Sunday 21 October 2012

HS2 is not a franchise (thankfully)

On October 12 at 1815, crowds gather at London Euston, defying the so-called 'cliff' effect when off-peak fares become valid at 1900. HS2 Ltd predicts significantly lower volume growth on inter-city servives between London and the northwest than do recent franchise bidders.
Several days on from the storm breaking over the flawed award of the Inter-City West Coast operating contract, it is becoming increasingly apparent that the government’s franchising reforms are in tatters and the very franchising edifice itself is teetering on the brink.

As a regular passenger as well as an industry observer, it does not especially surprise me that the West Coast Main Line should be the straw to break the camel’s back. Although not always recognised as such, the route is arguably Britain’s most strategically important and certainly its most high profile, thanks most recently to Sir Richard Branson’s vituperative criticism of the ICWC bid process. Whilst the government’s decision to halt all live franchising processes clearly indicates the flaws in the system are structural, not route-specific, it is inevitable that the West Coast would be cited as a microcosm of the railway’s wider woes.

The response to the fiasco has in some ways been predictable – widespread calls for renationalisation, and speculative assertions about the implications for High Speed 2. Personally I am not quite sure how alleged errors by civil servants and/or ministers helps the case for renationalisation, but that is no doubt a debate that will run at length elsewhere. But on HS2, the accusation is simple: if the Department for Transport can’t tot up the sums on ICWC, surely the same applies to HS2? According to the Daily Telegraph’s London Editor Andrew Gilligan, ‘some of the same statistical models are being used, in different ways’ to appraise the project, whilst former cabinet minister and Amersham MP Cheryl Gillan claims ‘elements like inflation figures and passenger numbers are common to both’ ICWC and HS2.

Ms Gillan is now leading calls for ‘a root-and-branch re-examination’ of HS2. But why? Whilst the franchising system is no doubt in crisis, it pains me to state the obvious: HS2 is not a franchise, it is an infrastructure project, just like Crossrail or the prospective third runway at London Heathrow. In terms of scrutiny, HS2 could not contrast more starkly with ICWC. Consider the reams of documentation available on the website of project promoter HS2 Ltd and DfT's own site, whilst franchise data is squirreled securely away, notionally on ‘commercial confidentiality’ grounds. Since GNER’s controversial ‘back the bid’ campaign in 2005, DfT has banned bidders even from releasing details of proposed service changes for fear of compromising the byzantine competition.

But on HS2, parliamentary committees, academics and a plethora of dubious think tanks have all had their say. Those parties which strongly disputed HS2 Ltd’s consultation process and economic appraisal have challenged it under judicial review, which inevitably brings further scrutiny. Yet the comparison with ICWC here is perhaps telling: it took Virgin Rail Group’s lawyers barely a week to seize upon DfT’s flawed model for calculating franchise bid guarantees; DfT’s own lawyers then instructed it not to contest the judicial review. However, more than three years since the formal launch of HS2, and no equivalent ‘smoking gun’ appears to exist, and HS2 Ltd insiders confirm that legal advice has been taken at every decision point, for example when whittling down potential London termini from 50 to a single option. Can Mr Gilligan’s ‘it’s the same, but different!’ argument really carry the day?

As for Ms Gillan’s arguments, inflation is straightforward on HS2: inflation is by definition excluded, with all costs cited at ‘Year X’ prices. That is appropriate to make a go/no go decision or choose between options on the basis of a benefit:cost ratio where inflation affects B and C equally and so has a neutral effect. This does not apply to a commercial contract like a franchise where there is revenue and expenditure coming in and going out, as would be the case for any operating concession let for HS2 in the early 2020s.

But the most compelling ‘like for like’ comparison is to read across the volume growth assumptions for both the next West Coast franchise and HS2, noting of course the substantially similar markets that both would serve either side of 2026. So whilst First Group’s now-aborted franchise bid assumed volume growth (ie. passenger numbers irrespective of price point) of 6% per annum, HS2 Ltd's figure would be less than 2.5%, and this includes a premium for new journeys created by the faster journeys and the inevitably-significant reliability gains from brand new infrastructure. (For the record, ICWC volume growth over the past 10 years has been 6.3% per annum). Furthermore, all growth on HS2 is forecast to cease at the ‘cap year’ of 2037, a mere four years after the completion of the Y route to Leeds and Manchester; this is analogous to all traffic growth on the M6 motorway ceasing in 1975!

Indeed, it is ironic that if and when a long-term West Coast franchise is eventually re-let for the years to 2026, its volume growth forecasts may well be more modest, and much closer to the ultra-cautious HS2 assumptions.

22 comments:

  1. If 6% growth continued, and HS2 took over long distance West Coast traffic, there would be ~989 million journeys per year after 60 years (way more than the Tokaido Shinkansen, which serves 16 towns).

    With HS2 directly serving just *three* towns on the West Coast route, it's no surprise that the model's passenger forecasts have to be summarily capped.

    ReplyDelete
    Replies
    1. "If 6% growth continued, and HS2 took over long distance West Coast traffic, there would be ~989 million journeys per year after 60 years"

      And how is this relevant? No-one is forecasting 6% growth for the next 60 years. Antis are claiming pros forecast this, but that's a different matter.

      "With HS2 directly serving just *three* towns on the West Coast route,"

      I counted London, Birmingham, Stafford, Wilmslow, Stockport, Crewe, Runcorn, Liverpool, Warrington, Wigan, Preston, Carlisle and Glasgow. I made that thirteen myself. Are we using the same definition of the number three?

      In any case, you've missed the point. Even with arbitrary limits on the rate of growth and period of growth, we are heading for a horrendous situation on the railways if we do nothing. The TSC hit the nail on the head: the dangers of under-providing on the railways vastly outweigh the dangers of over-providing.

      Delete
    2. "I counted London, Birmingham, Stafford, Wilmslow, Stockport, Crewe, Runcorn, Liverpool, Warrington, Wigan, Preston, Carlisle and Glasgow. I made that thirteen myself. Are we using the same definition of the number three?"

      Are we using the same definition of HS2? I'm using it to refer to the proposed new-build Continental gauge high speed line joining London, Leeds, and Manchester.

      Only three cities currently served by the WCML would have stations on the HS2 Y network track: London, Birmingham, and Manchester.

      As Stafford, Wilmslow, Stockport, Crewe, Runcorn, Liverpool, Warrington, Wigan, Preston, Carlisle and Glasgow are on the West Coast Main Line, but not on HS2, the maximum number of train paths to (or through) them is not increased by the HS2 project.

      Inter City West Coast is ~2% of Great Britain rail passenger volumes (number of trips). So it's unclear:

      * why that corridor is being singled out for creation of new build track, instead of upgrades; and

      * why that new-build track would only serve 3 cities directly.

      Delete
    3. "As Stafford, Wilmslow, Stockport, Crewe, Runcorn, Liverpool, Warrington, Wigan, Preston, Carlisle and Glasgow are on the West Coast Main Line, but not on HS2, the maximum number of train paths to (or through) them is not increased by the HS2 project."

      Not correct. The Liverpool branch and WCML north of Warrington still have spare capacity. It's the branches to Manchester and Birmingham that are running out of train paths and the stretch between London and Rugby is getting pretty desperate.

      In fact, you might have noticed that under First's franchise proposal (and this bit of the proposal is not under dispute), they were going to increase the number of services between Birmingham and Glasgow, but very little extra to London. There's no problem getting more trains through these northern stations - it's getting the trains between there and London that's the problem.

      HS2 simply diverts the trains serving the north on to a new pair of tracks. Yes, there are questions about how this will affect intermediate stations, but quite frankly, the service to most intermediate stations is already so atrocious it would be very difficult to make it worse. It's no different to laying an extra pair of tracks along the WCML (really the only remaining possible upgrade south of Rugby), except that you don't have to demolish thousands of houses the happen to be next to the railway.

      Incidentally, I note from your blog that you think you can solve the capacity problem by diverting London-Birmingham trains via the Chiltern Line. I don't get how a plan that removes three train paths between London and Rugby increases capacity, but a rail plan that removes more than three train paths between London and Lichfield (and further in Phase 2) doesn't.

      Delete
  2. "Not correct. The Liverpool branch and WCML north of Warrington still have spare capacity. It's the branches to Manchester and Birmingham that are running out of train paths and the stretch between London and Rugby is getting pretty desperate."

    The maximum number of train paths to (or through) Stafford, Wilmslow, Stockport, Crewe, Runcorn, Liverpool, Warrington, Wigan, Preston, Carlisle and Glasgow is not increased by the HS2 project.

    Unused path capacity on the West Coast route is not restricted to northern England, as London Midland's Project 110 demonstrated.

    "I don't get how a plan that removes three train paths between London and Rugby increases capacity, but a rail plan that removes more than three train paths between London and Lichfield (and further in Phase 2) doesn't."

    Neither HS2, nor routeing West Midlands intercity trains via Chiltern, nor routeing more freight over F2N, increase the theoretical maximum paths on West Coast. But they all free up paths on it. The differences are in the cost, and environmental damage (much higher with HS2).

    "HS2 simply diverts the trains serving the north on to a new pair of tracks."

    It diverts *some* trains serving the north on to a new pair of tracks, for *part* of their journey (in the case of Liverpool trains, for example). Even with 18 paths, it would not be possible to route all fast North - South trains onto HS2.

    "It's no different to laying an extra pair of tracks along the WCML"

    It's no different to laying an extra pair of tracks along the WCML but with just one crossover to the other tracks, outside Lichfield, and no platforms anywhere between London and Manchester, except in a field in Bickenhill. And with trains that are too wide to ever use the other tracks.

    "really the only remaining possible upgrade south of Rugby"

    I beg to differ.

    ReplyDelete
  3. "The maximum number of train paths to (or through) Stafford, Wilmslow, [etc.] Carlisle and Glasgow is not increased by the HS2 project. "

    But we are not yet at the maximum. It's not that difficult - HS2 services to Liverpool and Scotland use spare capacity on existing track when we can, and new track where we can't. It's the same logic as 51m's idea of relieving "pinch points", the only different being that frankly the entire southern section is a pinch point.

    "Unused path capacity on the West Coast route is not restricted to northern England, as London Midland's Project 110 demonstrated."

    Yes, had a look just now. Eeking out an extra 10% peak time when some LM trains coming into Euston are already over 145% load factor. All this proves is that there is very little spare capacity on the southern section.

    "Neither HS2, nor routeing West Midlands intercity trains via Chiltern, nor routeing more freight over F2N, increase the theoretical maximum paths on West Coast. But they all free up paths on it."

    Agreed.

    "The differences are in the cost, and environmental damage (much higher with HS2)."

    Is it? What about the costs of the Chiltern Line idea? An extra 20 minutes on Ldb-Bhm journeys (with an environmental cost with this causes rail passengers to switch cars), a potential bloodbath of local stops on regional services, and Coventry loses ALL IC services to London.

    And how do benefits compare? Fewer train paths freed than HS2, probably no improvement to intermediate stations, no capacity relief north of Rugby and no end to the laughable speeds of the cross-country services. Frankly, I'd rather struggle on with what we've got than try that as a solution.

    "It diverts *some* trains serving the north on to a new pair of tracks, for *part* of their journey"

    Some trains, for part of the journey but enough to get the traffic on the WCML down to something sane on the stretches where it matters.

    "It's no different to laying an extra pair of tracks along the WCML but with just one crossover to the other tracks, outside Lichfield"

    Yes, that is a downside, but the alternative of extra tracks along the existing line would require massive demolition of homes. I consider that too high a price to pay for adding extra crossover points.

    (I suppose there was an option of routing the line near the WCML but bypass the towns, but I don't see why the DfT should waste time investigating routes that neither pros nor antis are calling for.)

    "and no platforms anywhere between London and Manchester,"

    But London-Manchester trains barely stop anywhere on route as it is. Most intermediate stations are more than compensated by increased stoppage of WCML services.

    "except in a field in Bickenhill."

    If by a field in Bickenhill you mean an interchange with the Airport, the Coventry-Birmingham line, and a car park so that people who have to drive to a station don't have to spend half an hour struggling into a city centre, yes, that one.

    "And with trains that are too wide to ever use the other tracks."

    Or, more accurately, for services THAT NEVER LEAVE THE HS2 LINE, trains that are too wide to use the other tracks. And why would you want to? They're "captive" trains.

    "Classic compatible" trains *can* leave the tracks. That is the whole point of having two different kinds of HSTs. Arguing that London-Birmginham HSTs are useless because they can't use non-HS lines is like arguing DLR trains are useless because they can't use London Underground lines.

    "Even with 18 paths, it would not be possible to route all fast North - South trains onto HS2."

    This was taken into consideration. If rail use rockets and HS2+ECML+WCML+MML isn't enough, expect HS3.

    "I beg to differ."

    Apart from a change to peak pricing and a number of improvements that won't make any difference to the WMCL, it's 51m's plan. Which has been dismantled repeatedly.

    ReplyDelete
  4. "Arguing that London-Birmginham HSTs are useless because they can't use non-HS lines is like arguing DLR trains are useless because they can't use London Underground lines."

    If there were a blockage on HS2, captive trains would not be divertible onto alternative routes. That is a massive design and resilience shortcoming.

    "Apart from a change to peak pricing and a number of improvements [RP6] won't make any difference to the WMCL, it's 51m's plan."

    51m = 12-car ICWC trains in medium term; London - Birmingham intercity stays on West Coast; Stafford by-pass is built.

    RP6 = 11-car ICWC trains in medium term; London - Birmingham intercity goes to electric Chiltern Main Line, no Stafford by-pass is built. (Etc)

    So, not the same thing at all.

    "What about the costs of the Chiltern Line idea? An extra 20 minutes on Ldb-Bhm journeys (with an environmental cost with this causes rail passengers to switch cars), a potential bloodbath of local stops on regional services, and Coventry loses ALL IC services to London."

    Under the HS2 scheme, Coventry loses ALL IC services to London. The WCML fast service would become a sort of inter-regio, serving additional stations like Rugby (see Centro's proposal).

    Journey time London - Birmingham with Chiltern intercity is around 90 minutes, which is faster than the car journey. So why would travellers switch to a slower mode?

    "If by a field in Bickenhill you mean an interchange with the Airport, the Coventry-Birmingham line, and a car park so that people who have to drive to a station don't have to spend half an hour struggling into a city centre, yes, that one."

    Go-HS2 (Centro) are now quoting the "45 minutes" journey time for London to Birmingham, i.e. for trains that would not even stop at Bickenhill.

    "Yes, had a look just now. Eeking out an extra 10% peak time when some LM trains coming into Euston are already over 145% load factor. All this proves is that there is very little spare capacity on the southern section."

    There is a lot of un- and mis-used capacity on the southern WCML. Euston overcrowding is mainly outbound and much less of an issue than at other termini. As I've pointed out before, current LM crowding is largely traceable to rolling stock shortages.

    ReplyDelete
  5. "If there were a blockage on HS2, captive trains would not be divertible onto alternative routes. That is a massive design and resilience shortcoming."

    Don't see how that any worse than now. Apart from the Northampton loop, there's no alternative route between London and Birmingham already. However, Eurostar is constrained to HS1 by the lack of overhead electrification elsewhere. If this was a problem, one would have thought it would have been noticed by now.

    As an aside, if you're so in favour of divertable trains - and certainly here in the north-east we do get occasional problems on the ECML - why are you opposed to these new diesel/electric hybrids which will enable just that?

    "RP6 ... London - Birmingham intercity goes to electric Chiltern Main Line,"

    Ah, beg your pardon, I hadn't realised you were going for the one solution that I find ever worse than 51m.

    The Chiltern Mainline already takes roughly 6tph over most of its route. Add the Main Bhm/Ldn services and we're already looking at 9tph on a mixed-speed double track route. Bearing in mind the WCML struggles on 15tph on a four-track route, that sound wildly optimistic.

    "Under the HS2 scheme, Coventry loses ALL IC services to London."

    Let's forget semantics and look at numbers. Under post-HS2 plans, trains will call at all three of Rug/MK/Wat instead of one. An extra 8 minutes journey time maximum (and great if you're travelling to one of these destinations as you're up from 1tph to 3tph). Under your RP6 idea, the only way to get to London is on the LM service, with 11 calling points via the Northampton loop, and an extra 45 minutes. I know what I'd prefer.

    "Journey time London - Birmingham with Chiltern intercity is around 90 minutes, which is faster than the car journey."

    If you're going centre to centre. Not necessarily suburb to suburb. Selly Oak to Barnet, for example, is a 2 hour car journey. That's quicker than train via Chiltern line when you factor in the local journeys. So yes, there is an incentive to switch.

    "Go-HS2 (Centro) are now quoting the "45 minutes" journey time for London to Birmingham, i.e. for trains that would not even stop at Bickenhill."

    Hang on, one moment you were dismissing this station as a field and now you're complaining trains won't stop there. In any case, what is the problem about some trains stopping and some not on a 9tph+ service?

    "Euston overcrowding is mainly outbound and much less of an issue than at other termini."

    Only terminus with a worse PIXC is Paddington, which will be gaining a lot of extra seats through IEP and Crossrail.

    "As I've pointed out before, current LM crowding is largely traceable to rolling stock shortages."

    Thanks for showing me that document. I've had a look at the full version which shows that the 12-car trains are mostly nearly full, full or unbearably full. That's not something you can blame on rolling stock shortage.

    ReplyDelete
  6. Your response suggests that you don't realise that the Chiltern line was built with significant sections of four tracks. (By the way, 6 trains an hour on one track is a pathetic use of capacity.)

    I don't know what your "8 minutes" refers to. Each additional stop on West Coast 'intercity' would impose ~5 minutes extra journey time.

    The RP6 concept keeps Birmingham intercity trains off the West Coast trunk, which passes through Rugby. It does not prevent fast trains from Birmingham, Birmingham Airport, or Coventry, running to London via Leamington Spa.

    As paragraph 4.2.2 of HS2 Ltd's Demand Model Analysis (February 2010) stated, "Over half of all trips (and 90% of rail trips) start within central London...Trips that start in Outer London are dominated by car trips... Car is likely to be more attractive for trips which start or finish a significant distance from a train station at either end." As Selly Oak to Barnet via HS2 would require transfer - most likely on foot - between two stations in Birmingham, overall travel time would be much the same as today.

    The facts on Bickenhill HS2 are (i) it's a field; (ii) it could not be a stop for Birmingham HS2 trains reaching London in 45 minutes.

    London Midland crowding is mainly outbound, and caused by inefficient path utilisation and rolling stock shortage. But it's much less of an issue than at other termini. PiXC is a part line estimate of percentage crowding. In volume terms, crowding on London Midland is much lower than on the former Southern and Eastern regions. No LM services are "unbearably full".

    ReplyDelete
  7. "Your response suggests that you don't realise that the Chiltern line was built with significant sections of four tracks."

    Not sure what you mean by "significant sections", because every document I've read describes the Chiltern Main Line as a predominantly two-track railway.

    Anyway, taking all of this into account, pages 56-57 of the WM+Chilterns RUS doesn't bode well for an extra 3tph. It will take a proper schematic diagram of the proposed service to make me believe this is viable.

    "(By the way, 6 trains an hour on one track is a pathetic use of capacity.)"

    Not on a mix-use mixed-speed railway it isn't. Where I live (Durham), we get roughly 5tph plus freight going through the station. We used to have a perfectly decent hourly-ish service to Saltburn too, but that had to be axed because it wasn't viable to run 75mph trains on a double track where everything else runs at 90-110 mph.

    "I don't know what your "8 minutes" refers to. Each additional stop on West Coast 'intercity' would impose ~5 minutes extra journey time."

    All right, let's assume 5 for argument's sake. So, two extra stops on the way to London: 2x5 = 10 extra minutes max. Still considerably better than making Cov passengers go via either London Midland or Chiltern.

    "It does not prevent fast trains from Birmingham, Birmingham Airport, or Coventry, running to London via Leamington Spa."

    Okay, I accept that's a better idea than running entirely along the Chiltern route. (It also protects services to Solihull.) On the downside, you lose the change to improve local Cov-Bhm services, it adds an extra 6 minutes on to Bhm-Ldn journey times (on top of the extra 20-28 from re-routing), Cov-Ldn journey times are still likely to be an extra 20 minutes longer over WCML routes, and all IC services to Rug/MK/Cov are lost. Better, but still not great.

    "As Selly Oak to Barnet via HS2 would require transfer - most likely on foot - between two stations in Birmingham, overall travel time would be much the same as today."

    Assuming local journey times remain the same:

    * Via HS2: 38 minutes less Bhm-Ldn, but (as a conservative estimate) 10 minutes more changing NS-CS than you would over the current NS-NS connection. (And for WM routes that go to Moor Street, it becomes 10 minutes less.) Total: c. 28 minutes less than the status quo.

    * Via Chiltern line: 20-28 minutes over the status quo (28-36 if routed via Cov), assuming no loss of reliability on the Chiltern line, no extra time required to walk from NS-MS, and no loss of London local journey time from coming into Paddington/Marylebone instead of Euston.

    So no, overall travel time is not about the same.

    "The facts on Bickenhill HS2 are (i) it's a field;"

    ... with a car park, airport and rail interchange on it. What relevance is "it's a field"?

    "(ii) it could not be a stop for Birmingham HS2 trains reaching London in 45 minutes."

    But you could compensate for that by stopping some Liv/Mcr services (which actually is a good idea in its own right).

    "London Midland crowding is mainly outbound"

    So what? A seat on the inward journey doesn't make the outward journey any more variable.

    "caused by inefficient path utilisation"

    I'll believe that when I see a schematic diagram showing something better.

    "and rolling stock shortage."

    But the 12-car trains are already packed. Lengthening a different train won't solve that problem. Unless you want to introduce peak pricing for season ticket holders - if you would like to endure the (justified) uproar by proposing that, be my guest.

    "No LM services are "unbearably full"."

    If your response to passengers having to travel on trains where people stand for more than 20 minutes EVERY DAY is to stop complaining because we're not employing pushers (yet), I can guess how they'll react.

    ReplyDelete
  8. "Via Chiltern line: 20-28 minutes over the status quo (28-36 if routed via Cov), assuming no loss of reliability on the Chiltern line, no extra time required to walk from NS-MS, and no loss of London local journey time from coming into Paddington/Marylebone instead of Euston."

    The RP6 concept uses Snow Hill (not Moor Street), as the Birmingham departure point; the journey time, via Solihull, would be around 90 minutes, which is not "20-28 minutes over the status quo".

    In the HS2 concept, the transfer time New Street to Curzon Street would certainly be more than 10 minutes platform-to-platform. And like Curzon Street, Euston would be an end-on design, so the average walk to the Underground would be longer than today.

    "If your response to passengers having to travel on trains where people stand for more than 20 minutes EVERY DAY is to stop complaining because we're not employing pushers (yet), I can guess how they'll react."

    My response is that HS2 is of zero use in addressing 2012 capacity shortfall on London Midland; and crowding is, in any event, a much bigger issue on the Southern and Eastern Regions.

    "Where I live (Durham), we get roughly 5tph plus freight going through the station. We used to have a perfectly decent hourly-ish service to Saltburn too, but that had to be axed because it wasn't viable to run 75mph trains on a double track where everything else runs at 90-110 mph."

    Because the HS2 trunk would have no stations between London and Manchester (apart from a field in Bickenhill) fast trains would have to be retained on West Coast to serve all the other towns. So if 90-110 mph fast trains have to be retained, how can significant extra capacity be freed for increased freight and stopping passenger trains?

    The HS2 concept is shot through with contradictions.

    ReplyDelete
  9. "The RP6 concept uses Snow Hill (not Moor Street), as the Birmingham departure point; the journey time, via Solihull, would be around 90 minutes, which is not "20-28 minutes over the status quo"."

    By "90 minutes", I assume you're referring to the expected journey time after the Evergreen 3 upgrade, because the current journey time is 127m. The expected journey time is in fact 92-100m. The fastest Bhm-Ldn journey is 72m. The typical journey time of 84m is faster, and that's before consider any upgrades that might eek out more speed on the WCML.

    Of course, if you route these trains by Solihull then that's BI and Cov out the window. You can't have it both ways, either you extend the journey time routing it that way, or you inflict a massive service reduction on these two stations (and probably the intermediate stations to Solihull too).

    Whatever the figures, there's no getting round the fact that this proposal makes services worse for everyone travelling from the West Midlands, with only questionable benefits to stations on the rest of the WCML.

    (Incidentally, if you were going to do this proposal, at least send the trains the New Street instead of Snow Hill using the existing Euston slots. MS/SH add the same transfer time that Curzon Street does without the benefit of a faster journey.)

    "In the HS2 concept, the transfer time New Street to Curzon Street would certainly be more than 10 minutes platform-to-platform."

    Yes, but that doesn't matter. The increase in journey time is the platform-platform transfer time between NS/CS MINUS the platform-platform transfer time inside New Street. Ultimately, the EXTRA transfer time is going to be roughly the time it takes to walk concourse-concourse. An entrance-entrance is 5 minutes, 10 minutes seems like a reasonable guess. Much less than the time saving once you're moving in any case.

    "And like Curzon Street, Euston would be an end-on design, so the average walk to the Underground would be longer than today."

    Euston is already an end-on design (of course it is, it's a terminus). What's the difference?

    "My response is that HS2 is of zero use in addressing 2012 capacity shortfall on London Midland;"

    True, HS2 doesn't help pre-2026. Unfortunately, we're running out of ways to improve the existing line sooner. Being unable to relieve capacity before 2026 is not an excuse for never relieving capacity.

    "and crowding is, in any event, a much bigger issue on the Southern and Eastern Regions."

    Still not seen your evidence backing this up, but, in any event, the Thameslink programme, Crossrail and IEP between them are improving seat capacity in most directions.

    "Because the HS2 trunk would have no stations between London and Manchester (apart from a field in Bickenhill)"

    That's the third time in a row you've used that soundbite and it's getting tedious, but moving on ...

    "fast trains would have to be retained on West Coast to serve all the other towns."

    Wrong. Currently 4tph already stop nowhere between London and Stoke/Stafford. They can be replaced with an HS2 service (on the phase 1 stretch at least) with no need to retain the WCML service to serve the intermediate stations.

    "The HS2 concept is shot through with contradictions."

    That's rich.

    ReplyDelete
  10. "Still not seen your evidence backing this up".

    Well, first try getting informed. Then, suppose trains going into Euston had twice the percentage crowding of trains into London Bridge. In that event, there would still be more people affected by crowding, on trains into London Bridge.

    Peak arrivals at Euston don't account for a particularly large amount of London commuting.

    "By '90 minutes', I assume you're referring to the expected journey time after the Evergreen 3 upgrade, because the current journey time is 127m."

    No, it's not 127 minutes. Chiltern is 90 minutes London to Birmingham, with two stops, before any upgrades that might eke out more speed.

    "Of course, if you route these trains by Solihull then that's BI and Cov out the window. You can't have it both ways, either you extend the journey time routing it that way, or you inflict a massive service reduction on these two stations (and probably the intermediate stations to Solihull too)."

    No, Solihull and the intermediate Chiltern stations in the Birmingham area are on a four-trackable formation. But I can't see demand justifying more than one train an hour Birmingham - Coventry - Leamington Spa.

    "Whatever the figures, there's no getting round the fact that this proposal makes services worse for everyone travelling from the West Midlands, with only questionable benefits to stations on the rest of the WCML.

    (Incidentally, if you were going to do this proposal, at least send the trains the New Street instead of Snow Hill using the existing Euston slots. MS/SH add the same transfer time that Curzon Street does without the benefit of a faster journey.)"

    I'm not sure how direct electric trains from West Bromwich, Walsall, Stourbridge, and Solihull, with a transfer time of *zero*, "makes services worse for everyone travelling from the West Midlands".

    "Euston is already an end-on design (of course it is, it's a terminus). What's the difference?"

    In the HS2 economic model, a one minute time saving is worth millions. With 400 metre long HS2 trains, the average walk to exit is going to be more - and take slightly longer - at both Euston and Birmingham.

    "'fast trains would have to be retained on West Coast to serve all the other towns.'

    Wrong. Currently 4tph already stop nowhere between London and Stoke/Stafford. They can be replaced with an HS2 service (on the phase 1 stretch at least) with no need to retain the WCML service to serve the intermediate stations."

    Most WCML trains do stop somewhere between London and Stafford, so Not Wrong. Fast trains would have to be retained on West Coast, to serve all the other towns.

    ReplyDelete
  11. "Well, first try getting informed."

    Okey-dokey. So - other stations such as London bridge take more peak passengers than Euston. That is not surprising as Euston only takes trains from one commuter line and London Bridge takes trains from all over South London, Kent and Sussex. But crowding is not determined by the number of passengers alone - it is also determined by the capacity offered by the number of seats. Or passengers in excess of capacity. Which you yourself showed is higher than any other terminal except Paddington in the afternoon peak. Not so much the morning, granted, but that doesn't help.

    But in any case, all of this is academic. London Bridge redevelopment starts next year, and that will massively increase capacity in every sense. London Bridge's crowding is being address imminently - what's the problem?

    "No, it's not 127 minutes. Chiltern is 90 minutes London to Birmingham, with two stops, before any upgrades that might eke out more speed."

    Ah, now I think I know what you're referring to. All Chiltern trains are between 1h58 and and 2h25 from BSH to LM, apart from two morning trains which are 1h47 and 1h41. Presumably, you think you can get all trains to go that fast and faster still.

    Network Rail clearly doesn't, as their envisaged time post-Evergreen 3 is 92-100 minutes. I'll believe you can cut it further when I see a credible plan to achieve that and not before.

    "No, Solihull and the intermediate Chiltern stations in the Birmingham area are on a four-trackable formation."

    Errm, no it's not. Solihull-Tyseley is double-track. I've checked Google maps.

    "But I can't see demand justifying more than one train an hour Birmingham - Coventry - Leamington Spa."

    And I can imagine what the reaction from Coventry will be, seeing as their original opposition to HS2 was this rubbish about being cut to 1tph. have to say, it seems a little hypocritical to complain that low-demand stations might lose service due to HS2 and then gloss over a big reduction to a major population centre as not enough demand.

    "I'm not sure how direct electric trains from West Bromwich, Walsall, Stourbridge, and Solihull,"

    But still fewer services than you could get from New Street.

    "with a transfer time of *zero*,"

    No better than Curzon Street which is *next* to Moor Street.

    ""makes services worse for everyone travelling from the West Midlands".

    Because even if you do get a more convenient change at Moor Street, you still get a slower journey time once you're headed towards London.

    "With 400 metre long HS2 trains, the average walk to exit is going to be more - and take slightly longer - at both Euston and Birmingham."

    Actually, I believe the HS platforms at Euston are going to be extended forward, not backwards, so the walk will be the same or shorter (and certainly easier if you want to get to Euston Square).

    Not that this is relevant to what we were discussing. You were trying to argue that this makes suburb-suburb journeys longer and less appealing to going by car. You can't seriously be arguing an extra 30 seconds walking along a station platform is going to prompt people to drive instead.

    (And if you're now opposed to longer trains as well as more tracks, you're quickly running out of ways of getting more seats.)

    "Most WCML trains do stop somewhere between London and Stafford, so Not Wrong."

    And the key word is "most". That's "most", not "all". Can you please explain to me why diverting the 4tph that stop nowhere between London and Stafford doesn't relieve capacity, but diverting 3tph from London to Birmingham (which may have to be partially replaced to compensate for the loss to Watford, Milton Keynes and Rugby anyway) does relieve capacity.

    ReplyDelete
  12. "Via Chiltern line: 20-28 minutes over the status quo".

    London - Birmingham Inter City West Coast, in 2012, is generally 83 - 84 minutes. In the current timetable, plenty of trains run in 97 minutes or less, London to Birmingham. So your assertion is incorrect.

    "And if you're now opposed to longer trains as well as more tracks, you're quickly running out of ways of getting more seats."

    The RP6 concept, allowing 16-coach trains to run between London and Birmingham, involves four-tracking sections of the Chiltern Main Line and Midland Main Line - including the Birmingham approach through Solihull, which you claim isn't four-trackable, because you looked at Google Maps (?).

    "'I'm not sure how direct electric trains from West Bromwich, Walsall, Stourbridge, and Solihull,'

    But still fewer services than you could get from New Street.

    'with a transfer time of *zero*,'

    No better than Curzon Street which is *next* to Moor Street.

    'makes services worse for everyone travelling from the West Midlands'.

    Because even if you do get a more convenient change at Moor Street, you still get a slower journey time once you're headed towards London."

    There is no "convenient change at Moor Street". The RP6 proposition is to run direct trains, that don't require a change at Moor Street, or anywhere else.

    If you're going to comment on a proposal, why not do comparisons based on facts, and read what's being proposed?

    "Can you please explain to me why diverting the 4tph that stop nowhere between London and Stafford doesn't relieve capacity, but diverting 3tph from London to Birmingham (which may have to be partially replaced to compensate for the loss to Watford, Milton Keynes and Rugby anyway) does relieve capacity."

    Quite. I'd like to know how running 3tph from London to Birmingham on HS2 relieves capacity, when they would have to be 'partially replaced', to compensate for the loss of connectivity to intermediate points.

    ReplyDelete
  13. "The RP6 concept, allowing 16-coach trains to run between London and Birmingham, involves four-tracking sections of the Chiltern Main Line"

    To put it bluntly, it's hard to comment on your proposals because it's impossible to work out what you're actually proposing. Once would expect a proposal to divert all London-Birmingham trains and to four-track lines to be listed. They aren't. But no worries, let's start again.

    So, we're looking at four-tracking lines to provide the capacity needed for an extra 3tph. There's probably enough space along the line to provide sufficient opportunities to add what you need without losing many local services. This would free up space on the WCML for more local services. So far, so good.

    Now, the catches:

    Cost: New track is not cheap, especially not when you're mitigating against a working railway and adjacent suburbs. A short stretch of four-tracking part way between Birmingham and Coventry is already estimated at £900m, so we can expect an equivalent stretch near Solihull to cost something similar. That alone wipes out your saving from the supposedly unnnecessary Stafford bypass. We can expect even more four-tracking in the London area, re-routing all the Underground lines and sidings that run alongside various sections. Add in the expense of the new/extended stations, and we're quite likely already eating up a large chunk of the £17bn you're planning on saving by not building Phase 1 of HS2.

    Snow Hill: I assume you want Snow Hill as the terminus because there's space to extent the platforms. This is probably correct, but Snow Hill is a far worse place to terminate London Trains than any other site. Unless you are changing for the Snow Hill Line or the Midland Metro (so not half the local services or any regional service), it's a longer walk to New Street than either Moor Street or Curzon Street.

    Also, Snow Hill is a through station. The few through platforms are already busy enough as it is, and holding three trains per hour for c. 15 minutes whilst it they turn round is a very dubious proposition.

    Intermediate stations: There is no rail route between Snow Hill and Coventry unless you add some chords, at further expense. You then have a choice of either extending all journey times to retain the existing Coventry service, or cut it. And we know how Coventry reacts to suggestions of cuts, real or imagined.

    Things are no better for Rugby, Milton Keynes and Watford. The majority of their service comes from the Birmingham trains. You could compensate by stopping some northern trains instead, but that's no good if you want to go to the West Midlands, as you only services now are the painfully slow London Midland ones.

    London: I assume the idea behind this Old Oak Common station is that there's no room to lengthen platforms at Marylebone or Paddington. That is indeed possible, but Old Oak Common, whilst a good side for a Crossrail interchange, is a terrible location for a London terminal. Unless your final destination happens to be on the Crossrail route, the faff of completing your journey into Central London on Crossrail is a big time penalty.

    One side-effect is that this knacks up local services on the Greenford branch line.

    Revenue: It is very difficult to imagine getting more ticket revenue when - even with every trick in the book - Ldm/Bhm journey times will be longer. HS2 may recoup some costs through increased ticket sales. There's no chance under this scheme.

    So if you want this proposal taken seriously, you will need to come up with a detailed proposal stating:

    * How much new track you propose to build,
    * What full service pattern you propose,
    * What cuts to local services are acceptable, and
    * How much you expect this to cost.

    "Quite. I'd like to know how running 3tph from London to Birmingham on HS2 relieves capacity ..."

    Because - as I have clearly explained - it's the northern trains that don't need replacing, not the Birmingham ones.

    ReplyDelete
  14. Snow Hill station's location - in the business district of Birmingham city centre - is an advantage, not a disadvantage.

    As is the fact that it's a through station. As well as facilitating the Walsall - Redditch Crossrail line, the Benson Road curve would also allow inter city trains to run from Walsall (etc) to London over the Chiltern Main Line. So holding three trains per hour at Snow Hill for c. 15 minutes whilst they turn round, is not part of the proposition. In the RP6 concept, 16-coach trains would only operate as far north as the West Midlands.

    Investment in rail junctions such as Benson Road, the Camp Hill chords, and St Andrews East (Adderley Road South to Camp Hill) could offer substantial connectivity benefits. While the Benson Road curve would allow Coventry trains to reach Snow Hill, there is no reason why a Birmingham - Coventry - Leamington - London service could not operate from New Street.

    The real reason for HS2_at_Euston appears to be real estate development, not North London connectivity. In HS2's Ltd proposal, Old Oak Common would handle most passengers to/from East and West London, with the construction of a London Crossrail station there. But by developing Old Oak's connectivity further, and providing interchange with the Overground and Underground lines that pass close by, passenger dispersal for an RP6 Chiltern station would not be an issue.

    The RP6 concept has a lower cost than HS2, and delivers better outputs. Because electrification of the central section of Chiltern is now part of the government's "electric spine", the incremental costs of switching West Midlands inter-city from West Coast are substantially reduced.

    The costs of four-tracking Tyseley - Dorridge and Seer Green – South Ruislip (Northolt Junction) were estimated by Atkins as £441 million and £890 million respectively. Why these figures are so high, isn't clear. But even if they were correct, Inter City Chiltern Electric remains much better value than HS2.

    ReplyDelete
  15. Right, now we are getting somewhere. We have an indication of what work you think is necessary, and the cost involved. Still no word on proposed service patterns and whether it involves cutting local services, but it's a start.

    However, assuming you got your figures from Atkins Scenario C (that involves diverting Ldn-Bhm on to the an improved Chiltern Line), you missed a glaring problem. In order to achieve this AND not cut any local services, you need a lot more improvements than two stretches of four-tracking. The total cost? £8.36 billion. Even if we deduct the bit of electrification covered by the electric spine and the Kenilworth two-tracking, that's a few hundred million off at the most. Suddenly it doesn't seem quite good value.

    Then you can add in: building Old Oak Common station (£660m I believe, more if you want to integrate the Underground and Overground lines); the work needed at Snow Hill (which will almost certainly require a bigger concourse built over a working station - the £250m bill for the King Cross concourse might give you an idea what to expect); three chords to allow trains to go to Walsall and Coventry (the Ordsall chord is £80m - these three I suspect will be more complicated); possible re-signalling working through Snow Hill to allow the increased train frequency; lengthening Coventry platforms; repositioning signals and points on the Chiltern Line to accommodate 16-car trains; new sidings and train sheds for the longer trains; and loss-making rolling stock running nearly empty between Birmingham and Walsall. We could easily be passing the £10bn mark.

    And the benefits? Well, longer trains to London, yes, and presumably more regional services on the WCML (not even sure if that's in your proposal, but let's assume it's a benefit). As against that, journey times will probably be longer than the status quo (and certainly much longer than HS2), and services to former intermediate stations will be either severely cut or scrapped completely.

    So what if Snow Hill is the business district of Birmingham? That is no use for a business traveller going to London because business travellers do not start their journey from their place of work in Birmingham - they start from where they live. And no matter how many chords you add, there's no way a 3-platform station (plus two tram platforms) is going to serve as much of the region as 13-platform New Street.

    Walsall-Redditch crossrail? If we wanted that enough we'd have it - the infrastructure to allow it is already here. You simply route some Lichfield-bound trains to Walsall instead at Aston. You won't make it any easier by re-routing the train from a New Street-serving line to a Snow Hill-Serving like and then back on to the New Street-serving line again. All you are doing is swapping one congested section of track for another.

    And as for London, most passengers are going to be headed for the city centre. Euston offers five lines to different bits of the centre (two Northern Line Branches southbound, Victoria Line southbound, and Circle/H&C/Met both ways), Old Oak Common only one. The thought of getting the lion's share of a 16-carriage train into a 12-carriage train heading into the City doesn't look good.

    If you want this proposal to look remotely like value for money, you'll either have to state which local services you want scrapped to make way for mainline trains, or how you propose to deliver the same service with less improvements (preferably with a proposed service pattern). Only then is it worth looking at the benefits. What happened on the WCML taught me to be sceptical of bargain basement solutions. This seems like no exception.

    ReplyDelete
  16. "It is very difficult to imagine getting more ticket revenue when - even with every trick in the book - Ldm/Bhm journey times will be longer. HS2 may recoup some costs through increased ticket sales. There's no chance under this scheme."

    For trains making two stops, and with diesel haulage, the 2012 journey time on Chiltern is 90 minutes. That's all of 7 minutes slower than West Coast between Birmingham and London.

    For Walsall or West Bromwich, the RP6 journey time is much faster. Because, in 2012, there are no through trains from those places to London.

    According to the MVA/Atkins/Mott April 2012 Demand and Appraisal Report, demand between Birmingham and London would increase by 113% between 2010 and 2037 without HS2 (i.e., with journey times similar to 2012).

    RP6 effects on Chiltern local services would be minimal, because they would run on separate tracks for ~20 km out of Birmingham, and ~30 km out of London. Furthermore, stations on the Bicester cut-off were built with separate platform and passing tracks.

    There is no requirement for '£8.5 billion Scenario C' infrastructure for a ~90-minute Birmingham journey. That timing is already achieved, just with diesel loco haulage.

    ReplyDelete
  17. Congratulations on not answering the question. I'll ask again.

    The £8.5bn of rail improvements listed by Atkins are what they think is needed to divert mainline services on to the Chiltern line without harming local services. That's without journey time improvements - just the minimum amount of infrastructure improvements for this idea to be workable at all.

    You are claiming it is possible to do the same thing for only a fraction of the cost, four-tracking substantially less of the line. Clearly you and Atkins can't both be right here. I asked you to demonstrate why you are correct and Atkins is wrong. You didn't.

    If you want anyone to believe your proposal is achievable, you'll need to do at least one of the following:

    * Produce a proposed service pattern;
    * Find a railway somewhere else in the world that manages this kind of service level (that's 9tph mixed local and intercity services on a mainly two-track railway); or
    * Find some expert or authority who backs up your claim.

    If you will not do any of those three, then it ultimately comes down to deciding who we trust to have assessed this correctly: Atkins, or an anonymous blogger. And I have to point out most people put anonymous bloggers at the bottom of the pile (especially those who are there to push a single agenda).

    No government in their right mind would commit to an infrastructure scheme without making any attempt to check the claimed benefits are possible.

    ReplyDelete
  18. "The £8.5bn of rail improvements listed by Atkins are what they think is needed to divert mainline services on to the Chiltern line without harming local services. That's without journey time improvements - just the minimum amount of infrastructure improvements for this idea to be workable at all."

    Journey time improvements through very expensive by-pass and tunnel construction are part of Scenario C. But such works are not required for a transfer of West Coast intercity to Chiltern.

    "* Produce a proposed service pattern"

    HS2 Ltd - with 1000+ staff - are yet to produce a service pattern showing how their new line could accommodate services to both Heathrow and Europe. Or even a regularly-spaced service between Birmingham and Leeds.

    "* Find a railway somewhere else in the world that manages this kind of service level (that's 9tph mixed local and intercity services on a mainly two-track railway)"

    The country part of the Chiltern Main Line doesn't carry 'local stopping services' as such; with the exception of Kings Sutton, all the local stations were closed 30+ years ago. On two-track sections, trains would need to have similar performance profiles, which is perfectly manageable.

    "No government in their right mind would commit to an infrastructure scheme without making any attempt to check the claimed benefits are possible."

    Well, that's what the government appears to be doing, right now, with HS2.

    ReplyDelete
  19. "Journey time improvements through very expensive by-pass and tunnel construction are part of Scenario C. But such works are not required for a transfer of West Coast intercity to Chiltern."

    Aarrggghhhh. [Goes away. Bangs head on wall 50 times. Returns to keyboard. Resumes.]

    Scenario C IS the transfer of West Coast services to Chiltern. (And, okay, Scenario C also includes improvements on the WCML too, but those £8.5bn of improvements are for the Chiltern line alone. You need those whether or not you're improving the rest of the network.)

    "HS2 Ltd - with 1000+ staff - are yet to produce a service pattern showing how their new line could accommodate services to both Heathrow and Europe."

    All right. Let's assume the worst and suppose there'll no capacity left to provide through services to Europe. So what? That's no worse than today, and it still solves the problem HS2 is intended to solve, which is the lack of capacity on the WCML.

    "Or even a regularly-spaced service between Birmingham and Leeds."

    The current Leeds-Birmingham service is hourly. Two irregularly-timed services per hour is still an improvement.

    For pity's sake, have you ever travelled Leeds-Birmingham on train? I have, several times, and I can tell you anything's better than what we've got now.

    "The country part of the Chiltern Main Line doesn't carry 'local stopping services' as such; with the exception of Kings Sutton, all the local stations were closed 30+ years ago."

    And Atkins agrees. Scenario C doesn't involve four-tracking the sections where there's no local stations to worry about.

    Unfortunately, that's not where the money's going. There are local stations all the way between London and Haddenham & Thame Parkway, and that entire section that needs four-tracking according to Atkins.

    No way around it - if you want to keep costs down to what you claim, you've got to explain how you're going to mix fast services and stopping services on long sections with stations.

    "Well, that's what the government appears to be doing, right now, with HS2."

    You're welcome to question the business case, but if you're questioning the post-HS2 service pattern, the absolute worst-case scenario is that they won't be able to squeeze quite as many services on the new lines as previously hoped. Still an improvement.

    Contrast that with Rail Package Beleben, where the worst-case scenario involves massacring local services to make way for WCML services as a result of going cheap on the improvements needed, I know which one I'd rather pick.

    ReplyDelete