that's also the case for railways.
In fact, far from paying tax, the railway receives actual subsidy to run trains.

Red diesel used for locos and DMUs etc. is taxed at 5% VAT as I understand it.

The road fleet used by the railway companies (Network Rail for example) uses normal fuel from normal garages so pay the same as any company that runs road vehicles. And no, I don't mean the buses or coaches used as rail replacement.

Yes, the railway receives actual subsidy.

The airlines don't have to pay for physical infrastructure apart from costs to use airports and hangers for aircraft maintenance and for air traffic control, whereas railway lines have to be constructed and maintained between stations, as well as the actual stations and depots. There are lots of differences.

it's all rather complex.

The point I am making is that we are trying to compare two different transport systems each of which have or are subject to rather different costs.

 

[fishwomp]

Red diesel used for locos and DMUs etc. is taxed at 5% VAT as I understand it.

airlines pay passenger duty.

The road fleet used by the railway companies (Network Rail for example) uses normal fuel from normal garages so pay the same as any company that runs road vehicles. And no, I don't mean the buses or coaches used as rail replacement.

many of the road fleets used by airlines will go on the main road, so also use normal fuel. planes and trains don't run on roads, so this is a bit spurious, and 99% of all fuel is in the plane or train, not the van used to ferry something or other around a depot or to a site..

Yes, the railway receives actual subsidy.

The airlines don't have to pay for physical infrastructure apart from costs to use airports and hangers for aircraft maintenance and for air traffic control, whereas railway lines have to be constructed and maintained between stations, as well as the actual stations and depots. There are lots of differences.

'apart from costs to use airports and hangers'. right, so that sounds like paying to use physical infrastructure to me.. what other physical infrastructure should be paid for - the air, when was that constructed and by whom?!?

it's all rather complex.

The point I am making is that we are trying to compare two different transport systems each of which have or are subject to rather different costs.

it's not complex at all..

 

[PLY2AYS]

Even if it was the case, and the entire crew were on a 12 hour RDW deal you'd be at less than 5% of the operational revenue. Note the figures given by the OP are a single journey.

The reason I'm stressing this point is because it's absolutely tiring to hear / read complaints here and elsewhere that high costs are due to high salaries.

Agreed.

If a peak Plymouth or Edinburgh to London was DOO, staff costs would be 0.37% of ticket revenue. Bearing in mind, all staff (usually) do a minimum of 2 journeys a turn… usually more. So you have to divide staffing costs by how many journeys they are diagrammed to do.

 

[ChiefPlanner]

Just out of curiosity - how much per mile is a "standard" bus mile.

Peak obviously more than off peak ?

 

[negone]

Hope I can step in here, with some questions that have always puzzled me. I hope its not too much nonsense on my part.
I have always wanted to know if the difference in the number of train passengers carried actually makes a real energy cost difference, considering the heavy starting weight of a train, and the lesser weight of passengers, same with miles run and its likely empty seats aren't considered a loss just business. Fill the seats and you then get overcrowding, with hardly any financial gain, And that the fare structure seems to be just overlaid and has no bearing on costs per mile, or minutes run.
Why am I getting a 35% railcard fare discount, I cant go anywhere else and get that sort of discount. eg the bank wont give me £100 of money for £65. or the supermarket, petrol station or gas, water, electricity.