About this time of year, there is often a change in the farm atmosphere. The relative inactivity of late summer gives way to serious preparation for harvest, and the size of that harvest is frequently on our minds – like constantly. Early hints of what is in our fields are the hottest gossip items whenever farmers gather.
During my career, and especially the last few years, official estimates of the products of this growing season have been viewed with increasing skepticism. Some doubts are based on the age of traditional methodology, some on suspicions of deliberate tampering and some on theories of interference from outside factors.
One idea that has gained traction is that surveys may not be as accurate as current technology – especially satellite imagery. This could be true, especially since the number of satellites has ballooned as the price to put them in orbit has plunged.
The technology packed in those satellites has improved vastly as well. Satellite maneuverability now makes it possible for real-time stationary overhead observation with everything from sophisticated radar to multiple light wavelengths to astonishing optics.
One popular idea with farmers is sophisticated technology could create better, more frequent crop estimates than in-field surveys like USDA surveys or even our Pro Farmer Crop Tour. Maybe so. Technology also seems to offer less human intervention to counter growing distrust of official institutions of all kinds in the United States.
For crop reports, the USDA struggles with public misunderstanding about the need identical methods to generate data comparable to the past. Much of agriculture also operates in years, not quarters or months. While the agency is cautiously introducing alternate technology, it could be a long time before that’s how crop production reports are routinely generated. Not only do you need several years of side-by-side reporting to assure the new technology actually is an improvement, the same big problem remains: how do you know the right answer?
There is no substitute for an annual calculation of grain sold and bought, and stuff stored on the same date each year – the famed January 31 report. Right now, the least statistically reliable number for this definitive report is farmer-held stocks.
Better crop reports will need first more farmers sending their most accurate numbers, not just different politics or hardware.
(John Phipps’ commentary from 8/26/23 U.S. Farm Report episode)
