Are traders 'on the mark' for tomorrow's reports?
Aug 11, 2009
Chore time for me isn't what it used to be when I was growing up on our eastern Iowa farm, but taking care of two horses in the morning before I head in for work gives me a little time to think about the day ahead. Each morning, stop at this spot to get a feeling for the "tone of the day" - and some attitude about agriculture and the markets.
I was thinking…
... about tomorrow morning's USDA Crop Production Report.
Big day... USDA's first survey-based estimate of 2009 corn and soybean crop potential. It's important to keep USDA's survey procedure in mind as they assess yield potential as of Aug. 1.
First, USDA is conducting farmer surveys, asking for their opinion of crop potential. USDA's National Ag Statistics Service's state offices will also conduct field surveys. In those surveys, they'll collect any data point that available. If all that's available in a field are stalk counts, they'll count stalks. If, however, ears have silked, they'll also count all "ears and silked ear shoots," explained one state enumerator.
That's the way NASS does it year-to-year, but in the August 2008 Crop Production Report, the "Crop Comments" only included comments about stalk counts (not the specific stalk populations, just comments about how stalk counts compared to record levels).
The reason I bring this stalk-count versus ear-count subject up is because of some comments I've been getting from Pro Farmer Members about some "blank stalks" in the field this year. In the latest-planted, soggiest areas of the Midwest, growers reported after emergence that plant populations were off a bit from their target and from year-ago. Now these growers are reporting the ragged emergence of the crop has resulted in some "spindly" plants, or "blank" stalks. One of these growers told me this morning, "There are always a few wimpy plants or blank stalks, but there are a few more than normal this year."
Ear counts are probably one of the "most trusted" data points we collect during the Midwest Crop Tour. Obviously, "ear pops" have been trending higher in recent years, helping to push yields up in "less-than-ideal" growing seasons. At some point, however, plant/ear pops have to reach the point of diminishing returns -- when the high plant pop starts to impact ear sets or reduce ear length and/or grain fill. And it's got to be "easier" to reach that point of diminishing returns in a less-than-ideal planting season.
The comments we're getting from Members suggests ear counts will (once again) be very critical data from the Crop Tour this year. But... I won't go into the Tour "assuming" each counts will be down from year-ago... we'll let the numbers show us the way.
Now... for tomorrow morning's report, the market is expecting USDA to put the national average corn yield at 157.1 bu. per acre. I think that's a very respectable number going into this report. Last August, USDA's survey method generated a national average yield estimate of 155 bu. per acre. Assuming a 1.5% year-to-year "trendline" gain in yield, that would project to a national average yield estimate of 157.325 bu. per acre in Wednesday morning's update.
Now... with that said... that's what the market may be expecting USDA to deliver tomorrow morning, but we continue to hear in reports from the trading floor that traders are looking for this year's crop to at least challenge the record national average corn yield of 160.4 bu. per acre set in 2004. So... if USDA delivers a "157-something" yield peg tomorrow morning, these traders will immediately start to add bushels to the Aug. 1 peg to "ratchet up" yield expectations for the September, October and November Crop Production Reports.
The average pre-report trade estimate for soybeans is 42.1 bu. per acre. Again... that's a very respectable estimate. Somewhat surprisingly, the average pre-report trade guess, however, is a half-bushel below USDA's July yield projection of 42.6 bu. per acre.
As in corn, NASS will take all the available data points it can collect when sampling for yield. If pods are set, they'll count pods. If only blooms are available, they'll just count blooms. But, any data that's available, they'll collect.
In August 2008, USDA's survey methods resulted in a national average soybean yield estimate of 40.5 bu. per acre. USDA will also take into consideration crop condition ratings and the pace of development.
I "like" the 42 bu. yield expectations... not only for tomorrow morning's report, but that "feels" like a solid estimate based on the amount of moisture available to the crop. Obviously, bugs and disease pressure from this point forward will have a lot to do with the final bean yield.
USDA is doing some resurvey work on acres, too. When they did that last year, it generated some weird results. In the July 2008 S&D Report, planted corn acres were put at 87.3 million; harvested acres at 78.9 million. In the August S&D Report (using data from the August 2008 Crop Production Report), planted acres were down 300,000, to 87.0 million acres; harvested acres were up 400,000, to 79.3 million.
On soybeans, the July S&D Report put planted acres at 74.5 million; harvested acres at 72.1 million. In the August S&D Report, planted acres were up 300,000, to 74.8 million; harvested acres ballooned 1.2 million, to 73.3 million acres.