Economics of Total Electrification

[HSTEd]

Concept and Capital costs

It occurred to me recently that, based rather simplistically on the costs of the Manchester-NW electrification, the electrification of the entire unelectrified mainland UK network (some 12000 track miles) would total roughly £18bn.

Although there are parts of the network likely to be far more challenging than the routes tackled in this example project, there are also areas that will be able to use simplified overhead equipment so I conclude that on average the cost listed above is reasonable.

This however is not the only cost associated with the electrification, new stock must be purchased.
If we assume that all cascaded stock from current projects such as the Thameslink and Crossrail programmes will be used primarily for strengthening (either directly or by further cascade) then we come up with a rather large order.

I conservatively estimate it to be, using the minimum of unit types to ensure very large orders and maximum fleet homogeneity:

746 3-car suburban/regional units
317 5-car "express" units
110 5-car electric IEPs.

I assume that all of the non IEP units would have gangways, and that the regional units would have 100mph speed limits and the "express" units would be capable of 110mph.

In generating this estimate I assumed that major growth in capacity is necessary and that the Voyagers used on the WCML routes could be replaced with 110mph units without undue damage to the schedule, furthermore I assumed that 100mph and 110mph units were sufficient for Thames valley services. (220s were replaced with "express units", 221s with IEPs)

The latter assumption is a bit 'dodgy' and so this estimate may have to be modified significantly to account for this.

If we assume a purchase cost for the non IEP vehicles of £1.5m and for the IEP vehicles of £2.4m (in line with DfT estimates) we get a total bill of £6695m for rolling stock, so ~£6.7bn.
I will assume that state money is used to purchase the rolling stock outright rather than through a ROSCO.
This brings the total capital investment to ~£24.7bn.

Assuming the use of a 35 year index linked gilt issue (yield of -0.01%) to fund the project, this would result in annual repayments over the 30 years of approximately £703.2m (in real terms, this will obviously increase with inflation as do most other railway costs and revenues).

Leasing Costs

There are currently 2892 DMU vehicles in service in the UK and roughly 225 Mark 3 trailers that will be retained in the aftermath of the IEP project for a total of 3117 vehicles.
According to Network rail the leasing costs (excluding fuel and maintenance) for a DMU vehicle is roughly £110k/year. If this is assumed to be accurate over the entire fleet and if it a reasonable estimate for HST/loco hauled Mark 3 vehicles then the leasing costs come to £342.8m/yr.
As the state owned vehicles envisaged by the plan would have no leasing costs this entire figure can be chalked up as savings.

This reduces the excess costs to approximately £360.4m/yr.

Fuel Costs

According to the DECC the Railway industry expends roughly 682,000t of oil per year, effectively all in the form of diesel.
Using standard values for the density of diesel (0.832kg/litre) this equates to roughly 819.7m litres of diesel fuel.
Current prices for red diesel are around the 71p/litre mark which translates to a fuel cost of approximately £582m/yr.
However, electricity has a non zero cost so this must be accounted for, a litre of diesel burned in a modern rail diesel engine has a useful energy value of approximately 4.18kWh/litre, once a 20% efficiency for the use of regenerative braking is included this reduces to approximately 3.5kWh/litre equivalent.
This means that the electricity consumption fo the network would be approximately 2.87bn kWh/yr. Using the typical planning price assumed by Network rail (roughly 6.5p/kWh) this translates to approximately £186.5m/yr.

This means the overall saving in fuel costs is approximately £395.5m/yr.

This reduces the excess costs to approximately -£35.1m/yr.

Maintenance Costs

According to Network Rail there are approximately 40bn tonne miles of passenger traffic on the network every year, using 40t as an estimate of the average vehicle weight of the current fleet this translates to roughly 1bn vehicle miles.
~40% of this is operated by diesel traction, equating to roughly 400m vehicle miles.
Network Rail estimates that an electric vehicle is roughly 20p/mile less expensive to maintain, translating to roughly £80m in savings per year.

The overhead line equipment also requires maintenance however.
Estimates of this are difficult to come by, with some values varying from 1-5% of the original investment cost of the equipment, however these generate very large values and I have come to the conclusion that it is roughly 1% of the cost of the equipment itself (note that the equipment being installed by NR is designed to generate low maintenance costs even at the expense of greater capital costs).
The equipment itself comes to approximately ~35% of the cost of the installation, which equates to approximately £6.3bn in this case.
1% of this value is approximately £63m per annum.

This means that savings in maintenance costs are roughly £17m/yr overall.
This brings the excess cost per annum in roughly -£52.1m/yr.

Other Factors

There are other factors relating to increased revenue from the larger number of vehicles available for service strengthening in this proposal, possible factors relating to the price of carbon in the future and the possible "sparks effect" resulting in increased patronage from the more desirable product offered by modern electric trains compared to older diesel ones.

As these factors are extremely difficult to model with the resources available I simply do not attempt to do so, in any case it is highly likely that these factors will improve the case further and this is unnecessary in light of the already positive business case outlined above.

Additionally it may be noted that I have not really taken note of rail freight in this assessment, as it is a rounding error of traffic on the railways, especially in fuel consumption terms, it is considered unwise to take too much notice of it.
However it is likely given complete electrification that the freight operators would shift to electric traction, especially as extremely high power CoCo freight locomotives now seem practical (powers approaching 10,000hp now appear within reach).

In summary, complete electrification of the railway and purchase of huge amounts of new rolling stock appears to lead to a net benefit to the state.
This is without considering nebulous factors related to carbon pricing and the "sparks effect".

 

[eMeS]

If you have an all electric network, how are you going to rescue stranded trains and their occupants when the wires come down? I assume it's got to be diesel.
So, how are you going to justify keeping available the expertise to service, properly, the rescue locos? Surely the more aspirational staff and engineers will migrate to where the action is, the electric locos and EMUs; leaving lesser grade staff to keep the emergency vehicles in ready-to-go order.

 

[HSTEd]

If you have an all electric network, how are you going to rescue stranded trains and their occupants when the wires come down? I assume it's got to be diesel.

Well IEP units will be able to self rescue as all of them will have at least some diesel plant available, as to the other units they will have to be rescued by diesels (unless an IEP proves able to haul itself and another unit at low speed under diesel power).
However the number of diesels required for certainly far fewer than the number of diesel locomotives on the network today.
They will also run far fewer miles than diesels do today with the associated reduced maintenance requirement.
They would probably have some use hauling engineering trains during isolations on possessed track systems, which will enable traction knowledge to be kept up relatively easily.

So, how are you going to justify keeping available the expertise to service, properly, the rescue locos? Surely the more aspirational staff and engineers will migrate to where the action is, the electric locos and EMUs; leaving lesser grade staff to keep the emergency vehicles in ready-to-go order.

The issue with low staff morale in the diesel locomotive units could be handled with pay differentials, requiring all drivers to sign the diesels or perhaps by more esoteric methods such as providing a standard "thunderbird" livery to the locomotives, perhaps with something like "National Rail Thunderbirds - TO THE RESCUE" emblazoned on the side.

 

[jopsuk]

the channel tunnel operation manages to maintain a small diesel fleet used almost exclusively for rescue and maintenance.

In all-electric, diesels would be retained for rescue, but also for engineering trains- would would imagine that the engineering fleet would in fact supply the rescue services as well. There would also need to be diesel shunters in various yards- loading and unloading containers is tricky with OHL, as is loading hopper wagons.

Your suggestion that all-electric would leave a low-skilled, low-motivation workforce working on the remaining diesels is rather something of a desperate strawman eMeS

 

[HSTEd]

In all-electric, diesels would be retained for rescue, but also for engineering trains- would would imagine that the engineering fleet would in fact supply the rescue services as well. There would also need to be diesel shunters in various yards- loading and unloading containers is tricky with OHL, as is loading hopper wagons.

Well Furrer and Frey have done considerable work on the loading of containers and hopper wagons under OLE, and it now appears that shunters will be unnecessary.

Overhead contact rails that can be moved out of position easily have been developed. Although primarily used in depots at the present time it appears they could easily be used to allow electric locomotives to pass through a loading dock and then moved out of the way to permit loading.
Especially if a barrier wagon was provided in the formation.

The provision of a section of moving conductor rail is undoubtedly cheaper than the provision of a shunter, especially once the existing fleet of Class 08s expires.

Additionally most engineering workings such as rail grinders (especially if high speed grinding was adopted) and RHTT could be operated by a single electric with an MPV style driving trailer being provided on the far end of the rake, the fact that most electrics now have two completely independent traction groups and two pantographs would provide the required reliability.

 

[Bald Rick]

Interesting.

I may have missed it, but what assumptions are there for transmission losses in the power distribution system and the power loss through the trains themselves?

Also, could UK plc get a better result from, say, electrifying the busiest 60-80% of the network and leaving the rest? I can't imagine there would ever be a case for electrifyong the far north line or central Wales line for example.

For OLE maintenance costs, I think you need to multiply your number by 2.5, based on some old data I've got.

 

[Waverley125]

As has been said, while there are good net costs for the entire network, that will be comprised of some lines with very high benefits (i.e. trunk express passenger/freight lines) and some with costs (Far North, West Highland line)

The latter will get electrified only when battery technology becomes good enough to run a train on those lines solely off battery, without the associated infrastructure costs of OHLE.

 

[HSTEd]

Interesting.

I may have missed it, but what assumptions are there for transmission losses in the power distribution system and the power loss through the trains themselves?

Well losses on the supply side of the substations are included in the price of the delivered electricity, and it would appear that autotransformer type systems reduce the losses on the rail side of the substations rather significantly.

There are also simplifications in this model that overstate the electricity demand (for instance freight workings expend nearly 40% of the fuel used at idle where no traction power is required at all).
And I have concluded that the entire diesel fleet has engines as efficient as the latest locomotives/multiple units.

I also think the 20% figure for regenerative braking is rather conservative and can probably account for some losses as well.

Also, could UK plc get a better result from, say, electrifying the busiest 60-80% of the network and leaving the rest? I can't imagine there would ever be a case for electrifyong the far north line or central Wales line for example.

Perhaps on the face of it, but the principle of an "all electric railway" generates significant political capital and has certain operational benefits that I am sure are clear. (No need for any DMU depots or passenger certified diesels, allowing the existing 66 fleet to take over engineering and rescue operations).
It also allows through operations with no requirement for traction changes or diesel under the wires operation (there are significant freight operations as far as Invergordon on the Far North line for instance, which would be diesel hauled for the entire route otherwise).

This becomes especially true once the Southern Region network has been substantially converted to overhead wiring as the "electric spine" plan seems to indicate.

For OLE maintenance costs, I think you need to multiply your number by 2.5, based on some old data I've got.

Are those figures derived from Mark 3 equipment?
This new Series 2 type equipment has various features to reduce its maintenance costs supposedly, including the use of fancy tensioning springs in place of the traditional weights.
We need a correction of some sort I would think.

The latter will get electrified only when battery technology becomes good enough to run a train on those lines solely off battery, without the associated infrastructure costs of OHLE.

That will never happen, and even if it does the batteries would be more expensive than simply installing OHLE.

 

[Wath Yard]

Two points:

1. The Government doesn't directly pay or raise the cash for electrification. Network Rail does and it can't borrow as cheaply as the Government. It also can't borrow unlimited amounts of money. Network Rail has been deliberately structured to keep its debt off the Government's books.

2. The Government is not going to purchase trains.

There are other problems with your analysis but those two will do for the moment.

 

[HSTEd]

Two points:

1. The Government doesn't directly pay or raise the cash for electrification. Network Rail does and it can't borrow as cheaply as the Government. It also can't borrow unlimited amounts of money. Network Rail has been deliberately structured to keep its debt off the Government's books.

Treasury Money has been used for numerous electrification projects in the past, just because governments of the post privatisation era have not chosen to do so there does not mean they can't if they chose to.

As to the debt off the government books thing, this has absolutely no effect on the bond markets as everyone knows Network rail has certain debts and that the government stands behind them.
All it does it generate sound-bites and produce more money for financiers who agree to go along with the lie.

And as to the effects borrowing £25bn to pay for electrification causing a bond panic... that translates to something on order of eight weeks of current government borrowing.
It is not going to cause any such thing.

2. The Government is not going to purchase trains.

There are other problems with your analysis but those two will do for the moment.

Indeed, the current government, and indeed all recent governments have decided that state ownership of anything is an anathema that must be obliterated in the name of the 'free market'.
But there is no reason they couldn't if they chose to.

 

[Wath Yard]

Oh. It's the old that isn't how things are done and nobody with any influence has ever stated things will be done differently, but I don't agree with it argument. :roll:

 

[HSTEd]

Oh. It's the old that isn't how things are done and nobody with any influence has ever stated things will be done differently, but I don't agree with it argument. :roll:

So I cannot point out what would be possible if the state chose to do it simply because it won't for ideological reasons?

This would reduce the government deficit overall by something on order of £50m and would thus improve its financial position.

 

[yorksrob]

Oh. It's the old that isn't how things are done and nobody with any influence has ever stated things will be done differently, but I don't agree with it argument. :roll:

Well, people clearly have done things differently at times, and I've heard and read reports of a number of interesting debates in political circles questioning the way things are financed (the debates around PFI and spending on infrastructure for growth being two such examples), however don't let that get in the way of your argument

 

[Wath Yard]

So who has seriously suggested they want NR's debt on the Government's books? Osbourne said when he became Chancellor he was relaxed about it but has certainly not done anything to suggest he wants it or is going to make it happen.

Perhaps you could explain how electrifying Habrough - Barton-on Humber is going to boost the economy. When economists and policticians talk about spending on infrastructure for growth it relates to specific, strategic projects (HS2 being one), not ######## £billions up the wall for the sake of it.

 

[yorksrob]

I doubt very much that the Barton Branch will ever be worth electrifying. With that you have a point.

However, to suggest that the Government would never buy or invest directly in trains (or anything else for that matter), whilst being true at the moment, hasn't always been the case in the past, and may not be in the future.

We seem to go through a major change in economic orthodoxy in this country every thirty - forty years or so, so to suggest that the Thatcher-Blair settlement will inform the way that things are paid for for all eternity strikes me as being about as naive as those people who thought that history had ended in 1992.

 

[Wath Yard]

The railway isn't going to be renationalised; through the front door, the back door or the bathroom window.

 

[yorksrob]

The railway isn't going to be renationalised; through the front door, the back door or the bathroom window.

It might not. That doesn't necessarily mean that the Government won't find some way of investing in it directly without resorting to over-leveraged private finance or expensive third party borrowing.

 

[HSTEd]

Perhaps you could explain how electrifying Habrough - Barton-on Humber is going to boost the economy. When economists and policticians talk about spending on infrastructure for growth it relates to specific, strategic projects (HS2 being one), not ######## £billions up the wall for the sake of it.

I make it fifteen kilometres from the junction north of Ulceby (where apparently active freight trackage from Immingham joins the line) with apparently only two overbridges on the entire route.

15km of single track with hardly any over-bridges is going to cost almost nothing.
Trolley wire is probably sufficient for the speeds on the line, you probably do not even require auto-transformers on that section.

You want to maintain a local diesel unit depot with all the attendant logistical difficulties simply to avoid the expense of wiring that branch?

If all we are caring about is avoiding capital expenditure... t he smart move is to rip the whole thing up.

 

[eps200]

this would be a nice project as infrastructure as stimulus, especially as it doesn't just all get spent in London.
Which is another reason it will never happen.