August 9, 2019 / 9:16 AM / 7 months ago

Column: Trade estimates for U.S. corn, soy crops could lead the market astray

CHICAGO (Reuters) - Monday is going to bring much-needed clarity to the agriculture market with fresh U.S. government data on the U.S. corn and soybean crops after an extremely volatile start to the season.

Soybeans being sorted according to their weight and density on a gravity sorter machine at Peterson Farms Seed facility in Fargo, North Dakota, U.S., December 6, 2017. REUTERS/Dan Koeck

The market is most unhappy with the U.S. Department of Agriculture’s current corn acreage number because many do not believe farmers were able to plant given the excessive spring moisture. Corn yield, soybean area and soybean yield are not too contested from USDA’s most recent view, but the yield ideas are pretty narrow and the corn acreage numbers are very one-sided.

Normally, acreage would have been settled in June following the release of USDA’s acreage survey, but farmers in early June reported they had an unusually high number of acres left to plant. This prompted USDA to resurvey producers in 14 states beginning in July, and Monday’s number should reflect those results.

For corn, the market is certain that acreage will fall significantly from USDA’s previous estimate, and it always creates a risky situation when all market participants are on the same side of the boat.

On average, analysts see U.S. corn plantings right around 88 million acres, well below the 91.7 million from the June survey. The estimates range between 83.5 million and 89.8 million acres, meaning that the trade expects at least 1.9 million acres to come off the June figure.

But harvested area, not planted, really needs to be the focus. Any corn planted, even if done so as a cover crop on a field that was prevented from planting, will be counted in the planting figure. USDA’s survey asks farmers about the intended use of the corn, so some of the very late-planted corn could end up harvested for silage and not grain, the latter of which is what the market cares about.

This could cause traders to have underestimated corn plantings, and it could lead them into misinterpreting the data on Monday if they favor planted acreage over harvested. Analysts see harvested corn acres at 80.05 million, some 3.55 million acres below USDA’s current peg of 83.6 million. The range of estimates is from 76.1 million acres to 81.9 million.

The soybean planted area that USDA published on June 28 was just as shocking to market-watchers as the corn area. That number was 80 million acres, more than 9 million fewer than last year. The market sees that number rising to 81 million acres on Monday with a range of 78 million to 83.5 million, allowing room on either side of the June number, unlike the estimates for corn.

Harvested soybean acres are seen jumping to 79.9 million from 79.3 million in June, but the difference between planted and harvested is not as controversial for soybeans as for corn since soybeans have a much higher harvested percentage. However, soybeans could have also been planted on a field that was previously prevented from planting.

The total universe of corn plus soybean plantings fell 5.7 million acres to 171.1 million from the March to the June surveys, and analysts see that falling another 2.1 million on Monday. This implies the market may be thinking that around 7.8 million acres of corn and soybeans were prevented from planting, but one should use caution because that is not always a one-to-one relationship.


Some market participants think corn and soybean yields could be drastically lower than USDA’s current estimate because of the late planting, but that is not reflected in the Reuters survey of analysts. Historically narrow trade ranges can sometimes lead to surprises on report day.

On average, analysts predict corn yield will come in at 164.9 bushels per acre with a low of 161 bpa and a high of 167.2. That is a range of 6.2 bpa, lower than last August’s 9.2 bpa and the recent five-year average for August of 7 bpa.

For soybeans, the average trade guess is 47.6 bpa with a 3 bpa-range of 46 to 49. That is smaller than the August 2018 range of 3.5 bpa but larger than the five-year average of 2.3 bpa.

USDA’s National Agricultural Statistics Service (NASS) this month will formulate its yield estimates largely on farmer surveys and satellite imagery. USDA’s yield estimates have been the model-based ones from the World Agricultural Outlook Board so far this season, and as of last month the yields stand at 166 bpa for corn and 48.5 bpa for soybeans.

In the past, the August projection from NASS also included objective yield measurements, but USDA announced in March that it would drop this practice in August and do the first assessment in September from now on.

The trade estimates for yield have been corroborated by other farmer surveys. Commodity brokerage INTL FCStone conducted a survey published last Thursday that suggested corn yield at 167.4 bpa and soybean yield at 47.2 bpa. Analyst Farm Futures put out a farmer survey on Monday that placed yields at 167.2 bpa and 48.4 bpa, respectively.

But that is not what Twitter users told me earlier. I posted a poll that concluded July 21 asking for people’s best guess of corn yield, and 63% of the nearly 1,700 voters said yield is likely 161 bushels per acre or lower. I also did a Twitter poll of soybean yield on July 30, and 72% of the 1,400 voters saw yields below 46 bushels per acre.

This means there are people out there with much lower yield ideas than what the analyst polls for Monday indicate. Perhaps some of those people do not expect those light numbers to be discovered in August or they have changed their minds since voting on Twitter.

Mob mentality on yield has led to August shockers in recent years, and they have not been bullish. USDA’s August soybean yield has landed above the top end of the trade range for four years in a row now. Corn yield landed above the trade average but below the top end last year, though the August peg had topped the trade range in the previous three years.

Keep in mind that if the corn yield comes in at 166 bpa, for example, that does not mean USDA opted to keep yield unchanged. Instead it would be a completely separate line of evidence that supports the earlier modeling work and conclusions from USDA’s World Board.

The opinions expressed here are those of the author, a market analyst for Reuters. 

Editing by Matthew Lewis

0 : 0
  • narrow-browser-and-phone
  • medium-browser-and-portrait-tablet
  • landscape-tablet
  • medium-wide-browser
  • wide-browser-and-larger
  • medium-browser-and-landscape-tablet
  • medium-wide-browser-and-larger
  • above-phone
  • portrait-tablet-and-above
  • above-portrait-tablet
  • landscape-tablet-and-above
  • landscape-tablet-and-medium-wide-browser
  • portrait-tablet-and-below
  • landscape-tablet-and-below