Well, Kragen is right about one thing — I do have a predisposition in favor of local agriculture!

However, his arguments are flawed. Shipping costs for lettuce (along with most other vegetables) are not calculated by weight. It is not a bulk commodity. It is per carton. In 2008, when fuel was slightly more expensive than it is now, shipping costs to the East Coast from California were about $10 per carton — more than the cost of production. See this article in “American Vegetable Grower,” for example. It is true, as per Kragen’s accusation, that I didn’t perform calculations to back up my assertion. I was basing it off of news articles I’d been reading about how shipping costs were affecting supermarket food prices.

I also disagree with Kragen’s second main point, that internalization of costs wouldn’t affect shipping costs by more than 50%. The ecological costs of the petrochemical and transport industries are huge. Concrete, for example, requires about 1.7 million BTUs of energy per yard to produce (including hauling of raw materials but not transport to site), most of which (at least 60% not including electricity) is produced with coal, whose burning has high ecological costs (mercury, SO2, etc.), not to mention the CO2. BTW, cement production generates CO2 beyond the fuel combustion used to make it — the calcining of limestone makes lots of Co2 also. Cement production also makes lots of water pollution. (see here for documentation of the environmental impact of cement and concrete.) Most of these costs are socialized or passed onto future generations. Consider also the true cost of oil spills. It is incalculable, actually, since we really don’t have a way to measure the long-term impact of ecosystem disruption.

Another way to see the hidden costs of our energy system is to imagine what would happen to fuel prices if we set a total CO2 limit in line with the scientific consensus on what it will take to prevent a climate change disaster, and then instituted a strict auction-based cap-and-trade system for emissions allowances. Or a carbon tax. Various authorities disagree on how high a carbon tax should be — small ones have no appreciable effect. Consumption growth was barely impacted when fuel prices quadrupled from 1998 to 2008. To cut down on consumption by the amount needed, we would probably have to double them again, at least, probably much more. And remember that fuel costs affect road-building costs, vehicle manufacture costs, etc. Moreover, there are other costs unrelated to CO2. I would guess that on a truly sustainable and healing planet, we should use about one-tenth the amount of fossil fuel we do today. In that case, long-distance transport will be many times more expensive than today, not just 1.5 times as Kragen says.

But even at 1.5 times, the impact on vegetable growers is significant.

Increased energy costs and reduced externalities will not cause a relocalization of agriculture. Eisenstein’s reasoning seems very weak to me, as if Eisenstein had a preconceived conclusion that food production should relocalize, and was seeking reasons to justify it. The reason he suggests is not at all plausible.

This reasoning seems very weak to me, as if Eisenstein had a
preconceived conclusion that food production should relocalize, and
was seeking reasons to justify it. The reason he suggests is not at
all plausible.

### Non-transport costs are irrelevant to buying local. ###

First, if we’re trying to figure out whether Pennsylvanians will eat
Pennsylvanian lettuce or Californian lettuce, the externalized cost of
the farming practices is mostly irrelevant. If those costs are fully
internalized, perhaps food production will become centralized in areas
where environmental impacts are lower, so Pennsylvanians might eat
lettuce grown in Cuba and Venezuela rather than California. But it
certainly won’t cause Californians to eat Californian lettuce and
Pennsylvanians to eat Pennsylvanian lettuce. So we are left with only
the shipping costs.

### Transport costs are very small because transport is very efficient. ###

Second, the shipping costs are currently minuscule. A kilogram of
lettuce costs perhaps US$15 at retail, according to netgrocer.com, but
different grades of lettuce vary in cost by more than a factor of 3.
Trucking, the least fuel-efficient form of transportation in common
use, costs [2400 kJ per tonne-kilometer] [1] in the US. At an
approximate rate of [42 GJ per tonne of oil] [2], that’s 5.7 × 10?? kg
of oil per kg-meter. Transporting each kilogram of lettuce the
4400 km from California to Pennsylvania thus costs about 250g of oil.
At an approximate rate of [6120 MJ per barrel] [3], that’s 3.9 × 10??
barrels per kg-km, or 0.0017 barrels per kg for 4400 km. At US$81 per
barrel (currently reported by [Bloomberg] [4]), that’s US$0.14 per kg,
about 1% of the retail price.

[1]: http://en.wikipedia.org/wiki/Fuel_efficiency_in_transportation#US_Freight_transportation
[2]: http://en.wikipedia.org/wiki/Ton_of_oil_equivalent
[3]: http://en.wikipedia.org/wiki/Barrel_of_oil_equivalent
[4]: http://www.bloomberg.com/energy/

Some figures from truckers say it cost them about [US$1 to US$2 per
mile to operate] [5] in early 2008, of which about US$0.75 per mile
was fuel. That’s for a truck with a gross weight of 80 000 pounds,
which might carry 26500 kg of cargo (the [typical capacity] [6] of a
forty-foot cargo container). That works out to US$0.001 per kg-km, or
US$0.12 per kg for the 4400 km from California to Pennsylvania. This
is quite close to the figure derived above from energy efficiencies
and oil prices.

(Most, though not all, of the costs of road maintenance, are included
in the fuel costs paid by the truckers.)

[5]: http://www.truckersforum.net/forum/f6/how-much-does-cost-per-mile-you-operate-2396/
[6]: http://en.wikipedia.org/wiki/Teu#Equivalence

#### Energy costs will not go so high that transport costs dominate. ####

So suppose oil prices — as a result of shortages or internalization
of currently-externalized costs — increase by a factor of 5, to
US$400/bbl. This raises the cost of shipping a kilogram of lettuce
from Pennsylvania to California from US$0.14 to US$0.70, raising its
retail price by 56¢, from about US$15 to about US$15.60.

Given that many people already pay US$45/kg or more for better-quality
lettuce at retail, it seems implausible that an increase from US$15 to
US$15.60 would cause a mass shift from buying cross-country trucked
produce to buying local produce.

So, how much might full internalization of costs raise the prices of
oil? Eisenstein doesn’t attempt to consider the issue. I think it’s
implausible that removing pollution, cleaning up oil spills, and
maintaining roads would increase the price of oil by a factor of 5.
And wars, of course, do not produce oil, nor are they a result of its
consumption; they simply determine who benefits from the oil and who
does not, at a terrible, useless cost in human life and economic
damage.

My best guess is that internalizing all transport costs would increase
the cost of transportation by 50% or so, a factor of 1.5.

### Higher energy costs would make transport more efficient. ###

Third, if energy costs were to increase greatly, a market economic
system would shift toward using more energy-efficient means of
long-distance transport, such as trains and ships, so lettuce prices
would rise even less than the figure I’ve listed above.

If all these costs were embodied in a head of lettuce, California lettuce would be prohibitively expensive in Pennsylvania. We would buy only very special things from faraway places.

Above critiqued at http://lists.canonical.org/pipermail/kragen-tol/2011-September/000941.html