Return to Archive -by date - by topic.

INFORMED SOURCES March 2005

West Coast InterCity – all to play for.

Who will break the cycle of FUD?

Last month I reported that Virgin Rail Group and the SRA had agreed to suspend the long running renegotiation of the InterCity West Coast franchise until the operational and commercial prospects for the business could be evaluated with a degree of certainty. Figure 1, which illustrates Table 1, shows why both VRG and SRA blinked.

Fig 1

Table 1

After the longest period of continuous economic growth since George and Robert got the blast-pipe working in ‘Rocket', after seven years of Richard Branson's entrepreneurial flair, passenger journeys on the West Coast are back to where they were in 1997/98. Passenger kilometres are below even the 1994 figure when the last recession was still biting.

With so much Fear, Uncertainty and Doubt (FUD) around, it is not surprising that neither party was prepared to risk money on predicting how the market would behave over the remaining seven years of the franchise. Yet, unaccountably, Informed Sources were confident as recently as last September that new franchise terms would be signed by Christmas.

Negotiations will now restart in September when the impact of the first year of the 125mile/h Pendolino timetable should be clearer. By then, too, VRG will be dealing with the Department of Transport's new Rail Directorate.

It will be a baptism of fire for the as yet to be named Head of Procurement in the Directorate who is separated from the rest of DfT Rail by a Chinese wall. He or she may wonder why the letter agreement of July 2002, which rescued the two Virgin franchises thought that West Coast could be renegotiated by April 2003. Some of us wondered just that at the time.

He or she may also wonder why the Letter Agreement allowed for the Cross Country franchise to be terminated in the event of failure to agree new terms while West Coast just goes on being run under a management fee. The answer is, of course, Virgin's WMD, the Tom Winsor crafted Passenger Upgrade 2 contract which still commits Network Rail to provide the 140mile/h railway Railtrack signed up to.

While PUG 2 might no longer bring Network Rail down, it would cause a lot of collateral damage to the Government. Would Sir Richard Branson use it? I doubt it, he's too establishment. Would Stagecoach, his partners in VRG, unleash hell? Look into the eyes of Brian Souter and his rail enforcer Graham Eccles and ask yourself ‘are you feeling lucky'?

FOD

Starting point for the negotiating positions will be the operating performance of VWC come September. Pace Sir Robert Reid, if your engineering and operations are not right, then you haven't got a viable railway. What the West Coast needs to recover is, above all, reliability.

After the ‘Red Revolution' started on 29 September last year, Alstom worked wonders to better the minimum of 30 Pendolini ready for service each morning. But from December 15, the requirement rose to 42, which is not proving easy to meet.

Track workers leaving large metal bits lying in the four foot after possessions has not helped. Trains, as well as planes, can suffer from what the RAF calls FOD (Foreign Object Damage).

With one Pendolino already being repaired after such an incident, 390.009 hit a Kango hammer near Preston at 110 mile/h. The Kango bounced up and down under the train, puncturing under-floor equipment cases, breaking roll bars and damaging three bogies. It then popped out of the side, bounced back of an Overhead line support mast and embedded itself in the bodyside of the train. Scratch another badly needed Pendolino for three months.

June challenge.

According to Informed Sources, a sustained 42 sets everyday was unlikely before March. And that is only an interim stage to Alstom's ultimate challenge – meeting the full 46 diagrams from 53 trains service in June. If this is achieved, and it is a big ask, then an even bigger task is making the trains reliable.

VWC's delays are currently about 15% worse year on year, thanks to a combination of Pendolinos still on the steep part of the bathtub curve while the residual loco hauled stock, which should have been out of service by now, swirls round the plug hole. The next software fix ought to bring Pendolino miles per casualty up to 5000, (from around 3,000) but this is still only half of what the best InterCity fleets are achieving.

So the first uncertainty for the negotiators is the speed with which Alstom can have the full fleet working with adequate reliability. In September, they will still have the results of only three months at most of full Pendolino operation from which to extrapolate business plans.

Infrastructure.

But the decline in InterCity traffic on the West Coast since 2002 is predominantly about infrastructure rather than train performance. As Fig 1 shows, passenger km never recovered from the Hatfield effect, while ridership has dived in the last two years.

This is what you would expect on a route which experienced on-going 54 hour possessions at weekends on top of blockades lasting weeks as the WCRM team played catch-up. As Chris Green points out, VWC has effectively been running a weekdays-only service, with weekend travel not for the faint-hearted. Indeed, the message from SRA has often been not to use the West Coast.

Against this background, to have lost only 10% of passenger journeys since the SRA's big blockages policy came in, is quite an achievement. Passenger km are down by 14% over the same period, with the average journey length decreasing from 194km to 184km. This looks to me like the airlines, low cost and otherwise, pillaging the railway's weakness in the 2hr/300km market which the WCML once dominated.

Now, South of Crewe , there is less scheduled disruption and weekend travel is starting to come back. North of Crewe weekend possessions continue to keep Fraser Eagle profitably employed.

But, note that I said ‘scheduled disruption'. There is a degree of what Charles Atlas used to call ‘dynamic tension' between Virgin and Network Rail over the reliability of the new infrastructure.

Points and counters.

NR60 points with HPSS drives and train detection by axle counters are the two technical issues. Neither have worked as advertised, but Informed Sources suggest that Virgin may be suffering from perception hysteresis.

HPSS, or the High Performance Switch System was one of those whizzo technologies from a firm (which I worked for, once upon a time) with an aviation background. It fell victim to a backdrive which tended to bind.

A major redesign has helped, as has Network Rail's taking maintenance in-house. With better trained and equipped maintainers the failure rate is claimed to be a third of what would be expected from conventional points of the same size and under the same traffic density.

Axle counters are more intractable. Solving one, generic, software problem only revealed that it had been masking another. Solving the second problem gave a 90% reduction in in-service counting errors.

But the real issues remain not the technology but the principle. Where a track circuit physically detects that a set of axles is short circuiting rails, an axle counter tells you only that so many axles have gone into the section and not come out the other end - yet.

This raises sundry philosophical issues, and such issues keep the safety Taliban happily employed for ages.  For a start, after a possession, how do you know the section is empty? After all, if you can forget a Kango hammer you might overlook a couple of ballast wagons and a Class 66, which a track circuit would detect but axle counters know nothing about.

So to ‘reset and restore' the axle counters after a possession, a ‘sweep engine' had to run through. The axle counter counted the axles in, counted them out and if the numbers agreed the signaller could reset and restore.

By the time this appears, signaller reset should be allowed if rail vehicles have not entered the possession. In March this should apply where only one machine, such as a tamper, has been used. After that the aim is to extend reset to multi-vehicle possessions.

But, of course, the same issues apply in the event of a system failure, which means that getting axle counter based signalling up and running again is a slow process. Currently, Network Rail claims that axle counters are now as reliable as the average track circuit, whereas their use was justified, not to say, heralded, on the basis of a major improvement on old technology.

Since I tend to mingle with the crusty old signal engineers, I am not a fan of axle counters and prefer track circuits – which are good enough for TGV. Interestingly on Chiltern, the axle counters put in under Evergreen 1 are being taken out under Evergreen 2.

So huge amounts of FUD on the reliability of tracks and trains. Of course the engineers can get it right: but the commercial performance of Virgin West Coast will depend crucially on how long it takes to get a boringly reliable railway routinely delivering PPMs in the high 80% region.  

Billions.

What sort of numbers will VRG and DfT have on the table in September? More FUD, I'm afraid.

Figure 2

Fig 2 shows Virgin's original subsidy profile plus what really happened, plus some interpolations and extrapolations of my own. In particular, I have added track access charges

Staring from the left you can see that, initially, actual subsidy followed the forecast. When track access charges went up by £150 million in 2001, following the Periodic Review, the subsidy went up by a similar amount.

Now I know the drop in track access charges in 2002/03 looks unlikely, but, since the source of the figures was a reply to a parliamentary question, I checked with DfT whether Hansard had made a mistake. DfT assures me that the figures are correct.

Similarly, the 50% increase in subsidy for 2003/04 is taken from the SRA's annual report. It looks improbable, especially SRA is forecasting the subsidy for 2004/05 at £185 million.

Personally, I believe that railway finances are now subject to Heisenberg's Uncertainty Principle . The very act of analysing them makes them even more impenetrable.

And the SRA always talks about net subsidies, net of all sorts of things, such as Schedule 18.1 adjustments . Thus the VWC subsidy of £185 million in 2004-05 is after you have allowed for payments to VWC from Network Rail for Schedule 8 penalties and compensation for Network Change which I estimate at around £50 million in 2004/05.

An alternative analysis of Fig 1 is that we are looking at a variant of Schrodinger's cat which, since it could be alive or dead but you didn't know which, is both. Thus Schrodinger's VWC franchise costs £185 million and £235 million and perhaps, even, £300 million simultaneously.

Subsidy profiles.

Virgin's original subsidy profile is a reminder that in another time and place all 53 Pendolini were expected(this column excepted) to enter service at 125mile/h in May 2002. This was to fund the £66 million drop in subsidy for 2002/03, turning into a £60 million premium the following year.

With 140mile/h running following in 2005/06, a further £60 million increase in premium was due in 2006/07. Premium would then grow steadily for the rest of the franchise'

Back in this universe, a point to note is that the track access charges shown on the graph are not the full picture. When the Periodic Review came into effect in April 2001 the SRA decided not to pass on to the TOCs the full increase in access charges. Instead, Railtrack, as was, received some of its extra income as direct grants.

When the Interim Review of access charges burst the SRA's budget, putting up access charges again from April 2004, it was decided to hold them at roughly existing levels for the first two years of the new control Period and make up some of the difference through increased grants.

Thus, in 2004/05 Network Rail is receiving £1,710 million in grants in lieu of access charges. To put it another way, the TOCs are receiving a covert additional subsidy of £1,710 million on top of the over £2 billion.

Compared with passenger revenue of around £325 million, VWC's fixed track access charges alone for 2004/05 are £469,038,488, rising to roundly £582 million annually for the next two years. For the final two years when, presumably, all the enhancements, such as Trent Valley quadrupling are finished, there is another rise to £682million a year. To this should be added the variable charges based on cost per vehicle per mile for Pendolino which are the subject of my first Freedom Of Information Act challenge.

While track access charges dominate the costs of the franchise, that is not the fault of the franchise. And, anyway, since train operators are indemnified against changes in access charges, we can focus on operating costs and revenues.

In January VRG began a management purge in the two franchises aimed at improving financial control and shortening chains of command. This should bring costs down.

Traffic returning.

Meanwhile, there are clear signs that traffic is returning south of Crewe . Revenue for the latest accounting Period is up by just under 20% on 2003-04. Since VWC is running more trains and Rio has ended, some of this, say 3-4% comes from the ORCATS system which allocates revenue on shared routes.

Volume is up, too, year-on-year, by just over 4%, with First Class featuring strongly at around 7%. Train mileage went up in December which should also generate growth.

All good positive stuff. But a lot depends on what you mean by growth. Having driven 1.5 million passengers away since 2001, getting then back is not true growth, merely recovery.

If you look at passenger traffic across the network, Period 11 (effectively January) is 7% up on Period 11 2003-04. For the WCML, the gain for all operators is 12% while VWC, as already mentioned, is up 19%, nearly all this increase coming since the new September 2004 timetable.

However, if you include Rio in last year's figures, growth on London-Manchester services is, according to Virgin, 29% up period on period. And, anecdotally, Chris Green is meeting Manchester business men on trains who say that they have returned from air.

But, while encouraging, these figures should be treated with caution. Although their recovery from Hatfield has been slow, the other InterCity TOCs have seen a steady increase in ridership.

Figure 3 uses this to extrapolate where VWC should have been had it kept pace with the average growth of its peers. Just to catch up woulkd require 40% more volume.

Figure 3

Future cost?

Finally, back to Figure 2 again and those two dotted lines stretching out to 2012. The bottom line starts at the subsidy for 2004-05 and extrapolates the effects of cost savings plus revenue growth assumed to be in line with the original subsidy profile after the Pendolino fleet was in service.

Now remember, that because of its shortcomings Network Rail paid VWC around £50 million in penalties in 2004-05. In theory, on a WCML where everything went right all the time, these penalty payments would vanish and Virgin would make up the loss from the extra revenue attracted by the more reliable service.

Hmm, so the upper extrapolation assumes that the £50 million is bunce and that if it is not paid, VWC will need extra subsidy. And, or course, the eventual slope of my dotted lines will depend on the future performance of track and trains, not to mention the economy.

Somewhere between these extrapolations there is a deal to be struck between civil servants and the partners in VWC, Virgin and Stagecoach. They'll need strong nerves at DfT Rail and I reckon it could take up to a year to reach an agreement.

Continues.........

Return to Archive -by date - by topic.