FORT COLLINS, Colo. (Reuters) - The U.S. corn harvest is about a month away from ramping up, but there is more uncertainty over production outcomes than there has been all season after prolonged dry weather and a severe windstorm damaged crops in the heart of the Corn Belt.
On Monday, the Department of Agriculture slashed U.S. corn conditions to 64% good-to-excellent, more than analysts expected and a 5-point fall on the week. Top producer Iowa fell 9 points to 50%, well below the five-year average of 73% for the week.
Iowa has been battered this summer by both drought and the Aug. 10 derecho that blew down countless crops across the state. Iowa is on pace to notch its 12th-driest July-August in 126 years, and the last three drier periods occurred in 2013, 2012 and 1976.
Dryness has also been a problem in some of Iowa’s neighbors, such as eastern Nebraska and northern Illinois, and many market participants have serious doubts over USDA’s latest yield forecast of 181.8 bushels per acre (bpa), which would be a comfortable record.
I posted a poll on Twitter asking users about final corn yield. The poll will not conclude until early Wednesday, but with more than 1,600 voters on Tuesday afternoon, only 17% saw yield above 179 bpa. Half of the participants are banking on something below 177 bpa.
That deviates strongly from the brief poll I posted three weeks ago, where almost half of the 609 voters pegged yield to exceed 181 bpa.
The rally in Chicago-traded corn reflects the shifting market sentiment. December futures are up 8.4% so far this month, their biggest August rise since 2011. It is atypical for prices to drastically rise at this time of year since there is usually more certainty around the U.S. crop size.
Corn jumped 2.8% on Tuesday to $3.54-1/2 per bushel, the highest since July 10. Weather concerns remained on the radar, but a big Chinese purchase of U.S. corn also roused the bulls. Tuesday’s deal of 408,000 tonnes brings known U.S. corn sales to China to a record 6.13 million tonnes for the marketing year that starts next week.
Crop conditions are often a good indication of general health, but reading too far into the ratings can sometimes lead analysts astray.
In 2003, corn conditions fell 10 percentage points in the week ended Aug. 25 to 50% good or excellent, and that followed a 5-point skid a week earlier. August 2003 was a bit warmer than normal across the Midwest, and the upper portion, including Iowa, was extremely dry.
USDA’s corn yield outlook fell 1.4 bpa or 1% from August to September in 2003, but the final number came in 3.7 bpa higher than in September.
In 2017, there was never a huge drop-off in ratings, though they were relatively low from the start. Corn conditions were 62% good-to-excellent toward the end of August, but final yield came in at a record 176.6 bpa, some 7.1 bpa or 4% higher than the August forecast.
The latest corn score of 64% is down from the season high of 75% back in early June. It can be difficult this time of year to discern worsening corn from corn beginning to move through its natural dry-down process, which can reduce health ratings.
There is evidence that a dry August causes the September corn yield to be too low. When taking the nine driest Augusts across the Midwest in the last three decades, final yield was below the September peg only once – 2011. In the other eight years, final yield was an average of 3.6 bpa or 2.9% higher than in September.
But many of those dry August years also featured a lower August yield forecast than what USDA had in July. That was not the case this year, as this month’s yield was 3.3 bpa above the July number.
The trade guess for corn yield ahead of USDA’s Sept. 11 report may come in well below 181.8 bpa based on the recent weather. In the last three decades, September corn yield has dropped more than 2% below the August outlook only three times: 1993, 1995 and 2011. September yield was between 1% and 2% lower than August in another five years.
In the last 30 years, the number of increases and decreases in corn yield from August to September are evenly split. On average over the last five years, USDA’s August corn peg has been slightly closer to the final than the September forecast, though September takes the advantage over the longer term.
It is important to remember that USDA’s yields are computed only on the area that is harvested for grain. If any acres in Iowa are abandoned due to the storm, for example, they will not contribute to yield. USDA’s September yield estimate will be the first to incorporate items such as ear counts, as the August number relied primarily on producer surveys.
USDA’s Aug. 12 corn yield forecast did not incorporate damage assumptions from the derecho that occurred two days earlier.
The opinions expressed here are those of the author, a market analyst for Reuters.
Editing by Matthew Lewis
Our Standards: The Thomson Reuters Trust Principles.