USDA Reports Review

Above Expectations, Below the Belt

USDA on Thursday estimated corn's national average yield at 169.5 bushels per acre, below last year's 174.6 bpa but well above the average pre-report guess of 166 bpa. (DTN file photo by Pamela Smith)

Sometimes it's best just to start at the beginning. And in the case of USDA's August Crop Production and Supply and Demand reports, that means new-crop corn and soybean yield and production estimates. To say Thursday's numbers were bearish could be considered one of the understatements of the year. These numbers weren't just bigger than the average pre-report estimate, they came in well above the high end of the range for pre-report guesses. In the end, USDA literally hit U.S. agriculture below the belt, as it seemingly ignored July weather and, as always, its own weekly crop condition ratings.

USDA estimated corn's national average yield at 169.5 bushels per acre (bpa). Yes, this is below last year's 174.6 bpa, but everyone on the face of the earth who has paid a modicum of attention to the crop since planting season knew that it should be. The problem was this was well above the 166 bpa average pre-report guess and outside the range, bounded by a high guess of 168.5 bpa.

Recall that July was blistering hot and painfully dry across much of the U.S. Midwest, the heart of U.S. corn and soybean growing country. Also, remember that NASS' weekly crop condition numbers for Sunday, July 30, resulted in a DTN Crop Condition Index of 144 points, a low number in line with what was seen during the 2011 drought season when corn yield eventually averaged almost 6% below calculated trendline. Applying that same math to 2017 would result in a possible national average yield of 161 bpa.

This being USDA's annual first field-survey-based set of estimates, the thought was, based on July weather and at least some relation to crop conditions, yield projections would take a first step in the direction of a dramatic decrease. It didn't, though deniers of USDA fallibility would suggest the government did just that with its August estimate of 169.5 bpa.

Production was similar, with the average pre-report guess coming in at 13.841 billion bushels with a high end of the range of guesses at 14.070 bb. Given USDA's 169.5 bpa yield and unchanged harvested area of 83.5 million acres (at least until October), the official production projection came in at 14.153 bb. Though it won't likely stand the test of time -- possibly changing again as soon as the September round of reports -- this estimate puts the 2017 crop at third all-time highest in size of production behind the massive 15.148 bb of 2016 (that the market continues to deal with) and 14.216 bb seen in 2014.

Soybeans were also hit with a foul blow as USDA actually increased its national average yield estimate from July's 48.0 bpa to 49.4 bpa. The pre-report average guess was 47.4, with a high guess of 48 bpa. As one would conclude, production jumped as well, from last month's 4.26 bb to what would be a new record of 4.381 bb. The 2016 crop reportedly produced a then-record 4.307 bb. Guesses going into Thursday's reports ranged from 4.260 bb to 4.165 bb, with an average of 4.202 bb.

The one bright spot for soybean market bulls was expected -- at least by those familiar with my analysis. USDA lowered its domestic ending stocks number to 370 mb from last month's 410 mb, and now sits 23% below its high estimate of 480 mb from last November and December. Recall that the last three 15-month reporting periods, ending with the September Quarterly Stocks report, have seen an AVERAGE decrease from its high estimate to Sept. 1 final of roughly 63%. If that holds true this year -- and the September Quarterly Stocks report is only a month-and-a-half away -- it would suggest a final domestic soybean ending stocks number of roughly 170 mb.

But nobody cares. And by that I mean the market. Its focus is now on new crop, where USDA again sits at record production and an ending stocks figure bumped to 475 mb. It is sleight of hand any magician would be proud of: Divert attention with new crop and change old crop. It happens every year.

What lessons did USDA's August round of reports reinforce? First, weekly crop condition numbers are a complete and utter waste of time and effort. Second, it doesn't matter what weather does if USDA says it doesn't matter. And lastly, as one farmer told me in a post-report email, "This (USDA reports) is all about creating winners by making losers. Farmers are the losers."

I agree with him. Sometimes that blow below the belt is hard to take.

Darin Newsom can be reached at

Follow Darin Newsom on Twitter @DarinNewsom