Porto Alegre, September 17, 2026 – The September WASDE brought some surprises for the soybean market. The adjustments made by the U.S. Department of Agriculture (USDA) were disappointing compared with market expectations, particularly regarding U.S. new-crop exports.
The USDA raised its crop projection due to a slight increase in yield. In the previous report, yield had been cut from 53 to 52.7 bushels per acre; now, expectations stand at around 52.8 bushels per acre. As a result, the U.S. crop is projected at 123.4 million metric tons, an increase of approximately 435 thousand metric tons from the previous estimate, which, if confirmed, would represent a very good crop in terms of supply.
The market expected a production cut, but above all adjustments to the export outlook. There were some adjustments, but they were very conservative. The USDA carried over part of the remaining balance from 2025/26 net sales, which is normal, but made very little adjustment to expectations for new-crop exports, which already stand at 18.9 million metric tons, well above the same period last year and also above the 2024/25 season. In other words, this figure will still require further adjustments, particularly if there are any supply problems with Brazil’s 2027 crop.
U.S. stocks were adjusted by 272.1 thousand metric tons, but still remain at a comfortable level of around 8.4 million metric tons, while the market had consistently expected something closer to 7 million. The problem is that this triggered a sharp corrective move on the exchange, which, to some extent, had already been anticipated due to the significant volume of non-commercial net positions held by funds, which were extremely heavy, at around 250 thousand net long contracts. As a result, following the release of the report, the market fell approximately 2.8%, equivalent to a decline of 30 to 36 points.
The U.S. soybean harvest is about to begin, with fieldwork expected to start and the first data beginning to be released from mid-September onward, around September 14 to 15. Corn, which is further advanced, has already begun to be harvested. As a result, the market will pay extremely close attention to harvest percentages and average harvested volumes.
These data will begin to provide a more realistic perception of the size of the U.S. soybean crop, which may and likely will still undergo changes. The pace of harvesting will also be extremely important, given the strong volume of export contracts already fixed and the pace of domestic crushing, which remains intense.
In Brazil, the week did not see large trading volumes. Even before Friday’s report, some volumes changed hands, mostly for the new crop, as available soybeans are now practically scarce. In the Northeast, there are virtually no soybeans available. In the Center-West, volumes are also very limited, with Mato Grosso having around 1 to 1.5 million metric tons available at most, part of which is already committed.
Regions such as Minas Gerais, Goiás and other areas are in a similar situation. Paraná and Rio Grande do Sul still have greater soybean availability, but producers remain reluctant to sell. Even with the excellent price levels being offered, they prefer to hold onto volumes and try to pressure buyers further.
The industry continues to face difficulties covering offers, which remain very firm despite high soybean meal prices. Export parity at the ports also continues to pressure the market. On Friday, due to the decline on the exchange, prices fell more sharply, by BRL 2 to BRL 3 or more.
In the local market, however, prices remain well above parity, with firm bids, while sellers are widening the spread even further, to around BRL 5 to BRL 7, bringing the market to a standstill. Although the market is now very slow, this is mainly due to limited soybean availability, with around 90% of the 2025/26 crop already sold.
Port prices may indeed rise further in October. Premiums are unlikely to decline unless there is a very strong shift in demand toward the United States. Even so, the key months for U.S. supply are October and November, while Brazil will continue to see demand.
If more severe weather problems occur in November, this could trigger an important warning signal and lead buyers to increase demand, pushing bids and premiums higher. In this scenario, Brazilian soybeans could still test levels of BRL 170 to BRL 175 per bag at the port.
On the other hand, if El Niño does not cause significant problems, players may source supplies from the United States and count on a strong influx of the Brazilian crop. Therefore, the most likely scenario is high volatility on the exchange and strengthening Brazilian premiums through the end of December.
THE AGRIBUSINESS ECOSYSTEM
FROM BRAZIL AND LATIN AMERICA
