Low-hanging fruit for fare reform

[DynamicSpirit]

As I've mentioned before, regardless of what may or may not be the case for other services, I would be surprised were HS2 services, certainly those involving the actual HS2 route, not compulsory reservation with fully airline style pricing and ticketing totally separate from the rest of the railway. Thus if HS2 is faster, you'll buy (presumably in one transaction) an HS2 ticket and a relevant local ticket.

Compulsory reservation pretty much destroys the whole point of a turn-up-and-go service if it means people will only be able to travel if they know in advance what train they'll be getting.

 

[A S Leib]

Compulsory reservation pretty much destroys the whole point of a turn-up-and-go service if it means people will only be able to travel if they know in advance what train they'll be getting.

Mandatory reservations if travelling from Curzon Street to Scotland or from Lancaster to London, turn-up-and-go for Manchester to London? (No, this is not a serious suggestion.)

I also can't see season tickets not being accepted for HS2, and I don't think mandatory reservations work well with season tickets (or in general, but to a greater extent).

 

[Bletchleyite]

Mandatory reservations if travelling from Curzon Street to Scotland or from Lancaster to London, turn-up-and-go for Manchester to London? (No, this is not a serious suggestion.)

I don't have stats for Manchester to London but I bet most people are using the train on their itinerary.

It is 3tph but that is more about capacity.

 

[HSTEd]

In the medium term, the dominant (in traffic terms) service on HS2 is likely to be some form of high intensity shuttle running London-BIrmingham Interchange-Curzon Street. If it ran at, say, six trains per hour, it could provide a turn up and go service that would make doubling back an attractive strategy.

Given that people doing that would be beneficial for the railway, we probably don't need to worry about it in the fare system, but routing guidess hould include lots of double back via HS2 options as acceptable.

 

[DynamicSpirit]

In the medium term, the dominant (in traffic terms) service on HS2 is likely to be some form of high intensity shuttle running London-BIrmingham Interchange-Curzon Street. If it ran at, say, six trains per hour, it could provide a turn up and go service that would make doubling back an attractive strategy.

Given that people doing that would be beneficial for the railway, we probably don't need to worry about it in the fare system, but routing guidess hould include lots of double back via HS2 options as acceptable.

I think it would be a problem the railway would need to worry about. Let's say, it turns out (plausibly) that London to Solihull is quicker via HS2 and walking from Curzon Street to Moor Street (outside the barriers) than taking a direct train from Marylebone. Clearly in any reasonable fares system you'd want the walk-up London to Solihull fare to be less than the London to Birmingham fare because it's not as far. But if you allow London to Solihull via HS2 on a lower-than-Birmingham fare, that's an open invite to everyone who wants to go to Birmingham to just buy a cheaper Solihull fare and not show up for the Moor Street-Solihull portion of the journey. So as a minimum, you really do need two London-Solihull fares: Via Birmingham and Not via Birmingham.

The other thing is, I would say a basic requirement for a good fares system is that most travellers perceive it as fair and reasonable. I would reckon most members of the public would see it as quite fair and would even expect that you pay more for a longer distance. And I'd imagine most passengers would also think it reasonable that if there are two vastly different ways to get to the same destination, you might pay a bit less for the slower route (because that's not providing such good service). On both of those measures, charging more to go via HS2 to Solihull would likely be seen as fair, and charging the same amount as a direct train would be seen as unfair.

 

[crablab]

And I'd imagine most passengers would also think it reasonable that if there are two vastly different ways to get to the same destination, you might pay a bit less for the slower route (because that's not providing such good service).

I'm not sure if I agree; the slower route provides a different type of service - one which stops in places the faster service doesn't.

Reading to Paddington is served by both GWR and Elizabeth Line services. Tickets you buy along that route are valid via either operator but obviously they provide different types of service. There is no price differential (off peak tickets have some restrictions).

Many Reading passengers just travel on the direct 'fast train' to and from London, but it's not clearcut since there are semi-fast GWR services do stop at intermediate stations which can be useful (eg. if you have an appointment in Maidenhead).

The lack of price differential is important. You *do not* want to incentivise the Reading commuters to get the Elizabeth Line - these trains fill up very quickly heading into London as it stands.

The current situation works nicely: most of the Reading commuters take direct services as they're the fastest. Those at intermediate stations take the most convenient (GWR semi-fast or EL) service for them.

Those that want/need to take the EL for whatever reason can.

I don't think making EL tickets cheaper or adding TOC restrictions would achieve many positive benefits, except perhaps the long distance travellers who intermix with commuters at Reading (but that's kinda tough; it's the best use of finite capacity and otherwise those trains would be carting fresh air around). Indeed, it would likely be overall quite negative.

I think the same principle applies to some other routes (eg. Peterborough to Kings Cross, Cambridge to London) which do have pricing distinctions.

All to say, I don't think this is a problem best solved by adding lots of complexity to the fares system, and the lack of these differentials doesn't cause issues elsewhere. If there's specific issues with overcrowding then set down/pick up timetable restrictions are available, which are enforceable ticketing wise.

 

[HSTEd]

I think it would be a problem the railway would need to worry about. Let's say, it turns out (plausibly) that London to Solihull is quicker via HS2 and walking from Curzon Street to Moor Street (outside the barriers) than taking a direct train from Marylebone. Clearly in any reasonable fares system you'd want the walk-up London to Solihull fare to be less than the London to Birmingham fare because it's not as far. But if you allow London to Solihull via HS2 on a lower-than-Birmingham fare, that's an open invite to everyone who wants to go to Birmingham to just buy a cheaper Solihull fare and not show up for the Moor Street-Solihull portion of the journey. So as a minimum, you really do need two London-Solihull fares: Via Birmingham and Not via Birmingham.

Ah yes, I see what you mean.
London-Solihull would have its kilometreage set by the classic line, even if hte dominant way to do the journey is via HS2.
Since you have to let people out of the barriers at Curzon Street with a Solihull ticket, it would form the functional London-Birmingham ticket .

Hhm, this is an interesting dilemmna.
We would certainly need HS2 and non HS2 tickets if we allow such aggressive doubling back.

The other thing is, I would say a basic requirement for a good fares system is that most travellers perceive it as fair and reasonable. I would reckon most members of the public would see it as quite fair and would even expect that you pay more for a longer distance. And I'd imagine most passengers would also think it reasonable that if there are two vastly different ways to get to the same destination, you might pay a bit less for the slower route (because that's not providing such good service). On both of those measures, charging more to go via HS2 to Solihull would likely be seen as fair, and charging the same amount as a direct train would be seen as unfair.

The other side of it is, though, that people going via HS2 is probably better financially for the railway.
The chiltern line is likely to be more heavily loaded at the London end, for obvious reasons, so if people use the ample capacity on HS2 and then the quieter end of Chiltern to reach their destination, it could easily outright cost the railway less.

But this is something of a flaw in a distance based fare system.

(EDIT: First guess makes Marylebone-Moor Street kilometerage about 5km more than Euston-Curzon street)

 

[Trainbike46]

Ah yes, I see what you mean.
London-Solihull would have its kilometreage set by the classic line, even if hte dominant way to do the journey is via HS2.
Since you have to let people out of the barriers at Curzon Street with a Solihull ticket, it would form the functional London-Birmingham ticket .

Hhm, this is an interesting dilemmna.
We would certainly need HS2 and non HS2 tickets if we allow such aggressive doubling back.

A potential solution would be to offer two fares from Solihull (and similar stations) to London - Via HS2, priced at the same level as a HS2 Birmingham-London ticket, and Via Banbury, charged at the distance based fare for Solihull-London

The other side of it is, though, that people going via HS2 is probably better financially for the railway.
The chiltern line is likely to be more heavily loaded at the London end, for obvious reasons, so if people use the ample capacity on HS2 and then the quieter end of Chiltern to reach their destination, it could easily outright cost the railway less.

Exactly this.

 

[renegademaster]

Have the yield management algorithm work like airlines, where if a train is quieter than expected the price can go down as time goes on, rather than the current system of it only ever going up as time passes

 

[Bletchleyite]

A potential solution would be to offer two fares from Solihull (and similar stations) to London - Via HS2, priced at the same level as a HS2 Birmingham-London ticket, and Via Banbury, charged at the distance based fare for Solihull-London

That (aside from a Milton Keynes Stations group) is how I'd solve the Bletchley via MKC double back issue too. Normal fare doesn't allow the double back, one priced the same as to MKC does.

== Doublepost prevention - post automatically merged: 30 May 2025 ==

Have the yield management algorithm work like airlines, where if a train is quieter than expected the price can go down as time goes on, rather than the current system of it only ever going up as time passes

I did think LNER might have been looking at that when they started talking about dynamic pricing, but it turns out they just wanted to whack the fares up

 

[renegademaster]

out they just wanted to whack the fares up

It's not like the airlines do it out of charity

 

[miklcct]

Although distance-based pricing comes with it's own problem, if we don't blindly use the actual track distance, but assign a tariff number to each piece of track and runs a "distance based fares" model on the resulting graph, geography-based loopholes (split ticketing into the same ticket types priced by different companies) can be eliminated.

And to avoid "break of journeys" complications, long distance fares should approach fare = rate * distance as distance approach infinity, with a smooth curve starting from a flat fare for short distances.

distance * rate per km + base rate * e ^ (-distance * rate per km / base rate)

The above formula has an effect that the fare is flat at 0 km, and approaches true distanced-based fares at infinity distance as the second term approaches 0.

For example, if we set the base fare to be £1.00, with rate per km to be £0.20, the formula will be

0.2 * distance + e ^ (-distance * 0.2)

With this formula (rounded to the nearest 1p), the fares will be:
0 km: £1.00
1 km: £1.02
2 km: £1.07
3 km: £1.15
4 km: £1.25
5 km: £1.37
7 km: £1.65
10 km: £2.14
15 km: £3.05
20 km: £4.02
200 km: £40.00

Note that the above formula has the two (distance * rate per km) which is a monetary value and can be substituted by a common letter, which means that if we don't use strictly a distance-based fare, each piece of track can be assigned an incremental fare, and given a "base fare" the formula will take care of itself to decide how much of the "short distance fee" to be added into the incremental fare. With such a formula there can be no split ticketing possibilities for cheaper fares. Also, breaking long journeys midway overnight won't result in a higher fare as well due to how the formula works - If you travel from Edinburgh to York, then York to City Thameslink on another day, the total price will be exactly the same for an Edinburgh to City Thameslink ticket, but if you first travel from Edinburgh to City Thameslink, then from City Thameslink to London Blackfriars on another day, you will need to pay another base fare (because the distance from City Thameslink to London Blackfriars is close to 0).

Then, it is up to the pricing manager to decide the fare by assigning numbers to each section of track, or for advance ticket, each section of each service, and all flexible tickets will have an exact routing where the passenger cannot deviate from it (similar to other distance-based systems in the world). There will be no choice of routes on any tickets.

For peak and off-peak differences, the pricing manager can decide, for each section of the track, the peak and off-peak rate as well, and because the price difference is trivial (under most cases, 0, if the splitting point is not close to one end) by splitting, multiple tickets can be issued according to the exact station where the train is at the cut off time, and it should be an accreditation requirement that peak / off-peak split tickets be offered when appropriate.

For Advance tickets, there will be no longer quotas and fare buckets - the price setter simply set an incremental price between each section on each service using yield management software, limited by the applicable peak / off-peak rate on the section of the track travelled, and the fare will be just the sum of the sections offered plus any applicable leftover base fare for short distance journeys (if travelling on an IC train for a short journey).

There will be no point in prohibiting break of journeys even for Advance tickets because a longer distance Advance fare can't undercut a shorter distance Advance fare on the same train (assuming that the fare on each section can't be negative - it can be zero without problem - if SWR wants to set its advances to have the same fare from a long distance origin to Clapham Junction and to London Waterloo, it can set the "incremental fare" between Clapham and Waterloo on those trains to 0. It will result in Advance tickets being available on long distance trains even for a Clapham - Waterloo journey (unless blocked by a set down only restriction) but the base fare will still be charged according to the formula, so it won't result in free travel).

In addition, the rate can also be differentiated by the service category as well, that means an IC service can charge higher than a regional service on the same track, and brand-restricted tickets can be calculated and issued. For tracks with both types of services, a brand-restricted rate can be set on those tracks, while on tracks with only one kind of service, only one rate (all brands) will be set. So, for example, the track between Peterborough and Huntingdon can have the following rates:
- Any brand First Class Anytime (as IC trains pass through the station so a rate is required, even the trains don't stop at Huntingdon)
- Any brand Standard Anytime
- Non-Intercity First Class Anytime
- Non-Intercity Standard Anytime
- Non-Intercity First Class Off-Peak
- Non-Intercity Standard Off-Peak

While the track between Watford Junction and Watford North will only have one rate:
- Anytime (assuming that no peak times exist in that route)