How Accurate Are The USDA Wheat Numbers?

Friday's USDA wheat production estimates threw up some interesting questions, like exactly how accurate are they for a kick off.

The IGC has global 2009/10 ending stocks almost thirteen and a half million tonnes lower than the USDA's Friday estimate.

That's a pretty big difference is it not? AgResource said Friday that the USDA estimate is 12-16 MMT overstated.

In Russia, the USDA increased their production estimate to 60 MMT, yet SovEcon reduced theirs to 55-60 MMT, meaning that our chums (or is it chumps?) are now going in right at the top end of other analyst's estimates.

For Canada the USDA said 23.5 MMT, yet Agri-Food Canada also released their estimate Friday, coming out with a figure of only 22.1 MMT.

The Buenos Aries Grain Exchange say Argentina will produce only 6 MMT of wheat in 2009, yet the USDA say 9.5 MMT, over 50% more than the local exchange.

That potentially leaves the USDA overstating global production by up to 11 MMT this year, in just those three countries alone. Do they know what they are doing? That's another interesting question. How much wheat will Argentina export in 2009/10? Also another interesting question, it depends on what you mean by 2009/10.

Table A from the USDA says 2.5 MMT World Wheat, Flour, and Products Trade

Table B from the USDA says 4.0 MMT Wheat Supply and Disappearance: Selected Exporters

Table A clearly refers to the marketing year running from July 1st 2009 to June 30th 2010. Table B is a bit more vague, when to when in 2009/10 exactly it doesn't say. But it clearly says 9.5 MMT in the production column, so we are clearly talking the coming season. And how are they going to export what they aren't going to produce?

There's more questions than answers here.