USDA cut corn and cotton production and raised soybeans to a prospective record, but all three yields remain below 2025. The apparent divergence is largely an acreage story, while this week's Crop Progress report shows conditions stabilising as harvest accelerates.
TL;DR
- USDA cut corn production by 213 million bushels to 15.800 billion, with most of the reduction attributable to the 2.2-bushel yield cut rather than acreage.
- Soybean production rose to a prospective record 4.535 billion bushels, but yield at 52.8 bushels per acre remains below 2025. Harvested area is 5.4 million acres higher.
- Cotton production was cut 3% to 13.20 million bales and yield reduced by 22 pounds to 776 pounds per harvested acre, 76 pounds below 2025.
- Corn's production cut was broadly anticipated. More important for the balance sheet, projected 2026/27 ending stocks of 1.567 billion bushels leave stocks-to-use below 10%.
- Crop conditions stabilised as harvest accelerated: corn improved to 57% good/excellent, cotton to 36%, while soybeans held at 58%.
- The analytical focus is now moving from condition ratings towards realised harvest yields.
US Crop Dashboard
| Crop / stage | This week | Last week | Last year | 5yr avg |
|---|---|---|---|---|
| Corn — mature | 42% | 25% | 39% | 38% |
| Corn — harvested | 8% | 5% | 7% | 6% |
| Soybeans — dropping leaves | 44% | 26% | 38% | 37% |
| Soybeans — harvested | 6% | n/r | 5% | 3% |
| Cotton — bolls opening | 57% | 40% | 49% | 48% |
| Cotton — harvested | 8% | 7% | 9% | 8% |
| Winter wheat — planted | 8% | 2% | 10% | 12% |
| Rice — harvested | 62% | 50% | 59% | 50% |
n/r = not reported in that week's national Crop Progress table.
National Crop Condition Dashboard
| Crop | Last week G/E | Very poor | Poor | Fair | Good | Excellent | This week G/E |
|---|---|---|---|---|---|---|---|
| Corn | 56% | 6% | 11% | 26% | 44% | 13% | 57% |
| Soybeans | 58% | 4% | 9% | 29% | 46% | 12% | 58% |
| Cotton | 34% | 12% | 22% | 30% | 30% | 6% | 36% |
Conditions have stopped deteriorating, but the year-on-year deficits remain substantial. Corn is ten percentage points below 2025 on good/excellent, soybeans five points lower and cotton sixteen points lower.
Lower Yields, Different Acreage
Friday's Crop Production report appears at first to show three different stories: corn production was cut, cotton was cut and soybeans were raised to a prospective record. Separating yield from harvested area produces a more consistent picture.
All three crops are currently forecast to yield less than in 2025. Corn stands at 178.5 bushels per acre against 186.5 last year, cotton at 776 pounds against 852, and soybeans at 52.8 bushels against 53.0. The difference in production comes largely from acreage.
That is clearest in soybeans. USDA carries 85.881 million harvested acres, up from 80.437 million in 2025. Applying this year's 52.8-bushel yield to last year's acreage would produce approximately 4.247 billion bushels, below 2025's actual 4.262 billion. The prospective record crop of 4.535 billion bushels is therefore an acreage story rather than evidence of superior crop performance.
Corn moves in the opposite direction. Harvested area is down about 3% from 2025 while yield is also lower. Within the September revision, approximately 195 million bushels of the 213-million-bushel production reduction can be attributed to the 2.2-bushel yield cut, with only a relatively small contribution from revised acreage.
Corn: Attention Moves to Stocks
Corn condition recovered one point to 57% good/excellent this week after reaching its season low, while maturity jumped from 25% to 42% and harvest reached 8%, two points ahead of its five-year pace. At this stage, the modest improvement in condition matters less than the yield evidence now beginning to emerge.
USDA's reduction to 178.5 bushels per acre was also broadly anticipated. Pro Farmer's pre-report survey averaged 178.2 bushels, making the official figure confirmation of the deterioration rather than a substantial reset of market expectations.
The more important balance-sheet development is stocks. USDA projects 2026/27 ending stocks at 1.567 billion bushels against total use of 16.180 billion, putting stocks-to-use at 9.7%. The tightening was cushioned by a 150-million-bushel cut to feed and residual use, which absorbed roughly two-thirds of the supply reduction and limited the fall in carryout to 86 million bushels. At 15.800 billion bushels, production would still be the second-highest on record, but another meaningful yield reduction would tighten an already relatively lean balance sheet.
Harvest now becomes the test. Illinois is only 8% harvested and Iowa 2%, meaning evidence from the two largest producing states remains limited. The next several weeks should tell us considerably more than another one-point movement in national condition.
Soybeans: The Record Is in the Acres
Soybeans held at 58% good/excellent for a third consecutive reporting point, five percentage points below last year. Development is running ahead, with 44% dropping leaves against a 37% five-year average and harvest at 6%, double its normal pace.
USDA raised September yield only marginally, from 52.7 to 52.8 bushels per acre. Production nevertheless reached a prospective record 4.535 billion bushels because the harvested-area base is exceptionally large.
That distinction matters for the risk around the forecast. There is no contradiction between weaker condition and record production: yield itself is slightly below 2025. Instead, 85.9 million harvested acres magnify any change in the yield assumption. One bushel per acre is worth approximately 86 million bushels of production.
The balance sheet is not especially loose despite the record crop. USDA raised 2026/27 exports by 25 million bushels to 1.685 billion and cut ending stocks to 310 million from 320 million. That reinforces the distinction between record production and abundance: the crop is larger because more acres are being harvested, while projected carryout has moved lower.
The soybean question is therefore simpler than it first appears. Harvest needs to establish whether 52.8 bushels is sustainable. The record depends much more on how many acres are being harvested than on unusually strong productivity from those acres.
Cotton: More Acres, Fewer Bales
Cotton offers the clearest evidence of crop deterioration reaching production. USDA cut all-cotton output by 3% to 13.20 million bales and reduced expected yield by 22 pounds to 776 pounds per harvested acre, 76 pounds below 2025.
The acreage comparison reinforces the point. USDA carries 10.45 million all-cotton planted acres and 8.16 million harvested for 2026/27, against 9.28 million planted and 7.83 million harvested in 2025. Harvested area is therefore 330,000 acres higher than last year, while implied abandonment has risen to roughly 22% from about 16%. More acres went in and more are expected to be harvested, yet fewer bales are coming out. Cotton is therefore fundamentally a yield-and-abandonment story rather than an acreage contraction.
Condition improved two points this week to 36% good/excellent, while 34% remains poor or very poor. That is a marginal stabilisation rather than a reversal. With 57% of bolls already open and harvest at 8%, the more important question is whether the current 776-pound yield survives as more of the crop is measured.
Harvest Is Replacing Condition
The wider report reinforces the transition towards measured output. Rice harvest reached 62%, twelve points ahead of its five-year average, while spring-wheat harvest is effectively complete at 93%. Winter-wheat planting, meanwhile, is 8% complete against a 12% average — behind pace, but too early to constitute a production signal.
Moisture stress also eased slightly without disappearing. Pasture and range rated poor or very poor fell from 53% to 51%, while topsoil rated short or very short declined from 55% to 51%. Both remain materially worse than a year ago. At this stage, those readings matter increasingly for forage and livestock conditions rather than as direct indicators of row-crop yield.
The central message from the past week is therefore more straightforward than the headline production changes imply. Corn, soybeans and cotton are all expected to yield less than in 2025. Different acreage profiles turn those lower yields into different production outcomes: corn and cotton decline, while soybeans reach a record. For corn, the focus is increasingly the stocks cushion behind the crop; for soybeans, whether 52.8 bushels holds across an unusually large harvested area; and for cotton, whether the September production cut has gone far enough as more of the crop is measured.
Outlook
Base case: USDA's September revisions broadly capture the summer deterioration in corn and cotton, while soybean production remains near record levels because of its larger harvested area. Harvest continues at or slightly ahead of normal and progressively replaces condition ratings as the primary evidence.
Upside risk: Corn and cotton harvested yields outperform current expectations, limiting further production cuts. Soybean yield holds at or above 52.8 bushels, allowing the expanded acreage base to translate fully into record production.
Downside risk: Corn yields fall below 178.5 bushels, tightening a stocks-to-use ratio already below 10%. Cotton requires another yield reduction, while even a modest soybean yield miss becomes significant when multiplied across almost 86 million harvested acres.
What would change the view: Consistent harvested-yield evidence materially above or below USDA's September assumptions, alongside any significant revision to harvested acreage in October.
Key Risks
- Soybean acreage: The record production forecast depends on 85.9 million harvested acres rather than a record yield.
- Corn stocks: A further production reduction would tighten a balance sheet already carrying stocks-to-use below 10%.
- Cotton yield: The two-point condition recovery does not reverse a yield forecast already 76 pounds below 2025.
- Early harvest evidence: Illinois and Iowa remain lightly harvested, limiting the usefulness of early national corn-yield reports.
- Moisture stress: Pasture and topsoil readings have improved but remain materially weaker than last year.
Intelligence Monitoring Points
- Corn yields and stocks: 178.5 bushels per acre and 1.567 billion bushels of ending stocks.
- Soybean yield: 52.8 bushels across 85.9 million harvested acres.
- Cotton yield and abandonment: 776 pounds per acre and roughly 22% implied all-cotton abandonment.
- Winter-wheat planting: 8% against a 12% five-year average.
- October Crop Production: the next major incorporation of measured harvest evidence.
FAQ
Why is soybean production at a record if yield is below last year? Harvested area is 6.8% larger. USDA forecasts 85.881 million harvested acres against 80.437 million in 2025, more than offsetting the small decline in yield.
Is the corn crop small? No. At 15.800 billion bushels it would still be the second-highest on record. The more significant issue is projected ending stocks of 1.567 billion bushels and stocks-to-use below 10%.
What happened to cotton? USDA cut production 3% to 13.20 million bales and yield to 776 pounds per harvested acre. All-cotton harvested acreage remains above 2025, making the production decline primarily a yield and abandonment story.
Did crop conditions improve this week? Marginally. Corn rose to 57% good/excellent and cotton to 36%, while soybeans held at 58%. All three remain below their 2025 condition levels.
What matters next? Realised harvest yields. As more corn, soybeans and cotton leave the field, measured output becomes more informative than weekly condition ratings.
Sources
- USDA NASS — Crop Progress, week ended 13 September 2026, released 14 September 2026
- USDA NASS — Crop Production, 11 September 2026
- USDA — World Agricultural Supply and Demand Estimates, 11 September 2026
Data note: National crop-condition and progress figures are from USDA NASS. Production, yield, acreage and stocks figures use the September Crop Production report and WASDE. Soybean harvest was not reported nationally in the previous week's table and is shown as n/r.
This document is published by Bloodstone Research for informational and institutional research purposes only. It does not constitute investment advice, an investment recommendation, an offer or solicitation to buy or sell any financial instrument, commodity or security, or a forecast of future performance. Market conditions and data may change without notice. Readers should conduct their own analysis and, where appropriate, seek independent professional advice before making investment decisions. For institutional enquiries contact research@bloodstonecapital.co.uk.