Heavy Lifting - thoughts and web finds by an economist
|I also contribute to Division of Labour||Load HL's Front Page|
Friday, June 15, 2007
Southeastern Farm Press reports the following:
A new analysis of food, energy and corn prices finds rising energy prices have twice the impact on the Consumer Price Index (CPI) for food than does the price of corn, the National Corn Growers Association (NCGA) notes.I would concur that, on the margin, energy prices would likely have a larger and more immediate impact on food prices - especially those at the grocery store.
However, the opening paragraph seems a bit misleading as the story reports later:
"Demand for corn is at unprecedented levels, and we fully expect unprecedented levels of supply as well. This spring U.S. corn growers planted the largest crop since 1944. Given normal weather conditions this summer, we’ll produce the largest corn crop in history, and that will allow us to readily satisfy demand for livestock feed, human food processing, exports and fuel ethanol."
I haven't located a copy of the study yet, but on the surface this numerical example seems a bit misleading. The post-Katrina shock to gasoline prices has been in place for about three years whereas the shock to the price of corn caused by the ethanol mandate has been in place for about six to nine months. To get a feel for what is happening to energy and corn prices over the past year, I went to the Energy Information Agency and grabbed weekly gasoline prices from June 2006 through June 2007. I also went to the Iowa Agricultural Marketing Bureau and grabbed weekly corn prices. Here's the line plot of the two:
Let's take a look at the correlations between the price of gasoline, the price of corn, and time (over the past year)
. pwcorr gasprice cornprice week,sig
The upshot is that the price of gasoline and the price of corn are inversely related. That's fine. The price of gasoline hasn't changed much over the past year: the correlation between time and the price of is 0.069 but the p-value of 0.62 suggests this is not different from zero. However, the price of corn and time are positively correlated at 0.86 with a p-value of 0.000, i.e., 0.86 is statistically different from zero.
The numerical example given in the story is a bit misleading. Over the past year, the price of gasoline has increased from 287.1 cents per gallon to 307.6 cents per gallon, i.e., a 7% increase. On the other hand, the price of corn has increased from 183.5 cents per bushel to 338.2 cents per bushel over the same time period, i.e., 154.7 cents. Assume the study's estimates are correct and we that cost and demand elasticities are relatively constant (although that is admittedly an assumptin). Over over the past year the price of gasoline/energy increased 7%, not 33%, thereby contributing a 0.126 to 0.189 (0.21x0.6 to 0.21x0.9) percent increase in the price of food. Meanwhile, the price of corn increased by $1.50 (or 84% instead of the hypothetical 33%), thereby contributing 0.75 (2.5x0.3) percent increase in the price of food.
Over the longer run, the increase in energy prices have been more dramatic than 7%, but then too the price increase for a bushel of corn would be more dramatic.
This is why the concept of elasticity is so important. The author's statement about a 33% increase in the price of gasoline causing a 0.6 to 0.9 percent increase in the food price index implies an "energy price elasticy of food prices" of 0.6/33 = 0.018, which is very low indeed. The "corn price elasticity of food prices" is implied to be 0.3/33 = 0.009, which is even lower. However, when the percentage change in the price of corn (84%) is much greater than the percentage change in the price of gasoline (7%), the actual impact on the price of food might be considerably different than originally conveyed.
This is why the concept of elasticity is so important in economics.
I am sure energy price increases contribute more to food price increases than corn price increases, on the margin. Yet, which calculation is "better" - the hypothetical 33% increase in gasoline/crude and corn prices or the all-too-real 7% change in gasoline prices over the last year and the 84% increase in corn prices over the same period?
Just a thought.
Stata Data file
[Update (6-16-2007): Here's a copy of the report sponsored by the NCGA - there doesn't seem to be any "new" analysis in the paper. Rather the authors borrow their elasticity estimates from a couple of published papers.]
Is it possible that there is a lag effect and that gasoline prices are partially determined by the past price of corn as opposed to the current price of corn?Post a Comment
What if you control for the concentration (HHI) of the retail gasoline market? I wrote a paper for Dr. Wilson related to this.
Le Chai - galerie du vin
Posts that contain Craig Depken per day for the last 90 days.
Heavy Lifting's Main Page
Money I Found Today
Heavy Lifting - Firehose style (56k warning)
- Don't blame the corn?
Modified maystar design
powered by blogger