Tuesday, February 8, 2011

Speculation is the main driver behind higher world food prices - the evidence is there in the data

World food inflation is the crisis of the moment. Prices for basic commodities are rising across the world, causing enormous distress, particular for the poorest and most vulnerable.

There are several ideas running around the internet seeking to explain this appalling state of affairs. As I mentioned in previous posts, these explanations can be grouped into three broad categories;

  • Rising demand due to rapid economic growth in emerging markets, expecially in Asia:

  • Supply shocks reducing the quantity of available food:

  • Speculation and the search for higher yield, which has been facilitated by negative real interest rates in the developed world.

There is a difficulty in adjudicating between these three explanations. All three are, to some extent, true. They are not mutually exclusive. Nevertheless, we need to arrive at a relative weighting. Which of these three factors is the real driver behind the devastating rise in food prices?

It helps to clarify how these explanations might show up in the data. The Asian growth story is about fundamentals. If it is true, it will turn up in the data as long run trend . Since Asia has been growing for at least a decade, this upward trend needs to be there from the early 2000s.

The supply shock story is about deviations around the trend in prices due to year-on-year climatic differences. It is also commodity specific. Different commodities are produced across the world under different conditions. One year, wheat supply is down, but banana supply is up. Read the next sentence carefully, because it is important. An aggregate food price index will smooth out commodity specific shocks. Supply shocks hardly show up in an aggregage index, and if they do, it is as comparatively small fluctuations around the trend, and only to the extent that climatic fluctations affect all commodities.

Now we can define a bubble - it is a temporary deviation from the long run trend. Prices shoot up, and then crash.

So, what do we see in the chart above? First, there has been a long run trend upwards in food prices. This trend started in 2003 and it is illustrated in the chart by the blue arrow. This, I venture to suggest, is the Asian demand story.

However, there are two undeniable massive deviations from this trend. The first occurred in 2008. Food prices shot up, jumping massively away from the trend and then came crashing down and returned to trend. Hands up, please. is there anyone out there who does not think that this was a bubble?

The second deviation is underway now. In terms of orders of magnitude, it is at least as dramatic as the 2008 bubble. Give it a year or so, and I reckon food prices will come crashing down, just like the 2008 bubble.

This is not an "either or" issue. World food prices have exhibited an upward trend for at least seven years. This is consistent with strong demand in Asia. Higher Asian demand can sit perfectly comfortably with a speculative surge generated by irresponsibly low interest rates.

However, neither Asian demand nor climatic fluctuations can not explain the two recent massive surges in food prices that occurred in 2008 and from June 2010 onwards. These are clearly deviations from long run trends; these are bubbles. To understand why food prices are now suffering from a speculative surge, one needs to understand how financial markets operate, especially when interest rates are abnormally low.


  1. yep. destruction of 1/3 the russian wheat crop, half the pakistani wheat & cotton crops, half of china in drought, south african wheat fields under water, a weather related poor crop in canada, and flooding covering an area the size of texas & california combined in the agricultural part of australia has nothing to do with it…

    to say nothing of 170 million tons of the US corn crop being diverted to ethanol, which would be enough to feed 330 million people at what the UN considers a sustainable level…

  2. re: 40% of US corn crop now going into gastanks (chart):


  3. "An aggregate food price index will smooth out commodity specific shocks." Do you have some support for that, other than "I said so."? That's like saying that an aggregate energy index will smooth out fluctuations in any of its components, or an aggregate price index will smooth out any price changes. Doesn't that depend on how the index is constructed [weighted]? Is there no shock possible that could move such an index? I'm pretty sure war in Iran would cause a large enough blip to move an energy index.

    RJS is right. Most of what you're ignoring is food being converted to fuel. What happened to oil prices in those two timeframes? But you should do a better job than "I said so" to support the index argument.

  4. The “increase in demand produces a proportional increase in cost” only works in the linear portion of the demand/supply curves.

    Towards the edge, the cost curve goes asymptotic as people are willing to pay significantly more instead of eating less.

    So, you could be understating the impacts the supply/demand side of things.

    Problem is, if farmers respond to high prices and flood next year’s markets and push prices back into the flatter portion of the cost curve, then how would you distinguish that from a speculative bubble?

  5. "I said so" is sufficient when something is self-evident. Look at the statement you are demanding "scientific" proof for:

    "An aggregate food price index will smooth out commodity specific shocks."

    Duh! Using an aggregate is sensible because people are probably going to eat the same amount of food day in and day out, and if it isn't "pizza", it will be "cheeseburgers."

    And if their budget is $5 for lunch, it is probably going to stay about the same, and if "Pizza Hut" runs out of of pizza dough and the $5 all you can eat buffet, then the person goes to the Chinese restaurant buffet, or buys 3 $1 double cheeseburgers at McDonalds.

    This search for a "scientific" study stuff can be way OVERDONE.

    Squeeky Fromm
    Girl Reporter

  6. There's no way that speculation can drive up prices without also raising inventories. Either speculators buy food to hold in inventory, or else they buy futures or forward contracts, which drives up the futures/forward price relative to the spot price and induces arbitrageurs to hold inventories. So where are the rising inventories that are predicted by the speculative explanation for rising food prices?

  7. When commodity markets are operating normally, there is a certain buffer in both supply and demand. If prices rise, supply (or substitutes) come online to dampen prices. There is likewise a certain amount of demand that will surface if prices drop.

    That is when commodity markets are operating normally. When the world is coming up against the limits of growth, there is no supply buffer. When prices spike, they spike very high until demand is reduced. The undershoot of supply relative to demand is cleared and prices very quickly come back down to the trend (i.e. prices crash).

    The main problem with the speculation argument is that some resources have no futures contracts, meaning there is no feasible way to speculate other than buying the physical. Iron ore is an example. One test of your thesis would be the price of iron ore. If it kept to trend, that would tend to support speculation as the driver for other commodities. However, iron ore prices have been just as whacked as food and energy. This I think falsifies your hypothesis. However, it is consistent with a "limits of growth" hypothesis.

  8. shargash brings to mind that chili peppers, considered a staple in indonesia, are up fivefold, and that the indian cabinet met in an emergency session to deal with the rising costs of onions, which were up threefold, because two previous govts were toppled by rising onion prices...




    as perishables, neither chilis or onions fits your speculation speculation...