What Would a Good System for Working out Fares Look Like?

[DynamicSpirit]

(On yorkie's implicit request )

Many of us agree the current fares system is in many ways ludicrous and inconsistent. This has been highlighted in this thread, which points out that on Virgin trains leaving London during the evening peak, you can pay a lot more to travel to Preston than you would if you remain one extra stop to Lancaster.

It's easy to criticise the system, but I'd suggest devising a system that is good and doesn't have glaring inconsistencies is a lot harder. So - what would a reasonable, fair (sorry for the pun) system look like? Would it still have cheap returns? Would it be distance travelled based? Or as-the-crow-flies-to-destination based? How would peak restrictions work? Would there still be season tickets (which you could argue are unfair to people who travel occasionally). Any other ideas?

As a bit of material to start your ideas, here are some relevant quotes from that London-Preston thread linked to above:

I actually suspect a system that looks completely fair by mileage would be impossible - there are too many places where lack of direct routes, or presence of alternative routes would muck up either the consistency or the fairness of any attempted system.

For example, in an ideal world, how would you price a Brighton to Crowborough ticket? By likely actual travelled distance or by as-the-crow-flies distance to the destination? Should it be more or less expensive than a ticket from Brighton to Oxted (for which the person is on the train for a shorter distance but actually going to a destination that's considerably further away - and therefore you could argue is getting much more utility from the ticket).

Or - in an ideal world, should Horsham to Worthing be cheaper or more expensive than Horsham to Shoreham? After all, depending on your route, you could reasonably reach either destination via the other one!

The whole point of the thread is surely to illustrate a ticket anomaly, of which there are many. There used to be a similar anomaly on the evening peak train from Paddington to Carmarthen for journeys beyond Swansea.

The sooner that there is a move back to distance-based pricing, the better. In addition, trains should be designated as wholly off-peak or peak with respect to ticket restrictions for each segment of the trip. However, split-ticketing can often reduce the price and minimise the effect of peak fares, particularly where there are intermediate stops beyond which journeys are not subject to peak restrictions.

 

[yorkie]

Mileage based pricing is unviable because it would result in congested routes being too cheap, and little-used lines becoming too expensive. Penzance to Plymouth would become rather more expensive than Reading to London!

Here are some questions I asked before; the supporters of mileage based pricing are often reluctant to actually tell us how it would work and what the implication for fares would actually be...

1) Please give me the proposed fares from York to Whitby for a day return, avoiding Darlington, and tell me how this fare is calculated and whether you think it will be competitive with the coach service provided by Coastliner that costs only slightly more than the current rail ticket price but offers a shorter journey time? Also if I was to start the journey at Church Fenton, would that add to the cost?

2) Please give me the proposed fares from York to London for a day return, and if this is lower than current prices, tell me where the revenue will be brought in to compensate EC for the revenue loss?

3) Please give me the proposed fares for a return from Peterborough to Leicester, and also Peterborough to Nottingham, and let me know how you are calculating the mileage and whether under your system it will be permitted to travel from Peterborough to Nottingham via Leicester or not, and if someone does go via Leicester what excess will be charged (if any)?

4) Please give me the proposed fares for a return from London St Pancras to Ashford Intl (and any routeing options you wish to provide) and would those fares apply from, say, London Charing Cross to Ashford Intl? Please give me your calculations for different termini, if they vary..

 

[daodao]

Mileage based pricing is unviable because it would result in congested routes being too cheap, and little-used lines becoming too expensive. Penzance to Plymouth would become rather more expensive than Reading to London!

Here are some questions I asked before; the supporters of mileage based pricing are often reluctant to actually tell us how it would work and what the implication for fares would actually be...

In my opinion, the rate per mile for mileage based pricing should vary depending on the route and the quality of the train service offered, but would be equitable for that route, e.g. assuming that for the route London-Reading there is a fixed rate per mile for the entire distance, the sum of the prices for the journeys from London-Slough and Slough-Reading should equal the price from London-Reading. This would reduce the attractiveness of split ticketing. Not long ago, I made a day return journey from Hartford to (Birmingham) University and splitting the ticket at Stafford approximately halved the price!

However, the route Plymouth-Penzance could have a different rate per mile, so a journey from Plymouth-Par (similar distance to Paddington-Reading) could be cheaper to reflect the fact the service quality and speed is worse.

Where there are potential alternative routes, the price set should be determined by the shortest route, with alternative permitted routes having the same fare, but the permitted alternate routes should be clearly specified.

 

[yorkie]

In my opinion, the rate per mile for mileage based pricing should vary depending on the route and the quality of the train service offered, but would be equitable for that route, e.g. assuming that for the route London-Reading there is a fixed rate per mile for the entire distance, the sum of the prices for the journeys from London-Slough and Slough-Reading should equal the price from London-Reading. This would reduce the attractiveness of split ticketing. Not long ago, I made a day return journey from Hartford to (Birmingham) University and splitting the ticket at Stafford approximately halved the price!

However, the route Plymouth-Penzance could have a different rate per mile, so a journey from Plymouth-Par (similar distance to Paddington-Reading) could be cheaper to reflect the fact the service quality and speed is worse.

Where there are potential alternative routes, the price set should be determined by the shortest route, with alternative permitted routes having the same fare, but the permitted alternate routes should be clearly specified.

In that case I await with great anticipation what the figures will be

You say Penzance-Plymouth would be a different rate per mile than Reading-Paddington, but how would Penzance-Paddington be calculated?

Poppleton to Sheffield would be priced for 49 mile journey, but is valid by considerably longer distances than that, e.g. via Harrogate, Leeds, & Doncaster. So, what price is it to be?

So how much would Harrogate to Conisbrough cost? That's over 4 miles longer than Poppleton to Sheffield (by the shortest route) yet you could buy a Poppleton to Sheffield, start short & finish short! I think you are going to have some interesting anomalies.

 

[Gathursty]

I recall creating a thread on fare clusters many months ago. I wonder if revising which stations belonged to which cluster may yield a fairer pricing structure...

 

[daodao]

In that case I await with great anticipation what the figures will be

You say Penzance-Plymouth would be a different rate per mile than Reading-Paddington, but how would Penzance-Paddington be calculated?

Poppleton to Sheffield would be priced for 49 mile journey, but is valid by considerably longer distances than that, e.g. via Harrogate, Leeds, & Doncaster. So, what price is it to be?

So how much would Harrogate to Conisbrough cost? That's over 4 miles longer than Poppleton to Sheffield (by the shortest route) yet you could buy a Poppleton to Sheffield, start short & finish short! I think you are going to have some interesting anomalies.

Penzance-Paddington would be the sum of the (mileage rate x distance) for each sector that had a different rate/mile.

There would need to be compromises where there are alternate roundabout, but reasonable routes, and might need to be 2 prices in such a case (in some ways contradicting my previous comment): one for specified route(s) and a more expensive one for any permitted route where more roundabout routes are deemed acceptable, e.g. to Poppleton via Harrogate, from Leeds and beyond, in the example that you quoted. In this case, prices to Poppleton via Harrogate would be priced based on the longer distance via this route.

 

[Bletchleyite]

The only mileage based system that will remove *all* anomalies is an Anytime-only per-mile (or as I said "fractional power exponential" type system) based on rail miles with fully explicit routeing (i.e. a ticket is valid for *one* route and must be excessed to use any other route, be it longer or shorter).

Any other mileage based system will create anomalies of some kind.

 

[najaB]

The only mileage based system that will remove *all* anomalies is an Anytime-only per-mile (or as I said "fractional power exponential" type system) based on rail miles with fully explicit routeing (i.e. a ticket is valid for *one* route and must be excessed to use any other route, be it longer or shorter).

Which is why I always say to be careful what you wish for - fare 'simplification' will mean more expensive fares, fewer passenger rights or both.

 

[sheff1]

Poppleton to Sheffield would be priced for 49 mile journey, but is valid by considerably longer distances than that, e.g. via Harrogate, Leeds, & Doncaster. So, what price is it to be?

Each journey would be priced according to the route you wish to take in exactly the same way as happens in countries such as Germany. There, when you enter the your destination in the TVM it lists the next few trains with a price against each one. You make your choice using whatever cost, time, train type filters you prefer and pay for the appropriate ticket with the option to print the supporting itinerary if you want.

My response to your questions in post #2 remains the same as the last time you asked.

 

[Bletchleyite]

For clarity, SBB doesn't work that way, though DB does.

 

[sheff1]

Thanks, I thought it did, but have amended the post accordingly.

 

[Bletchleyite]

The way SBB do it is that you get asked to choose a route (not a specific connection), though each choice tends to include a few options e.g. "via Olten or Bern" being one route selection for Geneva to Zuerich. You can, unlike the UK, specify any route you like, and it'll calculate the fare, being kilometric. But that option is really something for enthusiasts and I suspect is rarely used by normal passengers who probably just take the obvious route. (The network is less of a web than the UK system, and more a set of sometimes-interconnected branches off a couple of main spines, with relatively few journeys where multiple routes make sense).

 

[cookie365]

Split the map into cells, each border between 2 cells has a crossing price set according to how busy/popular it is. Your ticket is valid on any route where the total crossing prices is less than that of the main route.

 

[Deerfold]

There would need to be compromises where there are alternate roundabout, but reasonable routes, and might need to be 2 prices in such a case (in some ways contradicting my previous comment): one for specified route(s) and a more expensive one for any permitted route where more roundabout routes are deemed acceptable, e.g. to Poppleton via Harrogate, from Leeds and beyond, in the example that you quoted. In this case, prices to Poppleton via Harrogate would be priced based on the longer distance via this route.

So you'd charge more for the direct service which could well be on a pacer and takes longer than for a journey that would mostly be on a far more modern train. Northern currently offer cheaper fares via Harrogate - I suspect that if this was reversed the line via Harrogate may require higher subsidies to survive. Are you happy with this as a side effect?

 

[Bletchleyite]

So you'd charge more for the direct service which could well be on a pacer and takes longer than for a journey that would mostly be on a far more modern train. Northern currently offer cheaper fares via Harrogate - I suspect that if this was reversed the line via Harrogate may require higher subsidies to survive. Are you happy with this as a side effect?

Provided the income to the railway as a whole remained the same, which would be achieved by setting the per-mile figure appropriately, franchise agreements could be renegotiated in order to move the requisite money around to counter this problem. They'd need to be renegotiated for any substantial fares system change anyway.

With regard to train "qualities" I suppose we could look at an InterCity supplement, but really I think there is no particular need to differentiate, as preferred train types are, as we have often seen here, very subjective. After all, there are people who would pay extra to go on a Pendolino and people who would pay extra not to - and I have waxed lyrical to a friend about the PKP Stadler FLIRT (interior a bit like a 350/1 but using the GWR HST style seats) and he wasn't impressed with it.

And, as I said, if you make it *too* complicated with supplements etc you practically have to go to per-train ticketing anyway, even if you allow free of charge/fare-difference-only changes.

 

[Camden]

Perhaps a system of flat fares to cover distance ranges?

London to Edinburgh, Glasgow = Fare 1

London to Liverpool, Manchester, Sheffield, Leeds = Fare 2

London to Birmingham, Coventry, Leicester = Fare 3

Birmingham to Liverpool, Manchester, Sheffield, Leeds = Fare 3

Liverpool, Manchester, Sheffield, Leeds to Edinburgh, Glasgow = Fare 2

and then local fare zones for local travel within areas.

Obviously you wouldn't want jumps between fare zones that seem unfair (ie, you wouldn't want Wigan and Preston in different zones) but that shouldn't be too hard to work out where they should be to and from, with overlapping.

Then it doesn't matter what route you take, as your ticket would in effect be from "North West England" to "London" rather than "Liverpool Lime Street" to "London Euston".

Franchises could then be let out as concessions, paid a fixed sum with performance incentives of various types on top. There would be no TOC specific tickets.

That would keep things simple all round.

 

[DynamicSpirit]

Mileage based pricing is unviable because it would result in congested routes being too cheap, and little-used lines becoming too expensive. Penzance to Plymouth would become rather more expensive than Reading to London!

On purely fairness grounds, if Penzance to Plymouth is actually further than Reading to London, then wouldn't it make some sense for it to be more expensive?

I can think of one counter-argument: There's a much better service from Reading to London, with much faster trains, and you could argue that should attract a higher fare. I wonder if one way of accommodating that would be for a system that's (roughly) based on mileage-pricing, but not rigidly so. And for it to be combined with supplements (say, around £3) for travel shorter distances on designated long-distance trains where more local trains are available. Thus, you might find that travelling Reading-London would be cheaper than Plymouth-Penzance if you use local stopping trains, but you'd pay a small premium to use the non-stop HSTs. Since those HSTs actually become the 'local' trains in Devon and Cornwall, the supplement wouldn't apply there.

Originally Posted by yorkie View Post
1) Please give me the proposed fares from York to Whitby for a day return, avoiding Darlington, and tell me how this fare is calculated and whether you think it will be competitive with the coach service provided by Coastliner that costs only slightly more than the current rail ticket price but offers a shorter journey time? Also if I was to start the journey at Church Fenton, would that add to the cost?

Could you argue that, with an integrated transport system, York to Whitby would be much more sensibly made by coach (or perhaps by train to Scarborough and coach from there), and it's not unreasonable for the fare to reflect that - and to be higher if someone wants to go on what is - as the crow flies - a somewhat (although not massively) roundabout route in order to use the train all the way.
(Yes, I do think that a decent fare structure would not only be somewhat mileage based but would feature integration with bus services).

4) Please give me the proposed fares for a return from London St Pancras to Ashford Intl (and any routeing options you wish to provide) and would those fares apply from, say, London Charing Cross to Ashford Intl? Please give me your calculations for different termini, if they vary..

I don't see any incompatibility between having a (roughly) mileage-based system and having the Ashford-London ticket as designated Ashford-London terminals. After all the actual differences in distance are relatively minor. Presumably travelling via HS1 would attract a supplement for travelling on a high speed train when there are slower ones available.

 

[Bletchleyite]

(Yes, I do think that a decent fare structure would not only be somewhat mileage based but would feature integration with bus services).

That kind of scenario is probably best handled with zonal fares of some kind, probably based on the "honeycomb" type zones the Dutch used to use. But even that isn't perfect.

 

[najaB]

I wonder if one way of accommodating that would be for a system that's (roughly) based on mileage-pricing, but not rigidly so. And for it to be combined with supplements (say, around £3) for travel shorter distances on designated long-distance trains where more local trains are available.

An excellent plan with just three potential points of contention (highlighted above). What is 'short' - it can't be a fixed distance nationally as the distribution of stations isn't uniform. Who decides which long-distance trains, what are the criteria? What is the threshold for 'more' local trains?

 

[DynamicSpirit]

An excellent plan with just three potential points of contention (highlighted above). What is 'short' - it can't be a fixed distance nationally as the distribution of stations isn't uniform. Who decides which long-distance trains, what are the criteria? What is the threshold for 'more' local trains?

I don't think you could do that by formula - it would have to be done on a case-by-case basis, but with 'long distance' (or 'inter-city' or whatever you want to call it) very clearly marked in timetables, and on all announcements.

Examples of where you might reasonably mark trains as requiring a supplement, based on today's timetable might be:

• All Virgin services South of Crewe/Manchester (but not North of there because they effectively become the local services from Warrington northwards)
• Great Western HSTs between Paddington and Didcot
• HS1 trains west of Ashford (the one place where this kind of supplement is in effect already charged)

For what counts as 'short' - I guess you could drop the restriction to 'short journeys' altogether and say ths supplement is payable for any journey that uses a designated train - for passengers who really are travelling long-distance, the supplement would be too small a proportion of the actual fare to be significant.

In an ideal system, I'd envisage a system of supplements replacing walk-on tickets that are only valid on particular operators, which would no longer be sold (except possibly in the case of open access operators)

 

[Bletchleyite]

For what counts as 'short' - I guess you could drop the restriction to 'short journeys' altogether and say ths supplement is payable for any journey that uses a designated train - for passengers who really are travelling long-distance, the supplement would be too small a proportion of the actual fare to be significant.

That's roughly how the IC-Zuschlag used to work on DB. It was something like DM7 in the late 1990s, which was a lot if you were going one stop (though you could choose to do that if you wanted) but next to nothing if going the length of the country.

The trouble with the UK compared to Germany is that we don't have a system of parallel IC and regional trains on most of the network, and we'd need to take a substantial hit to frequencies if we did. What, for example, is TPE?

 

[Deerfold]

Provided the income to the railway as a whole remained the same, which would be achieved by setting the per-mile figure appropriately, franchise agreements could be renegotiated in order to move the requisite money around to counter this problem. They'd need to be renegotiated for any substantial fares system change anyway.

You're also redirecting people off a quiet line onto one of the busiest corridors in Yorkshire.

It helps any argument to close or not invest in the Harrogate line even if you redirect money - in fact it makes it easier to argue the line is sucking money off other more successful lines.

On purely fairness grounds, if Penzance to Plymouth is actually further than Reading to London, then wouldn't it make some sense for it to be more expensive?

So long as you don't mind pricing people off using the trains in the South West, it's fine.

I can think of one counter-argument: There's a much better service from Reading to London, with much faster trains, and you could argue that should attract a higher fare. I wonder if one way of accommodating that would be for a system that's (roughly) based on mileage-pricing, but not rigidly so. And for it to be combined with supplements (say, around £3) for travel shorter distances on designated long-distance trains where more local trains are available. Thus, you might find that travelling Reading-London would be cheaper than Plymouth-Penzance if you use local stopping trains, but you'd pay a small premium to use the non-stop HSTs. Since those HSTs actually become the 'local' trains in Devon and Cornwall, the supplement wouldn't apply there.

The problem with this is the local trains between Reading and London are already rammed by the time they get to London. If you price people off the faster trains, they're going to be rammed by the time they leave Reading and local passengers won't get on at all.

 

[CarlSilva]

What about a pay what you think it's worth scheme ? It has been known to give good results in the catering trade, so might be worth a go.

 

[Bletchleyite]

What about a pay what you think it's worth scheme ? It has been known to give good results in the catering trade, so might be worth a go.

GTR might have financial problems

 

[TrainfanBen]

GTR might have financial problems

Good. Such a shambolic operation would be forced to close down.

I think that people who use quieter services should be refunded a % of the difference in ticket price.

For example;

Anytime single Preston-Birmingham £47.50
Some advance singles only cost £24

If the person bought an anytime but used the service that would have been £24
they should get maybe get X % of the difference in price back.

Maybe 25% so £5.875
Provided they did not break the journey and provided that the train was not full.

"Here you go sir, your £5 voucher for using a cheaper priced service." oh such a vivid imagination.

To summarise, I think the operators should incentivise people to use quieter services. Spread the load and therefore be able to offer more reasonable ticket prices.

 

[DynamicSpirit]

That's roughly how the IC-Zuschlag used to work on DB. It was something like DM7 in the late 1990s, which was a lot if you were going one stop (though you could choose to do that if you wanted) but next to nothing if going the length of the country.

The trouble with the UK compared to Germany is that we don't have a system of parallel IC and regional trains on most of the network, and we'd need to take a substantial hit to frequencies if we did. What, for example, is TPE?

Agreed. That is why I suggested we'd have to have trains designated as requiring a supplement for only a part of their routes, rather than - as I think was the case in Germany - the IC-Zuschlag designation being for a train. Thus, there might be portions of TPE that should attract a supplement - if there are places where local passengers should be encouraged to use more local services, but you certainly would just mark the whole of TPE as requiring a supplement.

 

[yorkie]

So, we have no actual figures to work with, so my questions still don't have proper answers.

We will have mileage based pricing but at various different rates, we will have supplements for fast trains, and journeys like Poppleton to Sheffield have to be bought with a routeing of "Via: Harrogate/Leeds/Doncaster" or "Via: York/Pontefract/ NOT Leeds/Doncaster" or "Via: Leeds/Wakefield NOT Harrogate/Doncaster" and so on and so forth?

It would take an age to buy the ticket, by which time the train would be gone and the guard would be calculating an excess fare for you on the next train!

Presumably the supplement would only apply to the trains via Doncaster, as if you go via Leeds, the speed does not exceed 100mph?

And I thought the current system was complicated

 

[bb21]

Good. Such a shambolic operation would be forced to close down.

So many things wrong with that statement.

What you want is an improved offering, certainly not it being closed down.

 

[Bletchleyite]

To summarise, I think the operators should incentivise people to use quieter services. Spread the load and therefore be able to offer more reasonable ticket prices.

They do - Advance and Off Peak. Go any more granular than that and you are just getting hideously complicated.
--- old post above --- --- new post below ---

Agreed. That is why I suggested we'd have to have trains designated as requiring a supplement for only a part of their routes, rather than - as I think was the case in Germany - the IC-Zuschlag designation being for a train. Thus, there might be portions of TPE that should attract a supplement - if there are places where local passengers should be encouraged to use more local services, but you certainly would just mark the whole of TPE as requiring a supplement.

Again really complicated.

FWIW DB did have exceptions - some of the one-off ICs that ran onto branch lines would run as IC until the junction station, then would be designated RB/RE and thus supplement-free. (Thus giving some RB/REs with restaurant cars). The Hamburg-Berchtesgaden (Königssee) was one such train.

They also sometimes designated an IC supplement-free (Zuschlagfrei) where there were few local services or high local demand e.g. Muenchen-Salzburg.

All this was shown on the PIS - "IC zuschlagfrei", "IC mit Zuschlag" or "ICE - Besonderer Fahrpreis[1]".

However...there was then the complexity that IR and D required a lower supplement (I think DM3, which at the time was about a quid so barely worth collecting) for journeys under 50km. Which you'd have to work out yourself or wait to see if the guard bothered or not.

[1] ICEs have been market-priced since day one.

 

[Haywain]

GTR might have financial problems

Good. Such a shambolic operation would be forced to close down.

They won't be forced to do anything as they are on a management contract and are not revenue dependant.

I think that people who use quieter services should be refunded a % of the difference in ticket price.

For example;

Anytime single Preston-Birmingham £47.50
Some advance singles only cost £24

If the person bought an anytime but used the service that would have been £24
they should get maybe get X % of the difference in price back.

Maybe 25% so £5.875
Provided they did not break the journey and provided that the train was not full.

"Here you go sir, your £5 voucher for using a cheaper priced service." oh such a vivid imagination.

To summarise, I think the operators should incentivise people to use quieter services. Spread the load and therefore be able to offer more reasonable ticket prices.

I wonder how you think this would be done? Perhaps a smartcard system that required full payment in advance and touching in and out at the start/end of the journey and on the train during the journey? Sounds a touch complex to me, to achieve much of what our fares system already does.