Kub's Den

It's a strange year when I can go on a crop tour across the Western Corn Belt and the corn fields of North Dakota look tall, lush and robustly promising, but by the time I entered eastern Iowa, it was as if the nation's crops had suddenly played musical chairs and upended their usual geographic distribution. It seemed the farther south and east that I went, the shorter and more uneven the corn looked. And that's without even entering Indiana or Ohio, which I understand would have only depressed me more.

Have corn yields become wilder and less predictable in recent years? Is it, perhaps, Anthropogenic climate change that's causing all this strife and variability? (Just kidding; there's no way I would have access to enough global data points over thousands of years to address that whopper of a question.)

Anyway, a woman in eastern Iowa told me that watermelons don't taste as sweet anymore as they used to when she was a child. And she was from Muscatine, even! I had to suspect that she was incorrect, not that the USDA keeps a database (that I know of) showing average fruit sweetness levels over time. It was more likely, I figured, that our human memory just plays tricks on us. We remember the past as rosy, easy or full of wonder. Meanwhile, the present is always full of greater anxieties, due to its innate unpredictability. What we've lived through, we can mentally file away with equanimity. What we have yet to experience, we regard with misgivings.

As it turns out, U.S. average corn yields had been posting remarkably rugged yield performances recently, with a stretch of above-trendline national numbers going back at least five years (depending how you define the trend line). In 2018, national average corn yields were 3 percentage points above the expected trend-line yield (using a simple linear trend line calculated with the most basic least-squares method). In each of the two years prior to that, national average corn yields were 5 percentage points above trendline. You have to go back to maybe 2013 or certainly 2012 (23% below trendline) to really find a wild, unpredictable data point.

Similar stretches of relatively calm, predictable performances have been seen before. For instance, we might count the seven years from 2003 through 2009. But these years-long streaks of tranquility always get broken up by some wild weather and overall yield drag, eventually.

To measure the multi-year predictability over time, I considered the trailing 10-year volatility (standard deviation) of how far the nationwide average corn yield strayed from its trend line expectation, measured in percentage points. By that measure, we might say that year by year, corn yields were actually growing less volatile over time: dwindling from a trailing 10-year standard deviation of 10% to 14% in the 1980s to a quieter range of 5% to 8% in the 2000s and 2010s. The 2012 data point bumped it upwards, of course.

If the 2019 national average corn yield data point really does come to the currently-predicted 166 bushels per acre (which it may or may not), it will break the short-term series of above-trend-line performances we've enjoyed in recent years, but it won't be a shocking data point among the larger collection of historical yields.

There may be many reasons why we'll remember the 2019 growing season as absolutely unique among our collective memories -- notably the late, wet planting window and the overall number of acres that were prevented from being planted at all. But ultimately, the history books (and data tables) will only count yields from the acres that were indeed planted and will be harvested for grain. In those terms, then, what currently feels like a topsy-turvy, supremely unpredictable experience may ultimately look like nothing but a statistical blip.

Elaine Kub is the author of "Mastering the Grain Markets: How Profits Are Really Made" and can be reached at elaine@masteringthegrainmarkets.com or on Twitter @elainekub.

(BE/BAS)

Copyright 2019 DTN/The Progressive Farmer. All rights reserved.