USDA released ending stocks for the 2024-25 marketing year in September which showed ending stocks below our expectations in July. Carry-in from 2024-25 came in at 316 million bushels, down from our forecast of 355 million bushels as exports and crush both topped our estimates. Use increased 51 million bushels from our July forecast to 4.425 billion bushels.
Crush use increased 18 million bushels while exports came in 26 million higher. The remainder of the difference can be attributed to increases in seed and residual use. Despite China being absent from the U.S. soybean market in the past quarter, exports proved to be robust, with shipments being above historical average throughout much of the summer months. Since their initial ending stocks forecast of 295 million bushels in May, USDA has maintained a relatively tight soybean balance sheet for the past several months, ranging upwards of 310 million bushels and down to 290 million bushels.
Exports to all destinations started the marketing year at 1,815 million bushels and the latest forecast (from September) pegged exports at 1,685 million bushels. The market has done little to price in relatively tight stocks, apparently balking at USDA’s estimates, assuming exports would be lower and ending stocks higher. That helps explain why prices remain near multi-year lows despite ending stocks and stocks-to-use at the tightest mark since 2022-23, when prices were around 40% higher. University of Illinois ag economics professor Scott Irwin in a study recently estimated what exports could look like if China remained absent from the U.S. market. Ultimately, he found that even if China stays away, USDA’s forecast for the 2025-26 marketing year remains achievable as world’s exportable surplus remains relatively tight when compared to import demand.
That begs the question, what could U.S. exports look like with China potentially re-entering the U.S. market? The tight 81 million acres of soybeans leaves little room for big demand shifts. We continue to anticipate production totaling 4,255 million bushels, down from USDA’s estimate of 4,301 million in their latest report. That brings total supplies to 4,592 million bushels, the tightest since 2023-24. Crush use has expanded over the past several years with more plants expected to come online in the coming months.
Combining that with persistent robust strength in soyoil demand, we anticipate crush to mark another record in 2025-26, which would be the fifth-consecutive year of record-breaking crush use. If prices were to see a significant rally during the marketing year, one could expect crush use to modestly fall, but even when soy prices were in the teens in the early 2020’s, crush use broke records. EPA’s decision on refinery exemptions could also have an effect on crush use. Residual and seed use has totaled near 115 million bushels in the past five years, so that is a safe assumption for use in 2025-26. Taking those use categories out of total supplies leaves 1,907 million bushels left over.
Even if China returns to the U.S. market for soybeans, purchases are not going to match the levels of the past couple years. Our estimates are somewhat conservative, assuming China purchases around 10 MMT for 2025-26. If the recent agreement comes to fruition, it will likely shift exports northwards of 1.8 billion bushels, but price will eventually need to ration demand and it is difficult to imagine ending stocks shrinking much below 200 million bushels. The 2025-26 ending stocks forecast is 250 million bushels.
Below that mark, prices likely see sustained strength, rationing demand. Prices recently have shown signs of strength, but there remains a lot of skepticism in the marketplace. Our ending stocks forecast implies a tight 5.7% stocks-to-use ratio. The January Crop Production outlook will give key insights into where production ended up for the current crop year, and the November Crop Production report will also assist in piecing together the crop size.
The 2026-27 sees planted acres bounce back, supported by lower input costs, a tight balance sheet and the potential for a significant shift in the soybean/corn ratio. We did lower our 2026 acreage forecast modestly following USDA revisions to 2025 acres, as acres will not likely see a big swing year-over-year as U.S. producers like to plant corn. The 2026 production is set at 4.345 billion bushels under normal weather and not for harvest acres. Crush demand is seen as continuing to expand under continued increases to soyoil use for renewable fuels.
Our assumptions see ending stocks steady with the current marketing year at 250 million bushels. Our long-range forecast shows acres increasing to the 87.0 million acres mark. Yields see modest increases as ending stocks remain near 250 million bushels. Cash prices steadily increase towards $11.50.