Wednesday, May 14, 2008

Abandon the railways or just the facts?

The former editor of NZ Trucking Magazine Jon Addison has written in the NZ Herald proposing that the rail network be ripped up and the corridors used to build dedicated trucking roads. I don't agree, I think the cost of that is prohibitive in itself and much of the rail network is too narrow to make that worthwhile (don't forget most railway lines are single track, which means one way roads!). I also think that some railway lines are economically viable (2 hourly freight trains on the main trunk line isn't a bad use of that corridor).
However, he has made some very good points that those who worship rail as a religion, than simply a technology, often ignore:
"New Zealand's rail network is constrained by more than its ageing bridges and locomotives. Its 3ft 6in (1067mm) narrow gauge tracks limit the speeds at which trains can operate and its 150 tunnels are too small to accommodate double-stacked containers, which have boosted rail efficiency overseas."
"Taking the last of these first, the introduction of Euro 5 emissions standards in Europe this October and inevitably eventually in New Zealand will mean that in some urban environments the truck exhaust will be cleaner than the air entering its engine. Truck noise levels have also reduced significantly.
While the fuel used by an efficient train will be less than that used by trucks carrying the same weight, this ignores the fact that freight is invariably carted by trucks at one end of the train trip and often at both ends. And at the transition, fuel is used by forklifts or container cranes and increasingly used to maintain the temperature of freight while its waiting to be moved.
As far as driver efficiency is concerned, change is on the way. Most of the major automotive manufacturers are working towards driverless vehicles and some are predicting that they are less then 20 years away. All of the technology exists now, and some of it is already appearing in production vehicles.
Interesting stuff indeed. He concludes by suggesting that maybe converting the rail network to roads wouldn't add up, but it is worth investigating. Perhaps so. I am more convinced it could be worthwhile using the rail corridors in cities for roads, but in Wellington and Auckland too much money has been poured into passenger rail for this to be worth considering for now.
Meanwhile, Keith Ng in the NZ Herald has a column called "Just the Facts" where this week he challenges Richard Prebble's assertion that "it was a myth to say rail was environmentally friendly if the production of rail, locomotives and the need for trucks to take goods to destination were counted". He claims an EECA study is the most authoritative, because it claims the energy intensity of road vs rail is a factor of around 5 to 1. What he doesn't say is the basis for this comparison, because the factors vary wildly. For example, a truckload of freight from Wellington to Petone by rail will burn a lot more fuel per tonne km, than a trainload consigned from Wellington to Auckland.
but then he goes to Chris Kissling of Lincoln University. Yes, the same one I fisked a month ago for advocating bullet trains in NZ, "smart clothes" that automatically open doors and "steer people around hazardous places" and that the future of flying is that passengers will be drugged and stacked horizontally on beds! Kissling presumably supplies him with a European study about environmental costs, which makes rail look good, failing to point out to Keith the most recent comprehensive New Zealand study directing comparing rail and road freight environment costs -
I've quoted it before, the Ministry of Transport's Surface Transport Costs and Charges study.
It contradicts what Chris Kissling and Keith Ng says, showing that Keith isn't showing "just the facts", since he ignored one of the most authoritative studies. What did it say? Well it compared the environmental impacts of freight between Wellington and Auckland, Napier and Gisborne and Kinleith and Tauranga. The comparison is as follows:
Environmental costs per net tonne km in NZ$
Wellington-Auckland rail NZ$0.008, road NZ$0.006
Napier-Gisborne rail NZ$0.002, road NZ$0.002
Kinleith-Tauranga rail NZ$0.001, road NZ$0.004
So in other words, on average it is more environmentally friendly to send freight by road between Wellington and Auckland than by rail, but the opposite between Kinleith and Tauranga. In which case, Richard Prebble is pretty much right.


sean14 said...

Hi Scott

In the House yesterday, Mallard was chucking around a figure of $400 million to bring the tracks up to "the sort of standard that is necessary for a half-modern railway".

I have seen a report that costed double-tracking between I think Pukerua Bay and Waikanae at between $300-$500 million (and admittedly there would seem to be some major engineering challenges on that terrain). Given that is just one small piece of the national network, is Mallard in fantasyland?

Cheers, Sean.

libertyscott said...

He's probably not far wrong to be fair. The double tracking you mean involves a lot of new tunnelling through very active geology, so is expensive (and not worth it). $400m would replace sleepers, track and the most needy bridges.

That is actually the bigger issue - the bridges that need replacement which in a few cases are seriously expensive. However, don't forget the recent purchase doesn't affect this - the track liability was bought for $1 a few years ago!

sean14 said...

Thanks Scott.

Keith said...

Hi Scott,

I can't find a copy of the actual report (MoT's site is full of broken links), so I'm just going to have to do some back-of-the-napkin calculations, sorry.

The Q&A you linked to notes that the total environmental costs of trucks in 01/02 was $492m ($195m of it was LCV). The environmental cost of freight trains was $8.5m.

Let's exclude LCVs - which do short distances and are arguably more polluting - so we're comparing apples with apples. Trucks (excluding LCVs) are not used for anything *but* freight, so we can compare freight trains with trucks minus LCVs. So that's $8.5m vs $297m.

That's a ratio of 35:1.

As a very rough benchmark, in 98, according to the EECA's Dynamics of Energy Efficiency Trends report (see my Public Address post for the links), 62% of freight in NZ was moved by roads, 12% by rail. And that counts LCVs, so if we exclude LCVs, that ratio would be even lower (probably a lot lower, actually).

Unless the road-rail ratio changed radically between 98 and 02, we can safely say that the STCC data shows that the average environmental cost per unit of freight transported is *much* higher for trucks than for trains.

Surely, you have to reconcile the data you quoted with this fact.

(If the STCC has a breakdown of total t-km transported by vehicle type, it would be good to see it.)

Obviously, rail is not going to be superior in every conceivable situation, but on average, it's more efficient by a very significant margin.

Combustion engines are far more efficient when they're run at constant speeds with constant loads. Therefore, a train designed to run on a low-gradient, single-purpose track at a constant speed will virtually always be more efficient than a truck that has to navigate corners, slopes, traffic and other road hazards, using constantly changing directions, using mechanical brake and running the diesel engine at wildly variable speeds.

Electric engines (serial-hybrids in particular) might even the playing field, though. If you're interested, a Kiwi called Richard Wright is building electric supercars in the US, with the intention of adapting the technology in LCVs and trucks. (Probably the better part of a decade away, though.)

libertyscott said...

Keith, apologies for the link, it USED to go to a full copy of the report, which I have. You ought to ask MOT to send you a copy. The study is being updated, so the information from that should be fascinating.

The figures you quote from the study are correct. However, MCVs also do not compete with rail and you need to exclude another $145.9m environmental costs, to leave HGVs.

I question EECA's figure, because what does % of freight moved mean, and how much double counting exists. e.g. freight on the ferries is actually a link in a road or rail trip. EECA was never very good at stats on this, usually relying on MOT! STCC has a total rail freight value of about 3.7billion net tonne km, but no figure for road. I doubt if, using your EECA figure, 26% of NZ freight goes by sea and air. Air would be in the 1% by weight category, and 25% by sea would be extraordinary on a NTK basis. Note that official stats for transport are far from particularly reliable for largely private sector activity, the best source of data for trucking is actually RUC purchases (which includes weight, distance and can include a factor for evasion).

Environmental costs are a matter of exposure and impact. Trucks today are far cleaner than your average diesel locomotive (new locomotives would make a difference though). I disagree about trains running at constant speeds, as almost all track in NZ is single track and that means stopping at signals for passing tracks several times for any given trip. Don't forget part of the costs of rail are maintaining a very lightly used infrastructure.

STCC still did a fuller look at marginal costs of road vs rail freight. A study specifying marginal costs (I know two of the people who spent a good part of 2 years of their lives doing this) is far better than a "back of the envelope" calculation with dubious data from EECA, which frankly is shallow on this sort of stuff.

STCC says that while for line haul rail is more efficient than road, the local handling impact significantly raises costs and puts it a lot closer to road.

Here is part of the STCC study conclusions on road vs rail freight "One finding of interest is that the environmental externality marginal cost component tends
to be relatively small, and of similar magnitude on the two modes. For long-distance freight
this tends to be true because the routes pass through areas which are primarily rural and
where minimal LAQ damage occurs due to the population density of the receiving

It continues that the highest impacts are in urban centres, and both road and rail impose these costs. One reason why rail has lower average environmental costs is because most rail haulage is long haul, whereas much road haulage is short haul and urban. Road freight exposes more people to environmental costs because it is the mode used for short to medium trips. On long haul road and rail freight have similar environmental impacts.

I hope that helps, I'm keen to have robust debate on these things.

Keith said...

Hey Scott, apologies for abandoning the thread. It was really interesting, unfortunately, things have been pretty manic since the Budget. (And as a journalist, I'm professionally obliged to have the attention-span of a gnat.)

Are you keen to have a chat (on or off the record) next time a transport issue comes up? Drop me an email via,

Anonymous said...

It's really fucking simple.

perhaps trains made sense in NZ in 1900.
They don't in 2000.

People fly and drive; trucks drive.

Trains are completely pointless in NZ. Sell 'em for scrap, throw the lot into the sea, fire all the drivers (or send 'em to Aussie) and execute all the union reps!

Why? Because NZ businesss really need the $3 billion that governments will spend every few fucking years on rail back to their own profits.

Because it takes a Kiwi Bludging worker 90 minutes to do the work an Aussie does in 60 - and spending anything at all on trains only makes it worse,