Is Open Access abstractive or does it genuinely plug gaps?

Lumo West Coast also opens up additional options for everyone north of Stirling on the Highland Main Line. It’s an easy and pleasant station to change at.

Greenfaulds is also interesting, I think this could potentially be a winner as it has a large free park and ride carpark that has mostly been pretty empty since Croy carpark was massively extended.

I think some of them do plug a gap. Sunderland to London via the Durham Coast is one.

It's a shame some of the "legacy" operators didn't identify the paths themselves.....rather than moan about somebody else finding them.

On a similar vein, I'm interested to see what becomes of the LNWR plans to/from Manchester

HighlandStorm said:

Lumo West Coast also opens up additional options for everyone north of Stirling on the Highland Main Line. It’s an easy and pleasant station to change at.

Greenfaulds is also interesting, I think this could potentially be a winner as it has a large free park and ride carpark that has mostly been pretty empty since Croy carpark was massively extended.

Click to expand...

Especially since Croy is a lot more attractive to use as a park & ride compared to Greenfaulds if you want to travel to Glasgow. Much faster journey times

Regarding the question in the thread title - 100% abstractive. But that's what they're for. An OAO wouldn't start operations if the owners weren't sure it makes a profit, so naturally high-yield routes are chosen, where they can compete with the state operator.

Hull Trains is probably the only one where a gamble was taken. FirstGroup guessed that providing a few trains between Hull and London would grow the rail market in Hull, and keep the company in the green through offering cheap advances between London and Doncaster. In this case though, the important part wasn't necessarily filling the gap, it was highlighting the gap. If HT were to withdraw operations, I'm fairly sure that LNER would step in and offer more connections to Hull if the rolling stock was available. The risk - whether direct London-Doncaster-Hull trains were worth it - was taken by the OAO, but now that the benefits of it have been proven, any operator could provide the service decently, including GBR.

Windy routes to give smaller towns a direct London service, just to prove that a route isn't abstractive, aren't suitable in a high-utilisation dense network of railways. Nobody can really deny that Grand Central & Lumo West Coast didn't begin their plans by looking at how to give Hartlepool/Halifax/Stirling a London service. They saw high demand and high fares between London and Preston, York, Doncaster, Wakefield etc, and attached something random onto the end with no consideration of how well-used that section would be, or how much capacity it's taking up on local railways. Case in point - very few people will actually make the whole 6 to 7 hour trek from Stirling or the eastern Glasgow suburbs into London on Lumo, they're more likely to change at Edinburgh to a Lumo East Coast service in fact.

So it's quite clear that if you prioritise network integration - meaning reliable, frequent services and thought-out connectivity between local/regional and long-distance rail, OAOs don't really work. The only way I can see it working is if the ORR selects the paths (e.g. an hourly/2-hourly slot between London and Edinburgh) calling at specified stations, so no skipping York while serving Morpeth shenanigans, and then allows private operators to bid to run the service at their own financial risk. This is the approach Poland is taking as it plans to implement both a national integrated timetable and an open rail market, and I think it'll work quite well.

However, fares integration is a lot trickier. Most countries in Europe generally get to choose between two options, neither of which are particularly good:
1. Predictable fixed fares, usually flexible. This option makes connectivity a lot easier, particularly on a journey involving an intercity train with a regional connection at either end. However fares remain high which limits ridership growth.
2. Dynamic pricing on either all of the network or parts of the network (i.e. just intercity), resulting in some great bargains but also some unattractive peak time fares. Also means that you have very few rights (e.g. if a train is cancelled) if you book a journey with numerous connections with multiple operators.
There is a third option, a Deutschlandticket/Klimaticket equivalent, but the amount of subsidy and built-in network capacity required means that it's very much a long-term goal instead of an implementable strategy at the moment.

Given the overall lack of ambition around rail growth at the moment, with minimal long-term planning and pricing strategies that signal that they'd rather you take the car unless you travel in first class (e.g. the ECML 'simplification' scheme), I'd say that at the moment OAOs aren't doing that much harm, and the short-term gain of some well-priced fares on a few routes outweighs the issues. I doubt it'll get much better with the next government as there really isn't that much public interest in rail at the moment, following the EWR and HS2 fiascos.
However, if this changes in the future, then I can certainly see a world where unregulated open-access is done away with. You could practically replace all OAOs by:
- scrapping HEX or transferring it to GBR
- GBR-SNCF-SNCB agreement to run Chunnel services separately to Eurostar
- running 2-hourly London to Hull LNER
- running a few extra Avanti services to Preston/Glasgow
- maintaining frequent, reliable connections between mainline services at Leeds/Doncaster/Middlesbrough and local services to Teesside/Sunderland/West Yorkshire

However this doesn't solve the fares problem, as the main reason why OAOs are popular is the competitive pricing. I can't think of a solution to that which can easily and affordably be implemented, so I guess the UK will be stuck with GBR's half-done network integration plans on most of the network alongside a chaotic mess of various operators on the mainlines for the forseeable future.

One plus for the UK is that even if we do go down the path of letting FirstGroup any operator run on any line, the lack of High-Speed Rail means that no government can ever really ignore regional connectivity in favour of profitable intercity operation and market liberalisation on the same level as Spanish/French governments have done over the years, as all of our important mainlines remain mixed-use. That's not to say that we shouldn't be building High-Speed Rail, but the key benefits of it (mainly the capacity release) can only be properly utilised if the high-speed line is designed timetable-first, with the discussion of who runs the trains coming second. If you have a high-speed line with trains every 15-30 minutes between major interchanges, then it doesn't really matter who runs the trains as long as your ticket is accepted on all of them.

I have to admit, there is a part of me that hopes the ORR shocks everyone and justifys authorising another OAO access agreement... almost as a swan song... a final defiant act


To be clear, I do not mean grant access willy nilly. I mean authorise where it can reasonable be done so

Farigiraf said:

Regarding the question in the thread title - 100% abstractive. But that's what they're for. An OAO wouldn't start operations if the owners weren't sure it makes a profit, so naturally high-yield routes are chosen, where they can compete with the state operator.

Click to expand...

That presumes that the OAO is running a service instead of the state operator. The moment that the state operator would not operate the same service, then the OAO ceases to be wholly abstractive, as a matter of simple logic.

The argument about capacity is separate, and also falsifiable the moment that you rely on the state operator running the service provided by the OAO - all you are doing is changing who would run the service

35B said:

That presumes that the OAO is running a service instead of the state operator. The moment that the state operator would not operate the same service, then the OAO ceases to be wholly abstractive, as a matter of simple logic.

Click to expand...

The main reason that OAOs like Lumo and Grand Central are established is in order to run services along a profitable intercity route, such as London-Preston, London-Newcastle or London-York.
The fact that they have to add on circuitous routes with low loadings on to the end is a result of how the ORR abstraction test works. But in practice, while passengers from Sunderland or Stirling are certainly happy to have a direct, cheap London service, they are not the priority at all really.

It really depends how you want to define 'not abstractive' - if you define it as any new direct train connection, then presumably my new OAO startup from London to Smethwick Rolfe Street (also stopping at MK, Coventry and Birmingham) isn't abstractive. If you actually want to achieve what the ORR intended, which is growing the rail market somewhere by upping the service to a place where the state operator hasn't tried yet, then Hull Trains is the only one that really counts here - but again, that's succeeded not because it's run by a friendly company with a strong local image, it's because previously there was only one daily connection between London and Hull. Plus because they offer some cheap London-Doncaster advances to people who would otherwise use LNER (or drive).

Along most of the unusual routes that OAOs run on, such as to Sunderland, Hartlepool or Pontefract, you'd grow usage far more by providing a frequent regional service (with good LNER connections at Middlesbrough/Newcastle/Doncaster) and cheap fares. The state operator can't offer cheap fares without massive subsidy though, nor is it easy to increase train frequencies at short notice, while an OAO can devote its time to sorting this out.
At the end of the day, for cheap fares, it's still the passenger who pays. The difference is how that's done - whether it's through the ticket price, or through the tax bill, or through accepting a lower quality Ouigo-style service offering. Each has its pros and cons, but neither option prevents something other than a revenue-first OAO from achieving it.

35B said:

The argument about capacity is separate, and also falsifiable the moment that you rely on the state operator running the service provided by the OAO - all you are doing is changing who would run the service

Click to expand...

The only routes which probably could stay even if you were designing an integrated network focusing on connectivity over ticket revenue are the Heathrow Express (nothing better to do with the paths) and the London-Hull route (only service between Selby and Doncaster). Most of the other unique bits - Stirling via Whifflet, Durham Coast-York and Doncaster-Bradford, make more sense as local connections. No reason why you have to run these as London services - it just results in less seats on the mainline, and less seats means that you have to charge more per seat to stay revenue positive.

Lumo and GC running 5-cars on the ECML is a great example of this - that's a few hundred seats each hour that are going to waste, no amount of cheap pricing to attract car drivers or airline passengers will be able to fill those seats, as they don't exist. GC won't pay for platform extensions at Pontefract Monkhill or Low Moor, so I don't see a timeline where those paths on the ECML will be used to their full potential.

Out of the three cornerstones of good passenger-facing rail operations - network integration, fares integration and low fares, the current model means that you can only really pick 2 of 3 at most, though some countries only manage 1 of 3. To achieve all three, alongside having some really excellent planning, someone needs to cover the cost of the low fares, as Swiss-style frequency and flexibility costs money to run. It'll always be the passenger in some way, but whether that's through the train fare, the tax bill, or lowered onboard standards (i.e. Lumo-style barebones operations) is a political decision.

Farigiraf said:

The main reason that OAOs like Lumo and Grand Central are established is in order to run services along a profitable intercity route, such as London-Preston, London-Newcastle or London-York.
The fact that they have to add on circuitous routes with low loadings on to the end is a result of how the ORR abstraction test works. But in practice, while passengers from Sunderland or Stirling are certainly happy to have a direct, cheap London service, they are not the priority at all really.

It really depends how you want to define 'not abstractive' - if you define it as any new direct train connection, then presumably my new OAO startup from London to Smethwick Rolfe Street (also stopping at MK, Coventry and Birmingham) isn't abstractive. If you actually want to achieve what the ORR intended, which is growing the rail market somewhere by upping the service to a place where the state operator hasn't tried yet, then Hull Trains is the only one that really counts here - but again, that's succeeded not because it's run by a friendly company with a strong local image, it's because previously there was only one daily connection between London and Hull. Plus because they offer some cheap London-Doncaster advances to people who would otherwise use LNER (or drive).

Click to expand...

And there's whatever go-op wanted to do.
Which if they got their £2,000,000 or however much NR wanted to make sure some car drivers don't do stupid things I'm sure they'd be well on their way to being an embraced part of our rail network (until the money runs out).
Their goals were good, just unrealistic.

Farigiraf said:

The main reason that OAOs like Lumo and Grand Central are established is in order to run services along a profitable intercity route, such as London-Preston, London-Newcastle or London-York.
The fact that they have to add on circuitous routes with low loadings on to the end is a result of how the ORR abstraction test works. But in practice, while passengers from Sunderland or Stirling are certainly happy to have a direct, cheap London service, they are not the priority at all really.

It really depends how you want to define 'not abstractive' - if you define it as any new direct train connection, then presumably my new OAO startup from London to Smethwick Rolfe Street (also stopping at MK, Coventry and Birmingham) isn't abstractive. If you actually want to achieve what the ORR intended, which is growing the rail market somewhere by upping the service to a place where the state operator hasn't tried yet, then Hull Trains is the only one that really counts here - but again, that's succeeded not because it's run by a friendly company with a strong local image, it's because previously there was only one daily connection between London and Hull. Plus because they offer some cheap London-Doncaster advances to people who would otherwise use LNER (or drive).

Along most of the unusual routes that OAOs run on, such as to Sunderland, Hartlepool or Pontefract, you'd grow usage far more by providing a frequent regional service (with good LNER connections at Middlesbrough/Newcastle/Doncaster) and cheap fares. The state operator can't offer cheap fares without massive subsidy though, nor is it easy to increase train frequencies at short notice, while an OAO can devote its time to sorting this out.
At the end of the day, for cheap fares, it's still the passenger who pays. The difference is how that's done - whether it's through the ticket price, or through the tax bill, or through accepting a lower quality service offering. Each has its pros and cons, but neither option prevents something other than a revenue-first OAO from achieving it.


The only routes which probably could stay even if you were designing an integrated network focusing on connectivity over ticket revenue are the Heathrow Express (nothing better to do with the paths) and the London-Hull route (only service between Selby and Doncaster). Most of the other unique bits - Stirling via Whifflet, Durham Coast-York and Doncaster-Bradford, make more sense as local connections. No reason why you have to run these as London services - it just results in less seats on the mainline, and less seats means that you have to charge more per seat to stay revenue positive.

Lumo and GC running 5-cars on the ECML is a great example of this - that's a few hundred seats each hour that are going to waste, no amount of cheap pricing to attract car drivers or airline passengers will be able to fill those seats, as they don't exist. GC won't pay for platform extensions at Pontefract Monkhill or Low Moor, so I don't see a timeline where those paths on the ECML will be used to their full potential.

Out of the three cornerstones of good passenger-facing rail operations - network integration, fares integration and low fares, the current model means that you can only really pick 2 of 3 at most, though some countries only manage 1 of 3. To achieve all three, alongside having some really excellent planning, someone needs to cover the cost of the low fares. It'll always be the passenger in some way, but whether that's through the train fare, the tax bill, or lowered onboard standards (i.e. Lumo-style barebones operations) is a political decision.

Click to expand...

I'm not disputing - or criticising - operators from seeking to make a profit. The "not primarily abstractive" test is set at 30%, meaning that at least 30% of the revenue earned must be new to the system. Those services have grown a market that is not just about creaming off traffic.

Now, there may be good reasons to do things differently and your suggestions may be better. But that doesn't mean that Open Access is in and of itself abstractive, and the case you make is undermined by that misuse of language.

Living on the ECML somewhere served by HT, what I also observe is that they are an integrated part of the timetable (to my financial disadvantage when the Mallard timetable was introduced many moons ago), and that they offer capacity over and above that which DfT are willing to allow LNER to offer.

That reality, however much it sticks in the craw in other ways, suggests that many of the complains about OAOs are misconceived, and reliant on a counter factual scenario that simply doesn't exist.

35B said:

The original was correct. The failure of the Virgin franchise was because DfT had got greedy, and ECML passengers weren’t willing to be milked in the way they needed to be to meet DfT’s expectations.

Click to expand...

Who said anything about Virgin?

Farigiraf said:

The only routes which probably could stay even if you were designing an integrated network focusing on connectivity over ticket revenue are the Heathrow Express (nothing better to do with the paths) and the London-Hull route (only service between Selby and Doncaster).

Click to expand...

Hull Trains are far from the only operator running between Selby and Doncaster. Both Northern and LNER operate over this section of line.

In fact that is an interesting point. We always hear about the costs of withdrawing services over a section of track, but do the same rules apply to Open Access Operators? As far as I can think the only sections of line serviced only by OAOs are:

York Holgate Junction - York Yard South - York Yard North - Skelton Junction - Lumo East Coast
Shaftholme Junction - Knottingley West Junction - Grand Central

What would happen officially if one of these services ceased?

pokemonsuper9 said:

And there's whatever go-op wanted to do.
Which if they got their £2,000,000 or however much NR wanted to make sure some car drivers don't do stupid things I'm sure they'd be well on their way to being an embraced part of our rail network (until the money runs out).
Their goals were good, just unrealistic.

Click to expand...

They were effectively trying to mock Hull Trains' strategy - i.e. highlighting a part of the network that's not utilised to the extent it could be. In that respect, they were the most faithful to the goal of not providing abstractive services. However, revenue from local fares between Taunton, Westbury and Swindon will never match that of the London-Doncaster-Hull route, so it was never going to work as an unsubsidised route.
What would've been a better use of all that effort is to lobby the DfT to fund such a service for GWR to operate, on a trial basis. Then I'm sure it would've proved popular, but as Go-Op never managed to actually start running, they could never prove any demand.

To echo my previous post, if you want to offer an entirely new rail connection (or improve the offering along an existing route), someone does need to pay for it. Crowdfunding won't pay for the full total, as we've seen, so it's either the DfT (through tax money) or an OAO (through profits earned from abstracting revenue along an intercity corridor) who pays for it. Potentially London - Reading - Swindon - Westbury - Taunton may have worked, if there were spare paths along the GWML, but this does basically show that without also serving London you couldn't run a profitable service to an unprofitable destination. You couldn't crowdfund a new road (and its subsequent maintenance) either, the state or the private sector has to pay for it.

35B said:

Those services have grown a market that is not just about creaming off traffic.

Click to expand...

Would you say that growing that market is worth it while 70% of the OAO revenue is abstracted though? That 70% partly disappears (into the private sector) and partly funds that 30% of new growth.

While some of that growth is from offering new connections, it's minimal compared to how much a reliable connecting service would be used. One-seat train rides are more valued in the UK than elsewhere, however as I explained in a previous post, those one-seat rides are often at the expense of capacity elsewhere on the network, so it's far from the potential growth that could be achieved in this sector.

The other, more significant area where revenue isn't necessarily abstracted, is where cheap fares encourage drivers and flyers to switch to rail. I'm certainly in favour of that, but as we've seen in in countries embracing open access competition - primarily France, Italy and Spain, attempting to maximise modal shift just to intercity rail (rather than to all public transport in general) results in airline-style rail operations which means that other parts of the wider transport network are neglected (or left to the broke regions to sort out). It's why they keep building high-speed rail parkway stations in the middle of nowhere - onward connectivity is not the OAO's probem. This isn't the most effective way method of driving modal shift though since people still end up driving to/from the intercity rail station.

Looking at two pairs of countries with different approaches to rail: Switzerland & the Netherlands, compared to France & Italy. The first two have good network integration, high frequencies, but high fares, while the last two have managed to shift lots of long-distance traffic to rail, while paying less attention to regional rail. Potentially more people on average make long-distance journeys by car in the Netherlands or Switzerland due to high fares, but the actual rate of car ownership and car usage is notably lower than France or Italy as people don't take the car for shorter distance journeys as much.

So if you have to pick 2 out of these 3: network integration, fares integration, cheap fares, the combination that drives the most modal shift is network integration + fares integration. If on top of that, the state subsidised cheap fares (more than they already do), there would be world-leading levels of modal shift, but that's a political decision which hasn't been made. In comparison, if you head down the road of having cheap fares along key corridors, along with a couple of daily connections to various places (in order to pass an abstraction test), you won't get nearly as many people out of their cars, and trying to retroactively establish swiss-style connectivity at that point is considerably more complex and expensive, as you're effectively building up a regional network from scratch.

35B said:

But that doesn't mean that Open Access is in and of itself abstractive, and the case you make is undermined by that misuse of language.

Click to expand...

Open Access operators only stay revenue positive if they do abstract revenue. They're able to offer new connections and cheap fares by robbing Peter (the state operator) to pay Paul (potential rail travellers who currently drive).
Go-Op demonstrates that it doesn't work otherwise.

35B said:

they offer capacity over and above that which DfT are willing to allow LNER to offer.

Click to expand...

That's the main problem really. A lot of my points are on the basis that the state operator could achieve the same goals more successfully and efficiently, but will it actually? Not with current DfT thinking.
Accepting that fact, then I guess you could say that the service offering by the likes of Grand Central and Hull Trains is contributing to rail growth that wouldn't otherwise be happening. The Tees Valley Metro for example didn't end up happening even though it would've reduced the usefulness of Grand Central's eastern route. I'm trying to explore more optimistic alternatives in my postings.

IanXC said:

Hull Trains are far from the only operator running between Selby and Doncaster. Both Northern and LNER operate over this section of line.

Click to expand...

Once a day each, at 7am and 8pm. It really doesn't count. Someone going between London and Selby or Hull, without the presence of HT, would either have to go via York/Leeds (indirect) or Goole. Had there already been a regular service on this section of line before Hull Trains started running, they most likely wouldn't have passed the abstraction test.

IanXC said:

Hull Trains are far from the only operator running between Selby and Doncaster. Both Northern and LNER operate over this section of line.

Click to expand...

If you remove HT from the equation then the only trains from Selby to Doncaster are at 0737 (LNER) and 2059 (NT), and in the other direction from Doncaster at 1953 (LNER) and 2024 (NT). That is not a viable level of service for passengers. It would certainly stop me from getting the train to work if those were the only options!

IanXC said:

In fact that is an interesting point. We always hear about the costs of withdrawing services over a section of track, but do the same rules apply to Open Access Operators? As far as I can think the only sections of line serviced only by OAOs are:

York Holgate Junction - York Yard South - York Yard North - Skelton Junction - Lumo East Coast
Shaftholme Junction - Knottingley West Junction - Grand Central

What would happen officially if one of these services ceased?

Click to expand...

It would make no difference. The cost of closing a line relates to where there is a requirement for a passenger service to be run. Where a service is run by an OAO, there is no such order in place, because the OAO runs the service with permission rather than by requirement, and so there is no legal impediment or cost to removing it. (The political cost of cancelling such a service, on the other hand, may be significant!)

Farigiraf said:

Regarding the question in the thread title - 100% abstractive. But that's what they're for. An OAO wouldn't start operations if the owners weren't sure it makes a profit, so naturally high-yield routes are chosen, where they can compete with the state operator.

Hull Trains is probably the only one where a gamble was taken. FirstGroup guessed that providing a few trains between Hull and London would grow the rail market in Hull, and keep the company in the green through offering cheap advances between London and Doncaster.

Click to expand...

Not quite.

FG had naff all to do with the start of Hull Trains. Hull Trains, as a concept, was the brainchild of John Nelson and Mike Jones (under their Renaissance Trains umbrella) who approached GB Railways, the then owners of Anglia, for both financial backing and an operator who could take it from idea to reality.

The GB Railways directors asked the MD of Anglia, Tim Clarke, to find a resource to do just that. So he tasked his newly appointed Track Access (and New Services) Manager to get it done. The work started in April 1999 but it helped that he was an ex GN man so he knew the ECML well.

To start with the class 170 units had to be worked hard on the ECML on their three return trips and not stop too often, to keep their paths. Doncaster was an obvious stop but not primarily for competitive reasons. That came later with the class 222 units. It was because people from stations to Hull wanted to go to and from there but not in a 142 or 153! That left only one station that realistically could be regularly served and Grantham was chosen, again it had got good connectional value and the yields were better than Peterborough, which would have sunk the business because of the northbound crowding off effect. Newark was too abstractive.

The first full railway year that Hull Trains was in operation was 2001/2 and the HT London revenue from Hull alone was £2357k. It took only £264k from London-Doncaster. That was out of an all operator London-Doncaster total of £11966k and the HT “take” was less than the total all operator increase for the Doncaster flow that year.

Just to give a comparison of annual all operator London Journeys “before” (1998/9) and “after” (2005/6), Hull - 155044/251788, Brough 23869/51172, Grimsby - 47760/51980 and Lincoln 68411/72823. Those were amongst the figures submitted to the High Court in the 2006 GNER Court Case. As a result, GNER had to withdraw their abstraction argument on the first night of that case.

HT isn’t, even now, always the cheapest option. The business model has always been about a target yield per seat, value per train and the value of a through journey is in the price. Of course there are headline figures to entice the punter but it isn’t about cannibalising the average yield in an effort to fill a train.

The business model and length of train have to be matched to enable a return to be made. It isn’t a model of providing maximum capacity, stopping everywhere significant or even undercutting everyone else at all times. It’s far more targeted - you can have that focus when you operate only a few trains a day.

Hull is now a mature market so GBR could take it over and even do different things with it but the temptation to run even more trains to Leeds and York northwards at the expense of Hull would always be there. GBR budgets won’t really take into account any targeted economic benefit from a regular service between London and Hull.

In answer to thethread title.

Is Open Access abstractive or does it genuinely plug gaps?

Click to expand...

Yes.

OAO will "take" passengers from the main operator.

For example Dave takes the 08:04 to get him to work for 9am. A new OAO starts a new service at 07:46 and due to various factors Dave uses the new service and so is a "loss" to the main operator.

However, that doesn't always mean that the operator is financially worse off.

For example having seen the marketing for the new OAO both Claire and Sam look to travel along the route of the new service, and book onto the train which best suits their needs - which happens to be the one run by the main operator.

In an ideal world there would be no gaps for OAO to implement new services. As the main operators would test a few new stations for a few years with marketing andthe like to try and grow a market, sticking with those which work and scaling back from those which didn't.

The issue is that the DfT can't (or more probably aren't allowed to) do this, in such a world then OAO are a useful tool in increasing the number of passengers.

HSTEd said:

Given the low access charges that open access operators are charged, far below the fully burdened costs of those paths, I personally can't see how they can be considered anything other than abstractive.

Track access charges only cover a fraction of Network Rail's costs after all (last I checked it was ~25%).

Click to expand...

Which is a fair point, up to a point.

However, the first question is which NR costs do the OAO only cover 25% of?

From the Rail industry finance report:

Network Rail

Operating costs: £3.3bn
Maintenance: £2.5bn
Renewals: £3.7bn
Financing costs: £2.5bn
Other costs: £0.4bn
Total: £12.4bn

The two values which are key are the Maintenance (i.e. direct harm done by each train) costs which are covered by the track access charges and the Renewals (i.e. long term replacement of life expired equipment).

The 25% figure quoted by rail unions is slightly skewed to make the OAO look worse than they are, but it's still something worth being aware of.

OAO pay all their track access charges (Maintenance), but (as I understand it, but maybe I've miss undersood something) effectively 25p in the £1 for the Renewals from their fixed payment (a per train cost).

Therefore, if every train was OAO the table would be:

Maintenance: £2.5bn - £2.5bn received (100%)
Renewals: £3.7bn - £0.925bn received (25%)
Total £6.2bn - £3.425 received (55%)

Given that they do pay 100% of their track access charges, so presumably do cover all their direct harm (maintenance) costs, I'm not sure how else to understand the 25% figure other than being their share of the Renewals budget.

Of course, the fact that not all OAO even pay this fee (in fact currently only one) muddies the waters in something which isn't a clear binary case to start with.

Sorcerer said:

I'm in two minds about open access. On the one hand it can plug gaps such as London-Hull or London-Stirling, and the competition can reduce fares and increase ridership like we saw in Italy between Trenitalia and Italo as well as in Spain between RENFE, Ouigo and Iryo, and in France between SNCF and Trenitalia. Even Eurostar might well end up with lower fares with Virgin entering the picture.

But on the other hand, railways by their nature are a bit more constrained than airlines and road transport, and competition will only really be a select few cherry-picked routes (the notable examples I cited are all high-yield high-speed routes like Paris-Lyon or Madrid-Barcelona) because obviously when you're accepting all the revenue risk, you need to maximise your return on investment.

Ideologically I think rail should be a public service first, but pragmatically open access can work for the benefit of the passenger. If the space is available then I suppose it's fine really, but you could still arguably claim open access is abstractive if it's competing with a state-operator like Lumo and LNER on high-yield routes like London-Edinburgh. The WCML equivalent would be London-Manchester.

Click to expand...

I'm minded to agree, OAO aren't a perfect situation, but then they are a useful tool in a rail model which is constrained in such a way which means that is unable to deliver even a near perfect perfect situation.

Obviously, there will be those from either side who will upsell the advantages or disadvantages of OAO's, the bottom line is that they exist, they do create services which often wouldn't exist, they do take passengers who were previously using TOC services, they do create positive headlines for the rail industry and it appears that they mostly grow the market overall. There are other things which they do which are "bad*", for example barring access to final salary pensions schemes for new entrants.

(* again whilst not ideal, for anyone moving their from a TOC this is going to be a calculation they take on a personal level, it maybe they've maxed the value they can get and so move for the last few years of working to create a second pension pot which they can use to further increase their income within retirement, it maybe that they can live somewhere cheaper and not need to own a second car in their household and so can use that spare money to pay down their mortgage and be mortgage free sooner, or many other things).

On balance, I would suggest that it's probably fair to say that "they are a necessary evil" is something many could accept as a fair assessment that many could accept as a view a significant majority would include as part of their position (i.e. anything more positive than this would include that, and it's only those with views of OAO's which are worse, for example from "I can't quite reach they are a necessary evil" through to "OAO are so foul they need to be sent straight to the same place as [insert something so bad it needs to be removed not only from this universe but throughout the multiverse]"

The Ham said:

However, the first question is which NR costs do the OAO only cover 25% of?

From the Rail industry finance report:

Network Rail

Operating costs: £3.3bn
Maintenance: £2.5bn
Renewals: £3.7bn
Financing costs: £2.5bn
Other costs: £0.4bn
Total: £12.4bn

The two values which are key are the Maintenance (i.e. direct harm done by each train) costs which are covered by the track access charges and the Renewals (i.e. long term replacement of life expired equipment).

The 25% figure quoted by rail unions is slightly skewed to make the OAO look worse than they are, but it's still something worth being aware of.

OAO pay all their track access charges (Maintenance), but (as I understand it, but maybe I've miss undersood something) effectively 25p in the £1 for the Renewals from their fixed payment (a per train cost).

Therefore, if every train was OAO the table would be:

Maintenance: £2.5bn - £2.5bn received (100%)
Renewals: £3.7bn - £0.925bn received (25%)
Total £6.2bn - £3.425 received (55%)

Given that they do pay 100% of their track access charges, so presumably do cover all their direct harm (maintenance) costs, I'm not sure how else to understand the 25% figure other than being their share of the Renewals budget.

Of course, the fact that not all OAO even pay this fee (in fact currently only one) muddies the waters in something which isn't a clear binary case to start with.

Click to expand...

All operators (franchised, OAO and freight) pay only 10% of NR’s net infrastructure costs.

The system works like this.

The Governments decide how much they want to grant fund NR. That is the vast majority of NR’s income. Then the ORR works out the TOC/FOC charges, based on short run marginal costs only with mark ups paid by some OAO, all done in line with current law restrictions.

This leaves a significant amount still to be funded. So what the Governments do is effectively another form of Grant. This is “paid” via the franchise TOCs and then goes to NR and is called Fixed Track Access but it is really just a laundered grant. It’s a residual funding mechanism which varies from year to year, it’s not a specific infrastructure cost.

NR is currently working out how charging is going to work under GBR. The legal position changes and there is a general fear amongst the OA community that GBR may be legally obliged (by the Subsidy Control Act) to charge long run incremental costs (LRIC) as part of the base charge.

To give you the scale of the issue, NR identified in 2018 that 44% of all fixed costs were in fact LRIC, so the potential is for OAO to be charged annually several millions extra, per operation, from 2029 or to have to hand over their operations on 01/04/29 (the start of GBR FP1) to avoid running at a considerable loss. OAO strongly suspect some people at the DfT worked all this out long ago and will “bounce” their ministers at a suitable time.

The Ham said:

Given that they do pay 100% of their track access charges, so presumably do cover all their direct harm (maintenance) costs, I'm not sure how else to understand the 25% figure other than being their share of the Renewals budget.

Of course, the fact that not all OAO even pay this fee (in fact currently only one) muddies the waters in something which isn't a clear binary case to start with.

Click to expand...

To add to what's already posted, when looking at the concrete numbers it feels vanishingly unlikely to be the case. Hull Trains is likely similar to Grand Central but wisely haven't put the granularity in their accounts.

Lumo for year ending March 2025 would have paid £7.2m excluding traction electricity to Network Rail (after their infrastructure cost charge discount ends - actually £5m with discount). Of that £4.4m is the infrastructure cost charge that only Lumo pays which the other operators don't. However, Network Rail paid back £4.9m in restriction of use claims for engineering works. That makes £2.4m net for an operator who took £53m in passenger revenue not even 5% of that figure.

The ongoing East Coast Digital Programme has led to grants and other funding to open access operators, potentially being up to £2.5m to Lumo that year - it's lumped with some intra-group transfers of unspecified volume. Other OAOs received millions so it could be predominately that.
While this won't always happen and is investment funding, the presence of OAO makes this kind of investment more expensive. Overall Network Rail sent money to Lumo not the other way around for the year. (The infrastructure charge discount being larger than any uncertainty in Network Rail funding)

Grand Central don't pay the infrastructure cost charge and that makes the deal even better. For their last published calendar year 2024, their £2.3m in access charges was counteracted by £1.7m in schedule 4/similar claim payments. That's £650k for a business that took £57m in passenger revenue. Schedule 8 payments of £1.8m meant Network Rail paid Grand Central on an operating basis. (2023 was better Network Rail received £762k an absolutely tiny proportion of revenue). In both years there was East Coast Digital Programme funding of £2.5m 2024 and £2.0m 2023 for Grand Central.

When a transformation programme means taxpayer funded organisations are funding private organisations to undercut the taxpayer backed operators it's hard to see it as anything other than abstractive. Even after it ends they are contributing very little to infrastructure funding and that's before considering the revenue abstraction on the taxpayer backed operators.

But the OAO are not abstracting from the state operators because the Government pays nearly all of the fixed costs or investment in the network. It’s a false argument because the public operators are not paying these either.

The system was set up to enable modal shift so that is why all operators pay short run marginal costs. Latterly, the increase of OAO has forced the ORR to introduce the ICC as a mark up but that charge is subject to the legal “afford to pay” constraint.

Schedule 4 and 8 are liquidated damages regimes for lack of access and disruption. They need to be balanced with the loss of revenue that lack of access and disruption has caused. In the case of Schedule 4 there is also the Access Charge Supplement (payable by the TOC) to be added into the equation.

What has changed over the last couple of decades is the increasing unwillingness of the DfT to tolerate the current way of only seeing short run marginal costs being charged. They want either OAO to pay much more or for them to disappear. NR (for GBR) hasn’t quite got the message yet but the emerging legal position may well force them down that road.

The perversity of the Government position is best explained by the fact the Treasury likes OAO because they get the benefits directly and not through the DfT, who just see the disbenefit in their budget either immediately or in the future through the loss of future revenue opportunities.

What the Subsidy Control Act does, in conjunction with GBR taking over the charging (and internal attribution) system for infrastructure costs, is to constrain cross subsidy between public and private. There has to be clear Government policy (such as modal shift or focused economic policy) to enable any cross subsidy to happen and, so far, they have given no such indication they will go down that road for passenger TOCs. That could effectively stuff OA as from 01/04/29.

The value provided by OAOs of taking on risky new routes to find new markets wouldn’t be needed if the public operator is given the operational autonomy it deserves. the public operator could then iterate on new services a hell of a lot faster than the whole open access approval dance.

== Doublepost prevention - post automatically merged: 14 Jun 2026 ==

OAOs serve to optimise their function on too narrow a scope (individual routes) to be more valuable overall. We shouldn’t be optimising operations on the railway in isolation - instead optimising for the whole of the transport network. I don’t see how OAOs’ incentives can be aligned to fit with this.

== Doublepost prevention - post automatically merged: 14 Jun 2026 ==

If we had a national clockface timetable with simple, frequent intercity services then there wouldn’t be ‘gaps’ in the network - because interchanges would be quick and reliable.

Admittedly, this is an ambitious vision from where we’re at currently, but it will be doable once we build a segregated intercity high speed network across the country.

squingo44 said:

If we had a national clockface timetable with simple, frequent intercity services then there wouldn’t be ‘gaps’ in the network - because interchanges would be quick and reliable.

Admittedly, this is an ambitious vision from where we’re at currently, but it will be doable once we build a segregated intercity high speed network across the country.

Click to expand...

Unless this national clockface timetable had no white space on the graph, someone will still bid in to it.

Are there any examples of open-access operators being particularly abstractive from each other? From a quick look without knowing the area well, driving to Selby for Hull Trains looks quicker than a direct Grand Central train for parts of West Yorkshire to London (I know Bradford services started after Hull Trains), and a time-conscious traveller somewhat sensitive to price would go for Lumo over Grand Central's Sunderland route.

Stephen42 said:

To add to what's already posted, when looking at the concrete numbers it feels vanishingly unlikely to be the case. Hull Trains is likely similar to Grand Central but wisely haven't put the granularity in their accounts.

Lumo for year ending March 2025 would have paid £7.2m excluding traction electricity to Network Rail (after their infrastructure cost charge discount ends - actually £5m with discount). Of that £4.4m is the infrastructure cost charge that only Lumo pays which the other operators don't. However, Network Rail paid back £4.9m in restriction of use claims for engineering works. That makes £2.4m net for an operator who took £53m in passenger revenue not even 5% of that figure.

The ongoing East Coast Digital Programme has led to grants and other funding to open access operators, compromising £2.5m to Lumo that year. While this won't always happen and is investment funding, the presence of OAO makes this kind of investment more expensive. This meant overall Network Rail overall sent money to Lumo not the other way around.

Grand Central don't pay the infrastructure cost charge and that makes the deal even better. For their last published calendar year 2024, their £2.3m in access charges was counteracted by £1.7m in schedule 4/similar claim payments. That's £650k for a business that took £57m in passenger revenue. Schedule 8 payments of £1.8m meant Network Rail paid Grand Central on an operating basis. (2023 was better Network Rail received £762k an absolutely tiny proportion of revenue). In both years there was East Coast Digital Programme funding of £2.5m 2024 and £2.0m 2023 for Grand Central.

When a transformation programme means taxpayer funded organisations are funding private organisations to undercut the taxpayer backed operators it's hard to see it as anything other than abstractive. Even after it ends they are contributing very little to infrastructure funding and that's before considering the revenue abstraction on the taxpayer backed operators.

Click to expand...

Those payments are also made to TOC, as others have pointed out due to then not being able to run their services.

I suspect that for NR it's better value for money to pay those rates than only be able to access the network when there's no trains.

Is there a data source for those numbers as a quick search for "£2.5 million to Lumo from Network Rail" doesn't appear to show any relevant information?

As I said, the situation isn't ideal, and without GBR having a way to innovate to fill the role of OAO's it's fairly reasonable for a significant percentage of people to conclude that they are at least necessary evil.

There's always going to be those so politically motivated that they conclude that OAO don't even meet that bar. Although, likewise there's those who will be at the other end on the spectrum on this argument for whom OAO can do no wrong and so would be unhappy about agreeing to my phasing as it implies that they aren't as saintly as they believe.

JonathanH said:

True, but by the time IC withdrew, the services had reduced to just the 'business' trains.

Click to expand...

To clarify my point: some years prior to the demise of ICEC, by when the likes of Hull were served once daily as you say (and in most cases in marginal time), IC had operated 3-4 daily trains to each (including Middlesbrough though not Sunderland - though a memory says they were cancelled before they started, anyone confirm? I was referring to the cutbacks from that level and I think it's reasonable to say that OA followed in IC's footsteps.

The Ham said:

Those payments are also made to TOC, as others have pointed out due to then not being able to run their services.

I suspect that for NR it's better value for money to pay those rates than only be able to access the network when there's no trains.

Click to expand...

I'm not saying the TOCs necessarily cover their cost to Network Rail either (they don't). It's that OAOs don't changes the net public subsidy for the railways by increasing their profit. For TOCs it just changes what entity gets listed as requiring public support.

The Ham said:

Is there a data source for those numbers as a quick search for "£2.5 million to Lumo from Network Rail" doesn't appear to show any relevant information?

Click to expand...

All the numbers in my previous post are from the OAOs filings at Company House. Lumo is registered as East Coast Trains Limited and the latest accounts are here. Page 22 has the detail:

Off-charging of operating costs of £2,201k (2024: £1,911k) results from various recharges to Network Rail in relation to their East Coast digital programme and to fellow group undertakings. Restriction of use claims of £4,881k (2024: £4,051k) is compensation from Network Rail for revenue losses, net of costs, for engineering possessions, which result in amended passenger service timetables. Grant income of ₤281k (2024: £101k) is amortisation of a capital grant received from Network Rail in relation to their East Coast digital programme.

Click to expand...

The group undertakings might be shared functions and not all East Coast digital programme related, ordering is normally by value. I've revised my post to clarify that. Grand Central's accounts do call out millions in government grant and plausibly would be a similar level for Lumo. Hull Trains isn't specified but appears to be in their much higher other income figure than other operators.

I’m not sure that how “Network Change” is accounted for adds any value to the discussion here.

If NR imposes “Network Change”, all operators, not just OAO, have to be compensated for the consequences. ERTMS is one such “Network Change”.

So you will find that stock fitment, any additional staff for the training period and costs of the training itself are some of the items that the scheme has to pay for. This isn’t unusual.

All the principles of how it works are in part G of the Network Code.

As a fairly regular user of Hull trains to Kings Cross I would make the following observations:

The LNER 'We dont really want to provide a service' 1tpd each way leaves Hull too early for reasonable connections from my location. Hull trains provide a roughly every two hours service through the day. They offer good prices on advance fares, with 1st class fares being realistic for those of us who appreciate the extra comfort but are travelling on our own dime, not an expense account. We can also change at Cottingham if its a Beverley service to connect with our Yorkshire Coast Line service. That is a same platform change so of benefit if loaded down with luggage.

Contrast that with the alternative: Nothern to Doncaster, a reasonably OK train, but then you join a probably already overcrowded overpriced LNER service at Doncaster and spend the next 1h 40m in seats that I think were rescued from the Spanish inquisition torture chamber, as 1st class is usually eyewateringly expensive at the times we would be travelling.

I also know that if OA was abolished we would lose the service from Hull whatever lip service DfT/lying politicians paid to it. Hull trains parent company spotted a gap and exploited it, local gossip is that service is profitable but not a massive money spinner. They also offer better connectivity at Retford, which was ignored by LNER for many years, and have added Howden as a rail head which is now fairly well used for a small station.

As for Lumo, they are providing a limited stop cheap option on the London - Edinburgh route, which LNER dont seem interested in providing, so if Lumo were canned what would happen to the passengers that currently use those services, my guess is a lot would fly, some wouldn't make the journey at all, and not many would swap to LNER

Grand Central are in a similar position to Hull Trains, they are offering services that LNER are not interested in providing.

The problem is we seem to be getting into the worst of all worlds currently, with ToCs being taken into public ownership, but the DfT micromanaging, with the treasury pulling the strings. Anything where the bean counters have ulimate control is doomed to failure and will inevitably be a race to the bottom. I would prefer to see 'GBR' given an annual settlement, a minimum service requirement, and be told to get on with it. In that situation OA may be less important.

Clarence Yard said:

I’m not sure that how “Network Change” is accounted for adds any value to the discussion here.

If NR imposes “Network Change”, all operators, not just OAO, have to be compensated for the consequences. ERTMS is one such “Network Change”.

So you will find that stock fitment, any additional staff for the training period and costs of the training itself are some of the items that the scheme has to pay for. This isn’t unusual.

All the principles of how it works are in part G of the Network Code.

Click to expand...

And even then, a lot of network changes don't require compensation.

Stephen42 said:

I'm not saying the TOCs necessarily cover their cost to Network Rail either (they don't). It's that OAOs don't changes the net public subsidy for the railways by increasing their profit. For TOCs it just changes what entity gets listed as requiring public support.

All the numbers in my previous post are from the OAOs filings at Company House. Lumo is registered as East Coast Trains Limited and the latest accounts are here. Page 22 has the detail:

The group undertakings might be shared functions and not all East Coast digital programme related, ordering is normally by value. I've revised my post to clarify that. Grand Central's accounts do call out millions in government grant and plausibly would be a similar level for Lumo. Hull Trains isn't specified but appears to be in their much higher other income figure than other operators.

Click to expand...

Thank you

For two consecutive weekends in the past month Hull Trains ran additional services due to Hull City/Hull KR playing at Wembley, with all trains being fully reserved. They altered their planned maintenance schedule to ensure 100% fleet availability for those days. On those days, LNER didn't run any additional services to/from Hull. It's a bit rich for DfT and their TOCs to complain on one hand that OAOs are abstracting revenue while not ceasing commercial opportunities like this when they arise.

KHCT said:

For two consecutive weekends in the past month Hull Trains ran additional services due to Hull City/Hull KR playing at Wembley, with all trains being fully reserved. They altered their planned maintenance schedule to ensure 100% fleet availability for those days. On those days, LNER didn't run any additional services to/from Hull. It's a bit rich for DfT and their TOCs to complain on one hand that OAOs are abstracting revenue while not ceasing commercial opportunities like this when they arise.

Click to expand...

LNER probably didn't bother because Hull were doing it. They certainly have run "footexes" before and so have Avanti and GWR.

KHCT said:

For two consecutive weekends in the past month Hull Trains ran additional services due to Hull City/Hull KR playing at Wembley, with all trains being fully reserved. They altered their planned maintenance schedule to ensure 100% fleet availability for those days. On those days, LNER didn't run any additional services to/from Hull. It's a bit rich for DfT and their TOCs to complain on one hand that OAOs are abstracting revenue while not ceasing commercial opportunities like this when they arise.

Click to expand...

Bletchleyite said:

LNER probably didn't bother because Hull were doing it. They certainly have run "footexes" before and so have Avanti and GWR.

Click to expand...

On both the 23rd May and 30th May LNER ran additional "Footex" services.