Think HS2 is an alternative to flights? Think again!

on Saturday, 14 January 2012. Posted in News from 2012, James's blog

Why plans for hs2 will do nothing to replace wasteful short-haul flights

This week, the government confirmed its plans to proceed with the first phase of its highly controversial £17 billion HS2 (High Speed 2) rail project, which will see a new fast link between London and Birmingham taking 49 minutes when it opens in 2026.

Confusingly, the government is also seeking to pass legislation for the next phase of the project, a 'Y' shaped continuation to Leeds and Manchester opening in 2032, bringing the total projected cost to £34,000,000,000. This has led to many inaccurate claims and counter claims regarding the actual scope of the project, whereas it has been impossible to do a rational assessment of this second phase of the project, as details of exactly where it will go have not been released, except that there will be some form of station in the East Midlands - presumably East Midlands Parkway Version 2.0 (drawing on the failures of the first one no doubt!).

Since before setting up in 2002, I have for a long time held the view that High-Speed Rail is a logical replacement for short haul flights. These flights are extremely wasteful as they use so much energy in the takeoff and landing process. So, why do I feel that the HS2 project falls so far short of such ambitions?

Train versus plane

This is naturally the first criteria I should look at when assessing a high-speed rail project like this, especially as I have also travelled by high-speed train in almost all other European countries that have them, and I have seen how short-haul flight routes have changed over the last 10 years as these rail routes have developed. So how does HS2 compare in this respect?

Claim: Up to 6,000,000 passengers per year will shift from UK domestic flights to rail travel due to HS2.

Reality: The train has already won the battle for point to point travel between London and northern England, with flights no longer being available between London City and Liverpool, or between London Heathrow or Gatwick and Leeds/Bradford airports. The high-frequency shuttle flights from London City airport to Manchester dwindled to just a handful each day before being axed, whereas flights from Manchester to Stansted are also well and truly finished.

Therefore, the only flight routes which would run directly in competition with the new rail line are those which operate from London Heathrow and Gatwick to Manchester. Even this route is highly vulnerable to the pressures of Air Passenger Duty (APD), which results in two sets of charges for passengers taking domestic flights, against one set for those people taking flights to mainland Europe and beyond. Already, three quarters of passengers on the Heathrow to manage the flights are taking connecting flights elsewhere -- not only are these taxed differently, but they are less likely to be substituted by rail services if passengers can make a direct connection in one terminal.

The current total market for flights between Manchester and London is around 1,000,000 passengers per year, although the market for point to point flights, as opposed to flight connections is clearly much smaller than that.

Where does the 6,000,000 passenger figure come from?

Apart from inflation based on a growing market (clearly not the case), the figure is largely made up by assuming that passengers in Scotland will transfer from short-haul flights on to rail services as a result of the gains in speed further down the line.

However, the ability of the HS2 project to deliver this as it currently stands is highly questionable. The first phase, by providing a faster route south of Rugeley, might knock around 20 minutes from the total journey time between Glasgow Central and Euston, not enough to make any significant difference, and certainly not enough to justify the huge investment in rolling stock which would be needed to make this happen. Any savings or on a journey times between Edinburgh and London would be miniscule, as the majority of this traffic would almost certainly continue down the east coast.

I would like to assume that the second phase of this project will join the West Coast Main line somewhere around Warrington, and that there will also be a join with the East Coast main line somewhere around York. However, this has not been put forward as part of 'Y' scheme, otherwise it would be more accurately described as a double 'Y' (Why o why I hear you say!).

Even if joins are provided at Warrington and York, the savings in time will still be slightly less than the one hour saving provided to Leeds and Manchester, as these will operate on a direct route, whereas the joins will operate as a spur. There is also a much more significant question about how different services will compete against each other for track access paths.

Trains operating entirely on the high-speed lines will be up to 20 carriages long, and they may also be double-decker, whereas trains operating on conventional lines will be restricted to single deck, and can only be as long as platform space allows at the stations they call at, although selective door opening may make length less of a problem, providing the terminus station can accommodate the full length.

High-speed line only trains could just be bought off the shelf -- for example the next generation TGV Duplexes could provide optimum capacity by joining two trains together and using a double-decker arrangement, whereas the hybrid trains would have to be customised to British classic line specifications, and therefore they may end up being more expensive, even though they would carry less passengers.

Viable route?

This does not, of course, mean that Glasgow or Edinburgh to London train services could not use the high-speed line at all, but it does mean that they are at a potential commercial disadvantage when compared to trains which use the high-speed line for the full length of the journey.

This may mean that discounted advance purchase tickets are either not available, or are not sufficiently attractively priced compared to low-cost flights operating between London and central Scotland. Don’t believe me on this – try finding an advance ticket on the current HS1 service from St Pancras to Ashford and beyond.

More likely is that if the time-saving compared to current journeys is only around 45 minutes, so there is less likelihood that lucrative business passengers will transfer from air to rail at this point – yes, the timings will be more appealing for some passengers, but not fast enough for a major switch by the premium passengers.

In effect, there is a significant squeeze both at the top and bottom end of the market in terms of operating a competing train service, compared with flights from London to Edinburgh and Glasgow. And European experience shows that a good fast train service may result in a reduction in frequency of flights, but it rarely actually kills the route entirely. For example, despite the TGV from Lyon to Paris offering fares from just £23 each way, and a service to both central Paris and CDG airport for connecting flights, Air France still operate multiple daily flights from to both Paris Orly and CDG airports from Lyon.

Either way, the figure of 6,000,000 is completely unrealistic, unless of course, the high-speed line was to continue all the way through to Scotland.

In terms of routing, it might make more sense to go up the West Coast and to have a split around Carstairs, although this leaves the question of how a high-speed line might get through this sensitive Lake District. A route heading East from Leeds via York could also serve Newcastle, but this may end up in a bit of a circuitous routing to reach Glasgow.

One thing that proponents of the scheme have failed to ram home is that although the cost per mile of the first phase stands at a prohibitive £145m per mile, much of this is due to the estimate £6bn cost of getting as far as Old Oak Common. The cost for phase 2 falls to around £63m per mile – still expensive even by European standards, but far more reasonable compared to phase 1.

A continuation to Scotland may be a phase3 – or should we call it High Speed 3 – arriving sometime in the 3rd Millennium at current rate? From the point of view of providing an alternative to short-haul flights, this would become far more viable. If anything, the cost per mile heading north may fall further, as land prices fall and as economies of scale kick in – although if extensive tunnelling is needed in the Lake District, these costs could rise again.

Either way, it should cost much less than another £34bn to continue north to Edinburgh and Glasgow – and it may be possible for as little as £9bn, assuming a slight reduction in cost compared to Phase 2. This is not much more than the cost of getting between Euston and Old Oak Common. Now I accept that this link is an important part of the scheme, but ultimately it still only provides a link as far as the north edge of London’s Zone 1, and connectivity with the first high speed line, which has still to be costed, and which is still going to present some serious operational challenges.

If, however, the link to Edinburgh and Glasgow was built first, followed by the link between Old Oak Common and Euston, that would be a real game changer!

Comments (16)

  • Matthew Williams

    Matthew Williams

    14 January 2012 at 20:31 |
    A very well thought out article. Why do they always have to focus just on London.

    Doing as you suggest would provide a fast link from Birmingham and Manchester to Scotland too, and that would provide an alternative to Flybe's monopoly on BHX - EDI & GLA routes!
  • Tom Grainer

    Tom Grainer

    14 January 2012 at 20:36 |
    So if I read you right, you are saying in this article don't tunnel under London, but then in your article about Gatwick (which I otherwise agree with), you are saying tunnel under London to join the projects up.

    Make your mind up!
  • James Avery

    James Avery

    14 January 2012 at 20:56 |

    It is a question of phasing and priority. The High Speed 2 project has rightly been criticised for its extortionate cost per mile, much of which is due to this extremely short stretch between Euston and Old Oak Common. I'd rather they build more of the line first, then opened up into central London when the level of service justified it - remember, Old Oak Common links in with Crossrail, Central Line etc anyway, Euston does not.

    The distance between Euston and Victoria is only around 3 miles - so linkage could be provided between them concurrent with the new tunnels, but this would also only be viable later down the line, should Gatwick be expanded significantly.
  • Tom Grainer

    Tom Grainer

    14 January 2012 at 21:11 |
    No, still don't get you. Sounds like a politician's answer. Either you want to build tunnel or you don't?
  • James Avery

    James Avery

    14 January 2012 at 21:19 |
    Tom, it is a question of bang for buck, or return on investment if you prefer. The tunnel on its own provides only a few minutes' saving for a cost of around £6bn. Where in London would you want to go? West End - take Crossrail, Oxford St / Tottenham Court Rd - take Crossrail, The City, Docklands etc etc. Where the tunnel would be useful is if it offered through links to other places - such as through to Stratford and then Kent - but this has not been made into any business case yet. Essentially even this is about east-west links, which are provided by Crossrail. North-South links are very poor - for example it is much quicker to reach King's Cross from Brighton by train and tube through Victoria than it is via direct Thameslink. That is why I think the tunnel becomes more interesting if a link is provided through to Gatwick and beyond.
  • George Collins

    George Collins

    14 January 2012 at 21:31 |
    You have missed the point. High-speed two is all about creating extra capacity to release track paths on the existing lines so they can take extra commuter and freight services. The intention therefore is that the new lines will carry passengers who would otherwise have travelled on crowded Virgin Trains services where they would not being able to find a seat, and to encourage people away from congested motorways to take the train. Offering an alternative to short-haul flights is an incremental bonus, it does not make the case, as that has already been clearly made for the other reasons.
  • James Avery

    James Avery

    14 January 2012 at 21:42 |

    I haven't missed the point at all, I have merely given a detailed breakdown of the case as it was made with regards to short haul flights. Clearly, the figures cannot be trusted. Other critics have given their view on the rest of the business case, and this includes the ridiculous notion that the average value of time spent on the train is based on a £70,000 salary and that train time productivity is zero.

    If you want to look at ways to increase capacity, then I find it ironic that the route goes to Birmingham first, a city served by four routings to London already (WCML via Weedon, Northampton branch, Chiltern Line, via Reading). I'm afraid I don't accept that they are all going to be completely full all of the time by 2026, nor do I accept that other alternatives such as longer trains and less first class space, have been explored. People say they can't remove first class because it reduces revenue - but HS1 already uses that model (no first at all), and First Great Western only have 2 - 2 1/2 first class cars per HST set.

    Capacity is also not provided north of Rugby, as the HS2 route joins south of Rugely, instead of continuing just a few more miles north to Colwich, where the Stafford and Manchester lines meet. In terms of speed, I still agree that high speed rail can bring tremendous benefits, but it would make more sense to create the biggest time savings from day 1.
  • Ewan Lambert

    Ewan Lambert

    14 January 2012 at 23:25 |
    So basically what you are saying is that instead of going through with the carefully considered government plans, billions more should be spent on a high speed link to Scotland, only to dump all passengers at some kind of London Parkway instead of taking them to the established terminal at Euston?

    Scots already benefit from free prescriptions, free tolls and a new Forth Road Bridge at our expense, why on earth should we let them have high speed rail services ahead of people in Leeds, making everyone suffer from an inferior service as a result?
  • Joan Higgins

    Joan Higgins

    14 January 2012 at 23:36 |
    What about Maglevs?
  • Marla Durden

    Marla Durden

    15 January 2012 at 18:35 |
    Surely, this whole hs2 is a desparate measure to leave a legacy by a group of Notting Hill luvvies who wanted to do something to replace plans to build a third runway at Heathrow. As if it will make the slightest difference to London's desperate need for a new runway! All it will do is suck even more passengers from Birmingham and Manchester through Old Oak Common to board flights from Heathrow! And Heathrow will also pick up passengers from City Airport and the others by becoming more accessible through Crossrail - so are they letting us have a third runway to cater for all this new demand? Of course not, because everyone is going to use the train instead - to get to, err - Dublin, Rome, New York and Beiing! DOH! We really must club together to fight this nonscence!
  • James Avery

    James Avery

    15 January 2012 at 19:30 |
    No, I am saying that the amount of miles which can be built for the same money is much higher further up the line. Old Oak Common could provide numerous links throughout London, aswell as Crossrail services. As the interchange platforms would all be parallel to each other, it would be much easier to change from Old Oak Common onto Crossrail, than it would be from Euston onto the deep tube lines.
  • James Avery

    James Avery

    15 January 2012 at 19:35 |
    Joan, they would offer numerous advantages over conventional adhesion rail, but they also have drawbacks, and have sadly been dismissed by the government. I'll do something on this in due course.
  • David Peilow

    David Peilow

    17 January 2012 at 22:59 |
    Thanks for a thoughtful article that doesn’t stoop to the usual mud-slinging that abounds with this subject (and thanks for the site – I’ve been using it for years for trip planning).

    I also passionately believe that short-haul flights are extremely wasteful, for the reasons you describe, particularly on an island the size of ours. That was brought home to me after my first TGV ride back in 1996. The TGV is a byword for quick, but at the time I was amazed by the ground it covers in a short time without the associated hassle of the airport. I started investigating the benefits and effects of bringing it to the UK and came to the conclusion that they are substantial.

    However you’ve hit the nail on the head about rail/air market share in your last section. HS2 must continue to Scotland to give the full benefits that come with significant high speed rail market share; particularly on London – Scotland routes, but also Newcastle and to an extent, routes from Southampton. Additionally, HS2 must fully integrate with Heathrow to capture transit passengers and HS1 to allow (for example) Birmingham – Paris or perhaps Manchester – Paris flights to be reduced. It is a real pity that Arup’s 2009 proposal for Heathrow hub was not pursued by the then government (see as it would have done that extremely well. I note now Labour are calling for something similar – after much of the planning for HS2 has been done.

    It has been shown many times in many countries that where rail has a 2 hour journey time, it can capture 90% of the total market. This percentage decreases on an S curve so that at 3½ hours, market share falls steeply and is only 45% (see for one example).

    So, in order to capture significant market share, a London – Scotland HS2 (or HS3) route would have to have a maximum journey time of 2½ hours and preferably 2 hours. Is this possible? Well, the West Coast Main Line is 399 miles long and the East Coast Main Line is 393 miles, meaning an average speed of approximately 160 mph is needed to achieve 2½ hours or 200 mph is needed for 2 hours. By comparison, today a French TGV service operates at an average speed of 173 mph and in China a CRH3 service has averaged 194 mph in commercial operation. Both achieve this with lower top speeds than HS2 is being designed for.

    At the present time, both France and China are reported to be developing trains with 250 mph top speed. South Korea – with a demographic similar to us – is producing the KTX-III, a 232 mph train. It seems very likely that we will be able to buy 250 mph or faster trains for HS2 in 2026, that they will be able to average at least 200 mph (6 mph than the Chinese have achieved to date) and thus they will get from London to Glasgow in two hours.

    This illustrates why the Secretary of State for Transport must hold her nerve and ensure HS2 is built for at least 250 mph running with minimal speed fluctuations. This is not about “saving 10 minutes to Birmingham” as has been claimed by protesters, but about saving 2 hours on the current London – Scotland journey times to get high market share. To get to Scotland, one must go through the Midlands first…

    The total rail/air market to Scotland today is 6.8m people a year. If trains can run from London to Glasgow or Edinburgh in 2 hours with 90% market share, then on to Aberdeen or Inverness on classic lines with 50% and 20% market share, then this gives the 6 million passengers figure that has been reported. To put it in perspective - this is enough for 32 *non-stop* trains a day from London to Scotland (around one every daytime hour, each way) *in addition* to the trains serving English cities and other stopping services that will use the line.

    If rail captures that much of the Scotland market, the CO2 emissions – including construction emissions – can be as little as a sixth of the equivalent air traffic. That is derived using figures from today’s electricity generation mix (renewables only further improve the case). However, the high speed track must reach Scotland – it is not, as you rightly point out above, enough to stop at Manchester or Leeds, as we will then be much further down that S curve for London – Scotland journey times. Not reaching Scotland means that construction emissions would be spread over fewer passengers and may only break even with today’s per-passenger emissions from air.

    Claims that HS2 is merely “carbon neutral” (not better, not worse) than the status quo have been misstated. This phrase was previously used by the government to describe the HS2 first phase to Birmingham *in isolation*, where there is currently no air traffic to displace, but since then campaigners have seized upon it and applied it to high speed rail in general. This is very far from the truth and has been analysed in detail by many organisations worldwide.

    Emissions aside, the benefits of energy security and freed airport capacity for long haul flights shouldn’t be overlooked and clearly the airports know this. If high speed rail is used to the full, then enough short haul routes are removed from London airports to offset the capacity proposed to be created by the third Heathrow runway. This clears the way for more useful and more profitable long haul air traffic. At under 40 minutes, Birmingham International is closer to central London than Stansted.

    As for the project cost – it has been pointed out that we are currently spending £2bn on the Crossrail project. This budget is expected to be taken over taken over by HS2 in 2016 when Crossrail is finished. That allows the system proposed as far as Manchester and Leeds to be built by 2032. The Scottish government has said that extending the line to Glasgow and Edinburgh would cost a further £15bn, meaning if the same budget were continued then central Scotland could be reached by 2040. EU funding may even be available for such a link across a border. It should be pointed out this timeframe is similar to the length of time it took to build the Paris to Marseille TGV line – the first project phase was approved in 1971 and the third phase reached Marseille in 2001, a total of 30 years.

    HS2 is affordable. Leaving aside issues such as the 60% Treasury “optimism bias” already in the headline numbers and the fact that the French seem to be able to build similar lines for half the cost, the same £2bn budget that is funding Crossrail now (and was started in the recession) is just 0.1% of British GDP, 0.3% of government spending and 9% of the transport budget. This seems a reasonable sum to spend to create an entirely new – and badly needed – high speed transport infrastructure that isolates us from future increases in the price of imported oil. Nevertheless, despite its high capital cost, if HS2 achieves low journey time and high market shares, then construction cost can be amortised across a large enough population of passengers to actually reduce fares relative to today. This is another reason to ensure attractive journey times are possible.

    More details, references and data on these and other HS2 topics can be read here:
  • James Avery

    James Avery

    17 January 2012 at 23:51 |
    David, many thanks for an excellent response. I agree with a lot of what you are saying, although I still think there are far too many flaws with the current proposals, which I will elaborate on in the next few days. I have also been reading through the DfT's own documentation, which shows that trains on the classic lines may well be shorter and slower than the existing Virgin Pendolini - creating a further disadvantage for running trains down from Scotland. I've also had a look at the likely amount of train-miles the new line will enable, and it is clear that Phase 2 represents a 3-4x better investment than Phase 1. Network Rail have also estimated that the cost per mile of building from York to Newcastle goes down to £41m - or £3.5bn for that whole 85 mile stretch, far less than the £6bn figure I have seen for getting to Old Oak Common from Euston!

    If those figures are correct, it might be possible to continue to Edinburgh for another £4bn - although I'm not sure about the routing through the Borders. This might be around half the figure the Scottish government have quoted - although I'm afraid after the way costs of the Forth 'Replacement' Bridge have yo-yod I'm not sure I trust them!
  • Chris Brundle

    Chris Brundle

    19 January 2012 at 15:35 |
    I think it is time we accelerated full throttle - HS2 is grossly inadequate, not because the nimbys of the Chilterns don't like it, but because it doesn't do far enough, fast enough. Until we have a fast link to Scotland, the rest is pointless. Just look at what they are doing in Asia! It is time for us to wake up!
  • Andy 85

    Andy 85

    31 January 2012 at 06:34 |
    It isn't realistic to expect a high frequency service of trains from Birmingham to Brussels - they are struggling to keep more than about 3 flights per day on that route as it is.

    Unless Euston was a through station, so Eurostar (or whoever) could stop there, and then continue as a domestic service, there's little chance these services would ever work.

Leave a comment

You are commenting as guest. Optional login below.