Since the last USDA report, released in September, corn has sold off sharply. My bullish opinion was wrong. The September report had the yield estimate, 148.1, just slightly below expectations, while my estimate was 146. The real shocker: Demand was cut by 400 mm bushels by the quarterly grain stocks report.
On October 12th the USDA with release its next World Agriculture Supply and Demand Estimate (WASDE). Over the last month corn has fallen from $7.50 to right around $6, because of the estimated drop in demand and in my opinion the "risk off" trade which sells the commodity sector.
I have again created an estimate of the projected yield. The methodology is found in my August 10 article but here I used all 18 states surveyed instead of just 10. Because of Columbus Day the most current crop condition report will not be available until after the close, so I based these estimates on no changes from the October 2, 2011, crop progress report:
|Date||Crop Condition Index|
I then regressed the crop condition index to October's estimated corn yield each year and received the following results.
Click to enlarge
One can see that the model is highly significant with an adjusted R^2 of 89.6%.
- Currently my crop condition index (variable Ex_Oct stands at 51.96
- The equation is Oct. Yield Estimate = 86.2171+ Ex_Oct * 1.08015
- Sep. Yield Estimate = 86.2171 + 51.96 * 1.08015
- Sep. Yield Estimate = 86.2171 + 56.12
- Sep. Yield Estimate = 142.33 which would round to 142.3
The standard error of the regression was 1.863, which according to statistics should produce a confidence band of 99% around 147.9 on the high side and 136.8 on the low side. September's estimate was higher than predicted by this model but still within the standard error.
This estimation model makes it look like the USDA will again lower the estimated corn crop yield. The reduction in demand is priced in. Here is the commentary from the CME about expectations of the report:
Some traders see a lower yield while others have pushed up yield estimates by 2-3 bushels per acre due to good weather and reports from early harvest. Some traders see a small revision in harvested acreage while others see acreage falling by near 800,000 acres due to FSA data. The grains stocks numbers released at the end of September will cause a sharp adjustment in beginning stocks to 1.128 billion from 920 million in last month. Ending stocks estimates are up for grabs due to acreage and yield considerations, but most traders are looking for a number near 810 million bushels from 672 million last month. Production is expected to come in near the September forecast of 12.497 billion bushels, but there is a 350 million bushels spread on the estimates.
How much corn acreage was reduced by spring flooding can very much shift the supply of corn. It will be intensely watched. But don't be surprised if the USDA again cuts the yield estimate since this model's point estimate is much lower than the previous forecast. There is actually a .092% chance, according to this model, of the October corn yield estimate by equal to or greater than the last report's value of 148.1, and remember trading in front of the release of these numbers is always speculation.
If yield estimates are again reduced and the acreage numbers are bullish look for corn to have a quick pop. But also be ready for a quick exit, as farmers are looking to sell any "post harvest" rally of their un-hedged crop.
Disclosure: I have no positions in any stocks mentioned, but may initiate a long position in CORN over the next 72 hours.
Additional disclosure: I might also get long corn futures and options.