The USDA is nothing if not relatively predictable. Note that isn’t the same as being “accurate.”
USDA released this morning its May USDA Crop Production and World Agricultural Supply and Demand Estimates (WASDE) reports. Let’s take a quick look at one of my favorite report charts, the one showing USDA’s changing domestic ending stocks guess from its “initial” stab in May through the “final” number in the following September’s Quarterly Stocks report.
The last four cycles have seen USDA’s final stocks number come in 53% (average) of what the peak guess was over the reporting period. This year that peak has been the 555 million bushels (mb) guess in March that was trimmed to 550 mb in April, and now 530 mb in May. If we stay along this seasonal path, USDA’s September 2018 Quarterly Stocks number should come in at roughly 300 mb.
For what it’s worth, the 20 mb change in ending stocks was due to a like increase in projected crush demand.
USDA’s May corn ending stocks guess came in at a benign 2.182 billion bushels (bb), unchanged from April. Domestic old-crop ending stocks of wheat were increased 6 mb due to a 15 mb decrease in exports; understandable given the pace of reported shipments late in the marketing year and an 8 mb increase in food demand.
Global old-crop ending stocks guesses in the May WASDE pegged corn at 194.86 million metric tons (mmt), down from the previous month’s 197.78 mmt; soybeans at 92.16 mmt, actually up from April’s guess of 90.80 mmt, and wheat at 270.46 mmt, also down slightly from April (271.22 mmt). 2017-18 world production guesses showed Brazil’s corn crop dropping 5 mmt from April to 87 mmt, and Argentina staying at 33 mmt. Meanwhile, Brazil’s soybean production estimate jumped to 117 mmt, from 115 mmt last month, as Argentina’s was trimmed 1 mmt to 39 mmt.
So, with basically neutral numbers, why did the soybean market post a quick double-digit rally? Because of new-crop ending stocks numbers that were calculated using nothing but unknown numbers.
For instance, the average pre-report estimate for domestic 2018-19 soybean ending stocks was 549 mb. USDA guessed 415 mb. Global soybeans were expected to come in at 91.10 mmt, with the WASDE report showing 86.7 mmt. Global corn also came in below expectations at 159.15 mmt. The average guess coming into the report was 182.0 mmt. It’s a good thing pencils come with erasers.
The bottom line is that old-crop numbers were largely neutral, with some wiggle room for domestic corn ending stocks to possibly be trimmed a bit more before the September Quarterly Stocks report. Soybeans remain the wildcard, with the game being how much more USDA will systematically decrease its guess.
Regarding wheat: It’s bearish. Both old-crop and new-crop ending stocks continue to hover around the 1.0 bb mark (old-crop 1.07 bb, new-crop 955 mb), while projected production for all wheat came in at 1.821 bb, well-above the pre-report average guess of 1.757 bb.
Darin Newsom can be reached at email@example.com
Follow Darin Newsom on Twitter @DarinNewsom