USDA Report Review

By Darin Newsom
DTN Senior Analyst

As discussed in DTN's USDA Reports Preview, "Looking Back and Projecting," there were a number of surprises possible when the November numbers were unsealed. Last year at this time, domestic soybean ending stocks jumped to a 15-month crop-report-cycle high of 480 million bushels, up 85 mb from the previous month. And given the bearishness of nearby futures spreads for both corn and soybeans and deferred spreads for corn, there was a chance corn's numbers could be slightly more bearish than average pre-report guesses indicated.

Wow...did that turn out to be an understatement.

USDA unleased a national average corn yield estimate of 175.4 bushels per acre on the world Thursday. Not only was this well above the average pre-report guess of 172.3 bpa, but it was above the high end of pre-report guesses at 174 bpa and -- Are you ready for this? -- above last year's record high of 174.6 bpa. All counter to its own branch NASS' crop condition numbers that all summer and through September ran not only well below its 2016 ratings but below the average ratings for 2013 through 2016 as well.

Hey, Mr. and Mrs. Farmer, how's that for a kick in the nether regions?

This increased guess on yield upped USDA's production projection to 14.578 million bushels, only the second largest on record because harvested acres were left unchanged at 83.1 million. Keep in mind that all these numbers are set to be chiseled in sandstone in the January report and therefore subject to change.

Thankfully, out of the kindness of its heart (for there is no fundamental reason to explain the following), USDA increased new-crop domestic corn demand by 150 mb, offsetting some of the 298 mb increase in expected production. Now, it is well documented that USDA has not a clue this early in the game of what demand might actually be, but the increase kept projected ending stocks from shooting past 2.5 billion bushels and climbing above 2.6 bb. Instead, the estimated 2.487 bb created an ending stocks-to-use ratio of "only" 17.2%, implying an average cash price for the marketing year of $2.60 with a potential range of $2.90 to $2.15.

But half of the increase in demand, or 75 mb, was due to increased export demand. This is peculiar since earlier in the day in USDA's weekly Export Sales and Shipment numbers, marketing-year total shipments for corn of 226 mb project a more meager total demand number of 1.507 bb as opposed to USDA's November guess of 1.925 bb. But what's 400 mb among friends, right? Let's just say if corn shipments don't pick up pace and USDA sticks with its now current production and total supplies estimates, next September's Quarterly Stocks report could come in close to 3.0 bb.

If you're looking at the domestic soybean numbers, don't spend a lot of time trying to figure out the production math. The old "If X = Y and Y = 1, then X = 1" proof doesn't work. In October, USDA multiplied 89.5 million acres times and an average yield estimate of 49.5 bpa and came up with total production of 4.431 bb. In November, USDA multiplied 89.5 (ma) x 49.5 (bpa) and came up with production of 4.425 bb. And this 6 mb difference in production was the only supply and demand category that changed, resulting in an ending stocks decrease of 5 mb. Naturally.

Rather than the fun-with-numbers of domestic supply and demand, the soybean market seemed to focus on the nearly 2.0-million-metric-ton increase in world ending stocks. Brazil's 2017-2018 production guess was raised by 1 mmt to 108 mmt -- no surprise, and it's still early in that crop's life for changes to be made. But, much of the rest of the difference comes from an increase in beginning stocks for the 2015-2016 marketing year (yes, you read that correctly), meaning changes go back at least to 2014-2015 if not further. And, oddly enough, those aren't in the official release. Yet these undocumented changes were enough to help send the market down double digits for the day.

The most notable wheat number was the domestic ending stocks figure of 935 mb. This was 25 mb below USDA's October estimate, due in total to a like increase in expected export demand as it climbed to 1.0 bb. Curious... earlier in the day, in the same weekly Export Sales and Shipment report mentioned in the corn analysis, total wheat shipments for the marketing year sat at 397 mb at a time when, on average, 45% of total marketing-year shipments have been made. Given that average, total export shipments for 2017-2018 would be calculated closer to 880 mb, or 120 mb less than USDA's latest estimate. If that's the case, then any bullish exuberance that may have existed going home Thursday night will likely evaporate over time.

Darin Newsom can be reached at darin.newsom@dtn.com

Follow him on Twitter @DarinNewsom

(AG/)

Copyright 2017 DTN/The Progressive Farmer. All rights reserved.