Food and Oil Over Longer Times

NOTE (Update): This article should be read as a follow-on to "Why is this?" The question at hand is whether food costs are actually dominated by petroleum costs.

---

I claim no prior expertise in this matter. Indeed, I am confused; I had thought that natural gas prices, not oil prices, would drive costs of agricultural commodities. But the latest spike in petroleum has not been matched by natural gas.

So far we have seen a wide variety of opinions proffered: one is that the coincident recent price spikes are indeed coincidental.

That all commodities move together to some extent remains plausible from where I am sitting. Does anyone have some input on that?

Here are real commodity prices and real oil prices over a longer time span. There is some question as to when we are looking at paleo-agriculture. Certainly oil prices in 1870 are irrelevant.

I had a harder time finding traded food prices, but this one comes from Phil Trans Roy Soc, so I think we can take it as reasonably authoritative, but it predates the recent glitch

It's crucial to note that vulnerability of the poorest states to food prices is relatively new. Before urbanization of the third world, bulk food was not a market commodity for poor people; communities grew their own staples. So taking this history too far back doesn't really speak to food security in the sense of the organization of global society.

The results, taking these particular plots at face value, are equivocal. We see a very close match between oil spikes and food spikes since 1960, but there is a price spike which clearly has some other cause around 1950.

The food price chart is from Jeremy Grantham, with a hat tip to Keith Kloor. I found a number of similar oil price charts via google images; this representative one is via "Supply Chain Digest".

It looks to me like there is plenty of circumstantial evidence that the current price rise is tied to the oil price rise. The alternative theories:

• Direct forcing by oil prices
• Speculation yielding common pressures across commodities
• Increased middle classes in China bidding prices up for animal feed over human use
• Increased biofuel demand
• Crop damage due to widespread climate disruption

I had been inclined to #3 as the primary forcing, reinforced by #5. But the synchrony of the oil and food prices is striking enough to pull my attentions back to the first two possibilities.

It matters a great deal, I think, which causes are at work. I had been one of those less concerned with peak oil, but if the first possibility holds we really are in a bad way.

Again, we have to separate out the oil price and the energy price. energy prices have been stable except for petroleum, which is mostly about fueling mobile heavy equipment. Certainly mobile heavy equipment is important to modern farming, but I would not have anticipated that the fuel costs would figure so prominently. I can't think of any other basis for the first.

If it's just speculation, we have an entirely different sort of problem, which amounts purely to a regulatory issue and not to a real-world physical problem at all. That would hardly be unprecedented, but if that's true, why is this not a prominent issue?

Color me confused.

Update: Some recent numbers for agricultural fuel consumption in the US.

From here: US farming in 2005 was worth 127 billion dollars.

From here, a gallon of gasoline is 114 K Btu. Treating diesel and gasoline as roughly equivalent, we have 637 trillion/114 thousand = 5.6 billion gallons of liquid fuel. Even at $4 per gallon this amounts to only 1/6 the total price. So at first pass the food price is overreacting to the fuel price by a factor of about 4.

Possible confounds: meat double counting: a good fraction of agriculture is for animal feed rather than directly for humans. This will increase the fuel intensity dramatically for that sector. Non-food components, especially forestry, accounts for a significant fraction of GDP and is low energy use. Luxury foods, perhaps not counted in the FAO index, which will produce a high GDP per unit of energy because they are labor-intensive and/or high-profit.

So, we have a dollar's worth of food containing at least 18 cents worth of liquid fuel at present prices, only 9 cents at 2005 prices, at the moment the farmer sells to the distributor. This does NOT count the energy input into the fertilizer or heavy equipment or other items purchased by the framer.


Update via Rust Never Sleeps in comments:

"The energy costs of common foodstuffs range widely, due to different modes of production (such as intensity of fertilization and pesticide applications, use of rain-fed or irrigated cropping, or manual or mechanized harvesting) and the intensities of subsequent processing. The typical costs of harvested staples are around 0.1toe/t* (*ton of oil equivalent per ton) for wheat, corn and temperate fruit, and at least 0.25 toe/t for rice. Produce grown in large greenhouses is most energy intensive: peppers and tomatoes costs as much as 1 toe/t. Modern fishing has a similarly high fuel cost per kilogram of catch. These rates can be translated into interesting output/input ratios: harvested wheat contains nearly four times as much energy as was used to produce it, but the energy consumed in growing greenhouse tomatoes can be up to fifty times higher than their energy content.

These ratios show the degree to which modern agriculture has become dependent on external energy subsidies: as Howard Odum put it in 1971, we now eat potatoes partly made of oil. But they cannot simplistically be interpreted as indicators of energy efficiency; we do not eat tomatoes just for their energy, but for their taste, and vitamin C and lycopene content, and we cannot (unlike some bacteria) eat diesel fuel. Moreover, in all affluent countries, food's total energy cost is dominated by processing, packaging, long-distance transport (often with cooling or refrigeration), retail, shopping trips, refrigeration, cooking and washing of dishes - at least doubling, and in many cases tripling or quadrupling, the energy costs of agricultural production."(Smil, Energy, 2006. pg. 152)

More in Smil 2008 Energy in Nature and Society: General Energetics of Complex Systems.

Very much the sort of thing I was looking for, thanks.

I still don't understand why food prices should be as sensitive to gasoline prices as the last few years would seem to indicate. I presume the basket used by the FAO is not heavily processed, since we are mostly interested in what food for the poorest people costs. It seems everything besides mobile farm equipment and trucking can easily have substitute energy sources.

But if farmers and truckers have to compete with the whole market for tractor fuel, prices should indeed go up, if not quite as steeply as we see. Obviously, demand for grain is inelastic, and will perversely increase as the price goes up (and demand for meat goes down). Ultimately, food is what we need and everything else is what we want. That food should be cheap enough that we throw a good fraction away is bizarre...