Return to Archive -by date - by topic.
With standard designs established, bulk buying has brought down the cost of multiple units – but not as much as I thought
Oops! Wrestling some recalcitrant boiling frogs a few months back, I wondered whether the Ford Factor also applied to traction and rolling stock. Intuitively it didn't.
In the latter days of British Rail, it was a given that a new Networker electric multiple unit cost £1million a vehicle in early 1990s money. Some quick correction for deflation, and assuming £800,000 per vehicle for Desirostars, suggested that today's prices were some 40% lower in good King Green's golden days.
So that's progress. I mentioned it somewhere in this column to show that I am not totally negative and forgot about it until a chum challenged a statement of mine on BR funding. Resolving the issue meant going back to an article I wrote in 1993.
And there I found a chart I had forgotten showing BR multiple unit prices. ‘That ought to be in the GRICER Database', I thought and at the next opportunity keyed it in. But it never stops there, of course, and I just had to bring it up to date.
My chart shows the result. It lists the cost of pretty well every multiple unit in approximate chronological order up to the present day. The red line is the rolling average.
Does it matter, with the railways finances collapsing around our heads? I think so, because it provides both a sanity check and the odd awful warning.
As you can see, it starts in the 1980s with the first diesel multiple units to be bought in competition with BREL – BR's in house train builder. You will see that the private sector prices are in line with those of the state builder. This is not coincidence.
With the Class 156, Metro-Cammell knew they had to match the uncommercial costs of the nationalised industry. So Met-Camm and its sub-contractors took cuts on their prices to buy their way into the business.
But as you can see, up to the aluminium-bodied Class 158, prices remained steady and the average for the 12 orders up to and including the Class 456 was £613,000 per vehicle. These prices include variation orders.
As a regular traveller in bog standard and refurbished Class 317s, I think these 20m long variants on the Mk 3 body shell were exceptional value for money. York just kept churning them out and the specification gradually improved. And when the refurbished Class 317s came onto WAGN, I overheard passengers comment that they were new.
Enter Chris Green from the north, cue transformation scene from London & South East into Network SouthEast with red lamp posts. And, of course, Networkers.
Here was a new generation super-commuter EMU with every high tech gizmo Green and his engineers could think of. Aluminium bodies, plug doors, three phase drive, slick interiors with no visible means of fastening, and so on.
By now BREL was being privatised and with commercial reality breaking in, the high spec/high tech Networkers came with an appropriate price tag. Which then increased with subsequent orders.
But note the dip in that rising curve. Gordon Pettitt, the wily old head of Regional Railways, wanted new trains for his electric services around Birmingham , but didn't want to pay Networker prices.
Fortuitously, at least in the short term, a group of 13 managers and engineers had left Met-Camm and set up as train builders under the Hunslet banner. Since old Met-Camm was all about gritty value for money and quality rather than flash, the Class 332 was high tech (the first three phase drive with a 25kV AC safety case) but plain vanilla.
As the graph shows, Gordon got what he wanted, but then the 1064 day hiatus started. Which is where we come to the ‘oops' which opened this piece.
At £1.3million a vehicle, the Class 471 Networker – an express unit for Kent outers, was the most expensive of the breed. Or would have been, had it been built. So, if we stick with trains in the metal, the average price of the Networker era was indeed a touch over £1million a vehicle, but in today's money.
Now we come to the initial privatisation era, in which you will note some very expensive trains and some weird pricing.
Remembering that National Express was expecting to pay £400,000 a vehicle for new DMUs for its franchises, the first post privatisation order which broke the 1064 day hiatus – Adrian Shooter's Class 168 DMUs came in at £980,000.
lass 170s were even more expensive although, as you can see, prices tend to yo-yo up and down depending on specification. Alstom prices are puzzling. On the published figure a Class 175 costs the same as a 125mile/h high spec Class 180.
For 125mile/h DMU prices, the most consistent guide is Bombardier where the Virgin fleet is slightly more expensive than the Class 222, as one would expect given the tilt capability in the Class 221.
But the ultimate contrast is the two airport express EMUs. Heathrow Express is the most expensive EMU ever at £1.5m per vehicle. Siemens point out that it is a very high specification train including a fire safety rating for tunnel running, fancy passenger information and is fitted with ATP – which can add a quarter of a million to the cost of a unit.
Which all makes sense, but leaves you wondering how Alstom built the Gatwick Express Class 480 units, (20metres to the Heathrow's 23), for £755,000 per car – and I doubled checked both prices.
However, as you can see, when we get back to meat and potatoes EMUs the average price of Desiros, Electrostars and Junipers, is £850,000 per car.
Era cost/car £
BR York 31X 615,000
Networkers 1,060,000
21 st Century 850,000
So, Networkers were 73% more expensive than their immediate predecessors and the new generation EMUs are 22% cheaper than the Networkers. But, today's trains are 40% more expensive per car than their equivalents of the 1980s.
This is probably a better comparison for one important reason. We now have some proper production runs again. Networkers were built in dribs and drabs, even the Class 465 fleet was split, with 200 vehicles each from two manufacturers. Contrast that with production runs of 700 or more for Mk 1 stock replacement.
Compared to the EMUs of the 1980s, today's trains have a higher specification all round, with air conditioning, three phase drive, train management systems, more demanding crashworthiness requirements and better disability provision. So not cheap – but good long term value, in my judgement.
But the big lesson from the graph is that the railway only gets keen prices when it provides stable long term demand. Which is what privatisation was supposed to deliver by transferring railway investment to the private sector.
Mark 1 stock replacement has provided massive demand, but, alas, it now turns out to have been short term. I suppose I should have spotted this in all the euphoria.
Indeed, this column did point out, that until the Mk 1 replacement orders came along, EMU order rates were the same pre and post privatisation. And in a perfect world Network Southeast would have carried on building 400 Networkers a year throughout the 1990s, meeting the Hidden Inquiry assumption that all Mk 1 vehicles would have gone by the end of 1999.
So the current mega-orders are not a triumph of privatisation but the result of realpolitik. And while they included some growth builds, there are already second thoughts over the Class 450/2 five car Desiros for SWT .
As last month's column showed (‘Roger's excellent analysis' – Bowker R. ‘For an engineer, you make a good accountant' – Muir G. Worrying eh?) the railway is close to financial meltdown. With so much money going into keeping Network Rail afloat and propping up failed franchises, new trains are once again off the agenda.
A year ago the Rolling Stock Companies were talking of a further 4,000 vehicles to be ordered by the end of the 10 Year Transport plan in 2011. Now they are much more sanguine.
Questioned by the Parliamentary Transport Committee on 26 February, Angel Trains Managing Director Haydn Abbott said his company did not believe that capacity enhancing schemes including Thameslink 2000 and Crossrail would materialise. Angel's best estimate was that ‘only 700 to 800 additional new vehicles would be purchased up to 2010'. This compared with an estimate 18 months ago of 4,000 new vehicles to be ordered between 2002 and 2010.
Peter Aldridge, MD of HSBC Rail gave similar figures. He had been looking for 3,500-4,000 vehicles over the rest of the current decade. Now the expectation was that only 1,000 vehicles would be ordered ‘over the next nine or 10 years'. Porterbrook ‘don't disagree' with these numbers.
So the expected demand is back to 100 vehicles a year. Which means small fleets, short production runs and high prices. It's back to the early 1990s with a vengeance.