What Does USDA Stand For? Unreliable Stocks, Dodgy Acreage?

That is what it seems to me. Yesterday's surprises are probably really only a correction to mistakes made as long as twelve months ago if you ask me. In the June 2009 Acreage report you may recall that our chums shocked the trade by coming in uniformly higher across the board for wheat, corn and soybeans acres than the trade had been expecting.

Despite the widely publicised late plantings and extremely wet conditions that prevailed throughout the growing season last year, they then proceeded to "up the ante" by raising yield estimates to record levels as well. Cue much back slapping and high fives all round, see "2009 Crop Year is One for the Record Books, USDA Reports" for example, on the USDA's own website.

Suddenly, fast forward twelve months, and US corn stocks are much lower than anyone thought. Lower than the lowest trade estimate. How can this be, where has it all gone? We know, or should know, what the level of exports and usage have been, so why aren't the stocks there at the end of the day?

Could it be that they were never there in the first place? Is it possible that yesterday's corn stocks numbers were simply a realignment, getting us a bit more in touch with the reality of the situation? Last year's corn crop was overstated but it's taken them until now to redress the balance?

It's not difficult to imagine, we are after all talking about the same guys who to this day STILL have Argy wheat production for 2009/10 at 9.6 MMT, when just about everybody else (including the whole of Argentina) has it at around 7.5 MMT. Now 2.1 MMT difference might not sound like a lot, so let's call it almost 22% instead. Just imagine if 22% of the world's grain stocks didn't really exist, they'd just been overstated and the USDA are just a bit too embarrassed to correct it.

So if last year's US record corn crop was overestimated, it seem a pretty fair assumption that the record soybean crop was too. Wednesday's ending stocks there also came in at lower than the lowest trade estimate. What's going on here, stealth accounting? And have they decided to hit us with all the bad news on one go here, or are there more overstates to be clawed back in the future too?

All of this kind of poses the next question of exactly how reliable are these acreage estimates then too? If you like records, and it seems that the USDA do, then this season's soybean area will be record large at just shy of 79 million acres.

Yet widespread wet weather (again) may well have curtailed farmer's best efforts to get all those soybean acres in. Remember here that it seem like we are only finding out now that what they were predicting last year, in reality, probably didn't actually happen.

Iowa has received 10 inches of rain in June, a whopping 60-75% above normal, according to Martell Crop Projections. The worst summer flooding in 100 years occurred in 1993 causing widespread damage in corn and soybeans. That year a US corn finished with a very poor yield, down 18% below trend, they say. Yet Midwest Corn Belt rainfall from April to June this year was even wetter than this "once in a century" flood of 1993.

Are all those soybean acres really going to get planted? And what are the real implications for final yields in 2010?

The truth of the matter is that we probably won't really know for sure for another twelve months (or more), and that the market will continue to trade whatever numbers the USDA decides to throw our way, as we all hang on their every word because that's the way it's always been.

You only know that the pot is empty, when the pot is empty, and then it's too late.