What if rail travel was charged strictly by the mile.

[ryan125hst]

In which case why change anything?!

You have just literally described the existing system. The only difference is that rather than using market based pricing you're using mileage based pricing. And because that leads to some fares jumping in price and others falling dramatically you're also applying other modifiers to boost/lower the price. Also you're now introducing a whole raft of TOC/Route specific tickets where once there was just an Any Permitted. I'm unclear how any of this is an improvement for the passenger?!

That's a good point. It would probably be best then to simply price for the shortest route and allow the ticket on longer routes where they exist. After all, longer routes will mean longer journey times, which passengers (except rail enthusiasts ) won't want.

Charging 40p per mile appears to be reasonable for fairly short routes. I've just had a quick look at Retford to Sheffield which is 24 miles. It works out as £9.60 single and £10.56 return (Anytime tickets) (multiplying it by 1.1). This compares very well with the current fare as a single is, surprisingly £9.60, and a return is £10.50.

It appears that 40p per mile is reasonable for most short to medium distance routes (20 to 80 miles).

That said, Retford to Worksop costs £4.10 single and £4.20 return. Under my system, it would cost £3.20 for a single and £3.52 for a return. It's not too far off, but probably slightly too low for the TOC.

The main advantage of the system would be the fact that passengers would be able to see exactly how fares are calculated. It does appear to be close to 40p per mile for Off Peak fares (or in the case of the two I've just mentioned, the Anytime fare as Off Peak tickets don't exist), they do vary significantly and the Anytime tickets even more so.

To summarise my system:
For an Off Peak Single ticket:
40p per mile on journeys under 100 miles.
60p per mile on journeys over 100 miles.

Off Peak Single to Off Peak Return: Multiply by 1.1
Off Peak ticket to Anytime: Multiply by 1.4 (as 30% proved to be too low in the examples earlier.
Off Peak ticket to Super Off Peak: Multiply by 0.7 (So a 30% reduction).

Child fares: 50% off
First Class fares: Multiply the relevant fare by 1.5

Apologies if I appear to have taken over this thread. I don't know if Harlesden would agree?

 

[talltim]

I know we are talking theoretical here, but I can't help thinking that pure mileage is too simplistic
I think you need to also allow for speed of journey and frequency of service.
Miles x average speed x trains per hour x price multiplier

 

[Peter Mugridge]

I know we are talking theoretical here, but I can't help thinking that pure mileage is too simplistic
I think you need to also allow for speed of journey and frequency of service.
Miles x average speed x trains per hour x price multiplier

Please don't give the DfT any new ideas that would complicate things further!

 

[theblackwatch]

In the Czech Republic, there is a 'basic fare' charged per km, and it seems to work (given that I've not heard of any mass riots, or seen any when I've been over there). The basic structure is at http://www.cd.cz/assets/vnitrostatni-cestovani/jizdenka/ceniky-jizdneho/tariff1-9-12-2012.pdf but there are also other fares such at day tickets and tickets for groups (see http://www.cd.cz/vnitrostatni-cestovani/jizdenka/ceniky-jizdneho/-9526/ ).

 

[yorkie]

Thanks ryan125hst, you have looked into this in more depth than most people who have suggested it have done, and it's interesting that you and Harlesden reached opposite conclusion on whether there should be a discounted rate per mile for a longer distance ticket, or a shorter distance ticket. I'm not saying either is right or wrong, but it highlights the scale of the problem.

We can also reach the conclusion that it will still be cheaper to "split".

It's not been stated what peak restrictions would apply, but with the multiplier for Anytime, and discount for Super Off Peak, applying to the whole mileage, it will always be cheaper to "split" where the journey starts at a time considered "peak" but the train goes on to call at stations from where Off Peak tickets for part of the journey become valid.

To summarise my system.....

OK here are some calculations then (mileages are approximate, with rounding. Fares are not rounded to the nearest 5p), taking your proposals, and for Harlesden I've taken his mileage figures along with your multipliers (but will happily change them if Harlesden requests), and applied them to journeys from Leeds to both Burton and Tamworth, on the XC route.

I'll also see what happens if the passenger buys a ticket from a previous station before Leeds (Church Fenton - where the line via Pontefract diverges) or a ticket to the next station beyond Tamworth (Wilnecote).

Church Fenton - Burton (83 miles):

          Off Peak          Super OffPeak         Anytime
 m    Single    Return    Single    Return    Single    Return
83    £33.20    £36.52    £23.24    £25.56    £46.48    £51.13 ryan125hst
83    £62.25    £68.48    £43.58    £47.93    £87.15    £95.87 harlesden
--    £46.00    £47.00      n/a       n/a     £64.50   £108.50 NFM16

Leeds - Burton (86 miles):

          Off Peak          Super OffPeak         Anytime
 m    Single    Return    Single    Return    Single    Return
86    £34.40    £37.84    £24.08    £26.49    £48.16    £52.98 ryan125hst
86    £64.50    £70.95    £45.15    £49.67    £90.30    £99.33 harlesden
--    £41.20    £42.20      n/a       n/a     £42.50    £61.50 NFM16

Under your proposals, it becomes cheaper to buy a ticket from Church Fenton, as the mileage assumes travel via Pontefract, but you said that permitted routes remain valid.

Under your fare proposals, the fares are cheaper than at present, so XC would demand an increase in subsidy. In reality, anyone doing that journey would use "split tickets" anyway, unless they were a business user on expenses!

However it's unclear what the Off Peak / Super Off Peak restrictions would be, and until that's known, a fare for an actual journey could not be calculated.

Leeds - Tamworth (99 miles):

           Off Peak          Super OffPeak         Anytime
 m    Single    Return    Single    Return    Single    Return
99    £39.60    £43.56    £27.72    £30.49    £55.44    £60.98 ryan125hst
99    £64.35    £70.79    £45.05    £49.55    £90.09    £99.10 harlesden
--     n/a      £51.80      n/a       n/a     £49.00    £98.00 NFM16

Leeds - Wilnecote (101 miles):

           Off Peak          Super OffPeak         Anytime
  m    Single    Return    Single    Return    Single    Return
101    £60.60    £66.66    £42.42    £46.66    £84.84    £93.32 ryan125hst
101    £65.65    £72.22    £45.96    £50.55    £91.91   £101.10 harlesden
--      n/a      £51.80      n/a       n/a     £49.00    £98.00 NFM16

For Leeds - Tamworth, the same applies where your fares are cheap (apart from your Anytime single!), but look what happens to Harlesden's fare: Leeds to Tamworth is cheaper than Leeds to Burton! Tamworth is one stop beyond Burton. That is because a cheaper mileage rate applies. Great - more anomalies

With your fares, there is a huge jump if the mileage goes up to 100 miles. So a passenger doing a Leeds to Wilnecote journey would save money by purchasing a ticket to Tamworth, and hiring a TAXI both ways between Tamworth & Wilnecote!

This would mean that "split ticketing" would save even more than at present.

Apologies if I appear to have taken over this thread. I don't know if Harlesden would agree?

Not at all! You've given it a good go, and let me pick holes in it. Happy to do the same with Harlesden if he wants to add some amendments to his figures.

 

[fowler9]

In the Czech Republic, there is a 'basic fare' charged per km, and it seems to work (given that I've not heard of any mass riots, or seen any when I've been over there). The basic structure is at http://www.cd.cz/assets/vnitrostatni-cestovani/jizdenka/ceniky-jizdneho/tariff1-9-12-2012.pdf but there are also other fares such at day tickets and tickets for groups (see http://www.cd.cz/vnitrostatni-cestovani/jizdenka/ceniky-jizdneho/-9526/ ).

Yeah but is a bit less than a pound a mile. Watch in amazement as the UK's commuter railways die a death if they do that here.

 

[PermitToTravel]

Any change to the fare structure would have to:
- be revenue neutral for the TOC/NR (so the cash coming in to run the service stays the same)
- provide EVERY passenger with a ticket that is at worst the same price, (but everybody really wants a cheaper ticket for their journey under a new system).
- be low cost to implement (as otherwise there is no chance the money will be available to do it)
- be easy to understand (as the main moan about tickets is they are too difficult to follow)

If anybody can come up with a system that ticks all these boxes, they deserve a Nobel Prize (probably the one for fiction!)

The two bold criteria are mutually exclusive, unless no changes are made. If the price of any ticket is decreased, the price of another must be increased (violating rule 2) to maintain revenue neutrality (rule 1)

I can never see the fares being changed to a different model:

If fares on a route go up, the public will complain

If fares on a route go down, the TOC's will complain with tax payers footing the bill.

It doesn't seem worth while.....

I'm not sure that either of these are as strong an argument as mileage based pricing simply being inappropriate for the British railway network. Fares do regularly go up, and people do complain. TOC revenues per route do sometimes go down. If a new pricing system is viable, it should not automatically be dismissed just because it is different.

 

[ainsworth74]

The two bold criteria are mutually exclusive, unless no changes are made. If the price of any ticket is decreased, the price of another must be increased (violating rule 2) to maintain revenue neutrality (rule 1)

That's the point

 

[Metrailway]

Although having different charges for differing mileage, ryan125hst and Harlesden's pricing structure is not tapered and therefore several anomalies arise, leading to the incentivisation of split ticketing as highlighted by Yorkie.

IIRC McNulty estimated the cost to the railway of transporting a passenger a mile as approximately 35p (incl track access).

This cost could be the basis of a flat mileage system. If we add 10p profit for the railway, we'll make our flat charge 45p per mile.

For Yorkie's examples that would mean:

Church Fenton - Burton (83 miles): £37.35
Leeds - Burton (86 miles): £38.70
Leeds - Tamworth (99 miles): £44.55
Leeds - Wilnecote (101 miles): £45.45

and for some other examples (shortest route):
Aylesbury - Wendover (6 miles): £2.70
Manchester - Leeds (43 miles): £19.35
Euston - Manchester (184 miles): £82.80
Coventry - Penzance (316 miles): £142.20

In the past when we did have flat p per mile, generally the fare charged was based on the shortest route.

Personally I'm not in favour of a purely flat rate as it favours short hop journeys which could be done by bus or foot, and makes long distance journeys, where the railway is competitive with other modes, expensive.

However, it would be easy to administer.

We could have a more tapered fare system, which favour long distance travelling.

So lets say for arguments sake we have 4 tariffs: 0-10 miles at 60ppm; 11-40 miles at 50ppm; 41-60 miles at 40ppm; and 60+ miles at 30ppm. The tariff rate is charged for the mileage accrued in the tariff range. E.g a journey of 75 miles would cost £33.50 [(10*60)+(30*50)+(20*40)+(15*30)]

So for:

Aylesbury - Wendover (6 miles): £3.60 (Avg = 60ppm)
Manchester - Leeds (43 miles): £22.20 (Avg = 51.6ppm)
Church Fenton - Burton (83 miles): £35.90 (Avg = 43.3ppm)
Euston - Manchester (184 miles): £66.20 (Avg = 37ppm)
Coventry - Penzance (316 miles): £105.80 (Avg = 33.5ppm)

 

[yorksrob]

I know I'm beginning to sound like a broken record, but of course, a lot of the value for money issues would be solved with a National Railcard.

I might actually take some notice of ATOC if they instigated something useful like this, rather than banging on about how every good thing that's happened over the last twenty years is entirely down to privatisation.

 

[Flamingo]

The two bold criteria are mutually exclusive, unless no changes are made. If the price of any ticket is decreased, the price of another must be increased (violating rule 2) to maintain revenue neutrality (rule 1)

Exactly my point!

 

[soil]

IIRC McNulty estimated the cost to the railway of transporting a passenger a mile as approximately 35p (incl track access).

That doesn't make any sense.

Let's take an example, East Coast, which does 3 billion passenger miles each year.

https://www.gov.uk/government/publications/rail-subsidy-per-passenger-mile

They are a nationalised TOC, and according to the most recent accounts

http://www.directlyoperatedrailways.co.uk/PDF/DORReportAccounts2013.pdf

they run broadly at break-even (and with only a nominal subsidy of £15m), spending £690 million per year, around 23p/mile.

Northern Rail, meanwhile, do 1.3 billion passenger miles, spending £590 million, or 45.3p/mile a mile, basically double what East Coast spend.

http://www.scribd.com/doc/183059895

In addition, their passenger income is only £216m 16.5p/mile: the government actually pays more per passenger mile than the passenger dpes.

So it's basically nonsense to say that the railways costs 35p/mile

Also not all railways are run to the same priorities. Some regional public transport operators are subsidised for political reasons. In these areas passengers will pay less.

In other areas public transport is subsidised because it wouldn't work without subsidy. If Northern were to operate without subsidy, many stations would close. Even within their network some lines would be more viable than others.

This cost could be the basis of a flat mileage system. If we add 10p profit for the railway, we'll make our flat charge 45p per mile.

But the railway doesn't make that much profit, 10p per mile is 29% profit. That's ludicrously generous. First Group, for example, makes £335 million on £6.9 billion in revenue, below 5%. Why would you hand them such a fat profit rise.

 

[PermitToTravel]

That's the point

Exactly my point!

I'm an idiot

 

[Flamingo]

I'm an idiot

The current fare structure evolved, it was not designed in any real sense. Trying to change it runs up against those two conflicting aims.

 

[Metrailway]

That doesn't make any sense.

...snip...

So it's basically nonsense to say that the railways costs 35p/mile

Note these aren't my figures - McNulty suggested that railways as a whole were as cheap as that. But I suggest you look at the graphs in his report:

2.3

This chart shows that industry expenditure was ~22p/ passenger km which is ~35p/ passenger mile.


2.10

This chart shows that it costs 20.2p/ passenger km (excluding track access) which is approx 32p/mile.

And it isn't just McNulty. A BBC article in 2012 claimed that the average cost of the railway is approximately 20p per mile! :

The Association of Train Operating Companies (Atoc) says it operates services between 2,500 stations so cannot break them all down on a cost per mile basis.

Once upon a time, it would have been simple. Up until 1968 British Rail used a rigid price formula of 3.25d per mile (1.35p in decimal money).

Adjusted for inflation it works out at about 20 pence per mile. Perhaps surprisingly the average cost per mile today across the whole of the network is indeed about 20p. Railway expert Barry Doe says that in practice little changed until privatisation, at which point prices quickly diverged between the cheaper and more expensive routes.

An ATOC presentation in 2013 shows that the average fare paid per mile is 20.4p.

 

[Oswyntail]

The real problem, as illustrated in "Flamingo's Paradox", is that we are looking at fares in isolation. However, if we:-

• Changed the system of calculating fares to something more transparent, such that the customers could understand it
• Changed the routing system, and its evil twin income distribution, so that it was also clear to the customers
• Changed the method of financing TOCs so that they were more incentivised to provide what customers needed
• And did this simultaneously, while still recognising that local authorities can provide subsidies to meet their local priorities.
• And took running this out of the hands of TOCs and put it in the hands of a single body

Then a simple, balanced, comprehensible system could be devised.

 

[PermitToTravel]

I understand the desire for routing to be understandable for customers, but what would the advantages be for customers being able to calculate fares? It might be beneficial for a passenger who knew not the fare between two stations, but only the exact mileage by rail for the shortest route; I don't think anyone else would find it useful. It might be a nice-to-have, but not at the expense of constraining rail fares to being calculated by a formula.
Revenue allocation is the same. Some of us would like to know, but knowing is not worth changing the system for (and anyway, that one is calculated by a fixed formula, but not one that they share with us).

 

[orpine]

I like the idea of per-mile costs. Rather than a fixed fee charge a different one depending on whether it's a commuter or long-distance, but other than that fixed-fee would make life easier and fairer for customers.

I understand the desire for routing to be understandable for customers, but what would the advantages be for customers being able to calculate fares?

Because there are countless examples where the default ticket for a journey isn't the cheapest, despite the cheapest being valid for the same journey. An obvious example:
Warwick to Leicester - 28.90 - offpeak return.
versus:
Warwick to birmingham - £7.20 - offpeak day return.
Birmingham to leicester - £12.50 - offpeak day return.
Total: £19.70 for an offpeak day return
So I'm being charged £9.20 for the privilege of walking between Moor street and New Street. And of course having the option to return any time in the next 30 days (which I don't want). Most people don't know they have to "shop around" to not get ripped off. Almost every journey I do has a cheaper alternative than a standard ticket that has similar validity. If things were priced per mile that wouldn't be an issue (assuming it was done properly - big assumption ).

 

[soil]

And it isn't just McNulty. A BBC article in 2012 claimed that the average cost of the railway is approximately 20p per mile! :

Adjusted for inflation it works out at about 20 pence per mile. Perhaps surprisingly the average cost per mile today across the whole of the network is indeed about 20p. Railway expert Barry Doe says that in practice little changed until privatisation, at which point prices quickly diverged between the cheaper and more expensive routes.

That BBC article is awful.

It makes no mention of passenger miles, and is actually talking about season ticket (per week, month, year ???) costs per mile divided by some arbitrary number of (return?) journeys per week.

An ATOC presentation in 2013 shows that the average fare paid per mile is 20.4p.[/QUOTE]

Right, so costs per mile are 35p, and fares paid average 20p/mile.

So we need to double fares. Right?

Costs per mile are nonsensical.

Two choices:

1) Drive Bodmin to Plymouth, 65 miles round trip, driving time 50 minutes door-to-door, free parking, maybe £60/week fuel
OR
Train Bodmin to Plymouth journey time 1 hour door-to-door

2) Drive Maidenhead to London, 65 miles round trip, journey time 2 hours, weekly congestion charge £40, weekly parking £100
Train Maidenhead to London, journey time 50 minutes door-to-door

Which train is going to be more popular? Where are local wages going to be higher? What should each respective ticket cost?

Did you know that median wages in certain commuter areas close to London over £60,000 per year? What would they be in Plymouth?

 

[Oswyntail]

...but what would the advantages be for customers being able to calculate fares?

If the public were confident that rail fares were calculated fairly, along known, understood and clear principles, then there would be more confidence in rail itself. At present, virtually any anti-rail publicity can have a moan about expensive fares, and the only answer is that they are cheap if you commit to travel at a different time to a different destination.

...Revenue allocation is the same. Some of us would like to know, but knowing is not worth changing the system for (and anyway, that one is calculated by a fixed formula, but not one that they share with us).

It may not be revenue allocation as such that I mean, but the system whereby fares for particular routes, or parts of routes, are set by TOCs, presumably to meet their own requirements.
Recently, much political capital was made of the PM not being able to say what the price of a loaf of bread was. This was possible because, broadly speaking, people across the country can come up with the same answer, allowing for differences in type. The rail industry, imho, should be aiming for this clarity and consistency.

 

[sheff1]

So again I return from travel in a country where rail fares are calculated by distance only to find a thread on here discussing why such a system can't work.

C'est la vie!

 

[Flamingo]

So again I return from travel in a country where rail fares are calculated by distance only to find a thread on here discussing why such a system can't work.

C'est la vie!

Nobody is saying it would'nt work, just that there would be losers as well as winners and nobody wants to be a loser.

 

[swt_passenger]

Warwick to Leicester - 28.90 - offpeak return.
versus:
Warwick to birmingham - £7.20 - offpeak day return.
Birmingham to leicester - £12.50 - offpeak day return.
Total: £19.70 for an offpeak day return
So I'm being charged £9.20 for the privilege of walking between Moor street and New Street. And of course having the option to return any time in the next 30 days (which I don't want). Most people don't know they have to "shop around" to not get ripped off. Almost every journey I do has a cheaper alternative than a standard ticket that has similar validity. If things were priced per mile that wouldn't be an issue (assuming it was done properly - big assumption ).

They don't have 'similar validity' unless you come back the same day, because you are comparing a return with a set of day returns. What you are getting extra is the ability to spend more time at your destination.

Day returns are sold as a discount over (period) returns to encourage people to make local out and back day trips. Day returns and returns are available for journeys of different distances. They are not the same product.

 

[dzug2]

So again I return from travel in a country where rail fares are calculated by distance only to find a thread on here discussing why such a system can't work.

C'est la vie!

There can't be many such countries left

 

[sarahj]

Germany still has such a system. Basic fares are calculated on a km basis. If you wanted to travel by a faster train, then you paid the supplement on top. However, many ICE fares are not calculated this way. On top of this you now have special long distance fares which are not done distance either, but rather the same price for many destinations. I've also seen some sly ones as well, a low price, then in the small print (with Bahncard 50).

However, recently I've seen an ad for rival Hamburg-Cologne service and its tag line was how cheap in cents it was per km.

 

[radamfi]

The Netherlands, Belgium and Switzerland have quite pure distance based fares. You can work out the basic fare from a table such as

http://www.ns.nl/binaries/content/assets/NS/reizigers/tarieven-eret-2013

The Netherlands has various railcards giving discounts between 20% and 100% off depending on the time of day/week. The high speed line between Breda and Schiphol incurs a small supplement, as does the ICE train between Arnhem and Amsterdam. Switzerland has a 50% off railcard which is so widespread that the website gives prices with the discount by default.

In Belgium, it is rarely worth regular travellers getting the (already cheap) full price ticket as there are absurdly cheap deals, notably the Rail Pass giving 10 single trips, any distance, a year for €76 a year and a similar ticket for short distance trips.

 

[orpine]

They don't have 'similar validity' unless you come back the same day, because you are comparing a return with a set of day returns. What you are getting extra is the ability to spend more time at your destination.

Day returns are sold as a discount over (period) returns to encourage people to make local out and back day trips. Day returns and returns are available for journeys of different distances. They are not the same product.

Maybe, but most users won't be aware of that distincton.
Furthermore, why am I being charged more to return at a later date in the first place? The cost to the railway network is the exact same whether I return today or in 28 days. If I wanted to stay longer, that's what a pair of advances would start coming in useful for.

 

[bb21]

Furthermore, why am I being charged more to return at a later date in the first place?

How long is a piece of string?

 

[fowler9]

Blimey Orpine. When I was visiting the Ex in Warrington from Liverpool I had to get 2 singles because it wasn't far away enough to get anything else over the course of a weekend.

 

[orpine]

How long is a piece of string?

Double half it's length. Why?