When do fuel costs actually matter?

Kim Stanley Robinson gave a fun talk at Google a couple of years ago in which he brought up the possibility of large, slow, wind powered live-aboard bulk freighters, among other ideas.  I was reminded of it by this post from Alex Steffen.  Especially for commodities like coal, grains and ore — non-perishable goods that get carried in bulk carriers — what matters is the net flux of materials and the predictability of supply.  More (or larger) slow ships can deliver the same flux as fewer high speed ones.  International contracts for these goods can span decades.  If fuel prices became a significant portion of their overall cost, it would be worthwhile to make this kind of ships-for-fuel substitution.  However, it turns out that fuel is a vanishingly small proportion of the overall cost of most internationally traded goods.


Our neighbors in Pasadena moved back to Thailand, and packed their entire household into a single half-sized shipping container.  The cost to get it from their home in SoCal to their home outside Bangkok was $2000.  Their combined airfare was probably a larger fraction of the cost of moving across the Pacific.  You can get a full-sized shipping container moved from point A to point B, anywhere within the global shipping network, for several thousand dollars.  If your cargo is worth significantly more than that, then you don’t have to worry about Peak Oil destroying your business.  For a typical container carrying $500,000 worth of goods, the shipping costs (not all of which are related to fuel!) represent about 1% of the final costs of the goods.  If fuel prices were to go up by a factor of ten, the shipping costs would still only represent 10% of the overall cost.  This would have an effect on business, to be sure, but it would not cause global trade to collapse.

Higher liquid fuel prices will only affect economic activities in which fuel costs make up a significant portion of the overall cost.  Air freight and air travel?  Absolutely.  What about long haul trucking?  At $2/mile to move a truck overland, it costs almost as much to get a container from the port of San Pedro in Los Angeles to Denver as it does to get the same container to San Pedro from Singapore.  Howard Kunstler is fond of making fun of the trans-continental Caesar salad, in which lettuce is brought from SoCal to those of us in colder climates all winter long.  How much would its cost increase if gas were ten times more expensive?  If it were $20 instead of $2 per mile?  Say a head of Romaine lettuce costs $1, and say you can get ten heads into a cubic foot.  Each shipping container is about 2,000 cubic feet, so the total cargo of lettuce would be worth about $20,000… which would be about the fuel cost to get a truck from the Imperial Valley to Colorado, if fuel prices went up by a factor of ten (that’s $20-30 gasoline!).  So to maintain the same profit margin, one would have to roughly double the cost of lettuce in Denver.  People might notice.  In this marginal case, it’s quite possible that we’d see a good market for locally produced winter greens like the ones that Eliot Coleman grows in New England.  But this is a really marginal case.  Lettuce is a very low value density product, and we’re talking about oil that costs $500/bbl or more.  Even much higher fuel costs will only significantly affect such marginal cases.  If anything I suspect that the end of shipping so much ultra-cheap disposable crap will only increase our standard of living.  Well designed durable products cost more up front, but in the long run are usually the better investment, both economically and in the time and hassle of dealing with them when they fail.  The Dark Green Apocalypse of Peak Oil will not crash us into the 19th century.

Ecosense Ecovillage

This is not where you want to live when oil is $250/bbl.

The only place that much higher liquid fuel costs will be starkly obvious to most people in the developed world is personal transportation.  If your household drives 30,000 miles a year at 30 mpg, that’s 1,000 gallons of fuel.  At today’s prices, that makes the cost of fuel a minority of the cost of owning and operating your vehicles.  But you’ll certainly notice if your annual fuel costs go from $3,000 to $30,000.  Jet fuel is around half the cost of short and medium range flights.  If you’re spending $2000/year on flights to visit friends and family, and that suddenly looks like $11,000, you’ll become a big fan of Skype, or decide to live near the people you care about, or wish we had electrified high-speed-rail.  To his credit, Kunstler also heaps scorn upon our suburbs, exurbs, and insane attempts to retrofit city centers to accommodate cars at the expense of walkability.

This will also suck!

For all these reasons I could only sigh and roll my eyes at the Colorado Bioneers conference last fall, where virtually everyone was talking about retreating to the hills and living a subsistence agricultural life, or building eco-villages ten miles outside of town.  Such “solutions”, in a world with much higher transportation costs, would be non-viable or just as desperately isolating as life on 19th century western US homesteads.   If you have to choose among:

  1. being socially and economically isolated on a self-sufficient homestead,
  2. paying tens of thousands of dollars a year for fuel to drive to and from an eco-village where everyone grows their own winter greens, and
  3. living in a dense, walkable city like Groningen where the vast majority of trips are done by bike and on foot, and you can get your winter greens at the farmer’s market…

The choice seems pretty obvious.


This is what a community insulated from oil prices looks like.

If you’re really afraid of peak oil, you should be advocating for high density livable urban development that is optimized for human powered transportation, and which only considers motor vehicles as an afterthought.  If you get the travel bug, you’ll still be able to buy parts from Taiwan and Germany, to build a machine that lets you see the world without burning any oil.

Boulder seems to have a significant amount of political will that’s grounded in the Peak Oil narrative, but for some reason it isn’t being translated into much pro-density policy.  Maybe because there are many fewer people who will come out and vocally oppose supporting local food production than there are that will scream bloody murder when you charge them the true cost of parking.  Which isn’t to say that I’m against preserving agricultural open space.  If you’ve got good farmland nearby and water, why not produce food there?  However, in the context of high fuel prices, the main benefit of preserving that open space is not that it generates food, it’s that it encourages compact urban development by inhibiting sprawl.

Published by

Zane Selvans

A former space explorer, now marooned on a beautiful, dying world.

10 thoughts on “When do fuel costs actually matter?”

  1. I live just a few km south of Groningen and ride there by bike to get to work. It’s a lovely city, very good to live in. However, it’s not actually all that dense. I’ve a graph of population density vs. cycle rate. There are plenty of places with much higher density which have a much smaller percentage of journeys by bike and by foot, but even though the cycling rate of the Netherlands is considerably higher than anywhere else, no Dutch cities are actually particularly dense by world standards.

    It’s not the density that makes the difference so much as the infrastructure design. Make cycling attractive, and people will cycle, even quite long distances – just as they do here in the Netherlands, where it’s not all that unusual even for school children to make a daily 40 km round trip.

  2. Is there any data out there on the distribution of the lengths of trips that people make in the Netherlands by bike? Especially school children? I’d love to believe a 40km round-trip isn’t unusual, but I do find it a little hard.

    Clearly just building density without good infrastructure won’t increase cycling (as many US cities have shown) but trip length, density, and the city population are intrinsically linked. If people are willing to make a 40km round trip, and you want the entire city to be accessible to them (a circular city with a 20km diameter), and you have the density of Groningen (2400/km^2), then the largest city population you can have is about 750,000. At Boulder’s density (1500/km^2) you’d top out at 470,000. If you want a city of a million within this range, you need a density of 3200/km^2.

    In Boulder, we have good bike infrastructure (by US standards), with a significant amount of separated lanes, but the cycling rates are still low (only a few percent), even for trips within town. Certainly the infrastructure could be better, but I don’t get the impression that just the infrastructure is enough, and in any case, there won’t be political support for building more and taking space away from cars unless we can get the number of people cycling to increase. These two things are a feedback loop — cultural acceptance and good infrastructure. You can only get so far with one before you have to work on the other. I feel like Los Angeles (where I used to live) was unwilling to build anything, and so we could only go so far in advocating for people to ride, before they realized that actually, riding was pretty unpleasant in those circumstances. Here in Boulder, instead, I feel like we’ve got better infrastructure than the level of cultural acceptance would indicate. Politically, it’s hard to spend millions of dollars and take some space away from cars without that normalization of the bike as a mode of transportation.

Leave a Reply