Freight shipping and greenhouse gases


in Economics, The environment, Travel

Travelling 100km by car produces about 10.8kg per person of carbon dioxide (assuming an average of 1.5 passengers per car). Doing the same by bus produces about 1.3kg, while taking a modern electric train produces about 1.5kg (based on the energy balance in the UK). What is remarkable is that shipping freight by truck produces 180 grams of CO2 per kilometre, while doing so by train produces just 15. Clearly, switching freight transport modes offers considerable scope for emission reductions (as does reducing the total amount of freight shipped).

When you factor in how much damage heavy trucks do to roads – as well as the expense and carbon emissions involved in rebuilding them – it seems pretty clear that disincentives to ship freight by road make sense. Yet another externality that road pricing and carbon taxes could help address.

Report a typo or inaccuracy

{ 9 comments… read them below or add one }

Neal August 11, 2007 at 10:03 pm

One simple way to discourage freight on roads would be tie all funding for road construction to tolls and fuel taxes, as in Europe. Also, North America needs to wake up and immediately start investing in rail infrastructure, which has basically been left to rot over the last 50 years.

tristan August 12, 2007 at 12:21 pm

I don’t think disincentives are neccesary. What we need is the removal of the huge Incentives in the form of subsidies to trucks. What they cost to insure and drive on the road is not proportional to the damage they do. Private rail simply cannot compete with what is effectively public trucking.

tristan August 12, 2007 at 12:27 pm

I’m a bit confused at this figure you have of “10.3kg per 100km”. Maybe my math is a bit off.

To adjust from 1.5 persons to 1 person, I multiplied 10.8 by 1.5 and got 16.2. I then divided by the kgCO2/literpetrol (2.6 or so) and get 6.2. So, this seems to assume that the “average” car gets 6.2 liters per 100km. That’s 37 US mpg and 45 UK (the kind that gets posted in canada) mpg.

This might be true in Europe where everyone drives superminis because gas taxes actually reflect the cost of the roads. However, I think on this continent 9 liters/100km, or even 10, would be a more accurate average considering how many people drive trucks and minivans and SUVs, many of which would be lucky to get 12 or 13 even on the highway. So, that’s more like 15kg per 100km per person.

tristan August 12, 2007 at 12:28 pm

Also, if these are UK numbers, trucks are much more heavily regulated in the UK and have smaller engines. Thus, it is likely that the CO2 per KM for Canadian and American trucking is much higher. Especially Canadian, because we don’t even have seperate lower speed limits for trucks as they do on the interstates.

Milan August 12, 2007 at 12:32 pm


Those figures are all straight from Monbiot’s book. Oddly, he expresses them in terms of emissions produced driving from London to Manchester. I just converted that distance into kilometres.

tris August 12, 2007 at 2:53 pm

Still, probably good not to take any numbers as given before converting them into numbers which can be compared to direct experience.

What really irks me is the trend of expressing CO2 per kilometer, or as anything except as per liter of fuel. I have never seen the C02 per liter of petrol quoted, I’ve had to search it out whenever I forget it. It is stupid to think that people can’t think in terms of liters of petrol, for God’s sake – they pay for it by the liter! People know how many liters they use!

Maybe we could lobby to have, alongside “liters” and “dollars” at the pump, “kg of C02”?

Sarah August 13, 2007 at 8:39 pm

Alas, heavy freight trains do a great deal of damage to railway tracks which means that the tracks need a great deal more maintenance than they otherwise would. This suggests that a) different tracks would be used for freight and passenger trains, which would necessitate the building of new railways and hugely increase the greenhouse gas emissions of freight rail, or b) continuous repair & replacement of the tracks used for freight which would make it difficult to run passenger trains without causing disturbance.
As in other walks of life, there is a trade-off to be made: good passenger railways that attract people away from car usage (requiring fast-ish, regular trains every day including late nights) versus greater freight carriage on existing railways.

There is also the (fairly obvious) point that rail freight will generally require other modes of transport to be used as well, such as road or water, and that the need to direct freight traffic around terminals for moving between road or water and rail may notably increase overall distances travelled.

All this is not to say that freight rail isn’t a good idea – just that it may be substantially less attractive than your figures suggest.

Milan August 13, 2007 at 8:57 pm


Good points. Thanks for the information.

. November 4, 2009 at 4:18 pm

Carbon Pricing Will Help Warren Buffett Get Rich From Investing in Railroads

By Matthew E. Kahn

Warren Buffett’s Berkshire Hathaway Inc (BRKa.N) (BRKb.N) will pay $26 billion to buy out railroad Burlington Northern Santa Fe Corp (BNI.N) in what the billionaire investor called a bet on the U.S. economy. Did you know that freight trains have a fuel economy of 400 MPG? That’s better than a freight truck. You don’t have to be Jimmy Hoffa to anticipate that Mr. Buffett’s bet on shipping logistics will be more likely to payoff in a carbon constrained world where carbon is priced. Here is the story. Now for cities such as Detroit and Cleveland that have ample railroad network capability — will there be beneficial spillover effects? Are you getting ready to buy Detroit Real estate to get rich along with Warren Buffett? I talk about these points much more in Climatopolis just wait until it appears as a Basic Books publication some time next year.

Leave a Comment

You can use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Previous post:

Next post: