US Rice Crop Falls to 33-Year Low Despite Record Carry-In Stocks

The United States is heading into the 2026/27 rice marketing year with an unusual combination: the largest carry-in inventory in four decades, yet the smallest new crop in 33 years. The U.S. Department of Agriculture’s Economic Research Service now forecasts production at 158.2 million hundredweight, after trimming its estimate because a smaller harvested area more than offset a higher expected yield.

One hundredweight equals 100 pounds, so the production forecast is equivalent to about 15.82 billion pounds of rice. USDA raised total all-rice supply by 4.4 million hundredweight to 266.4 million, or 26.64 billion pounds, primarily because beginning stocks were revised up by 4.6 million hundredweight to 58.4 million. Imports remain forecast at 49.8 million hundredweight.

The headline increase in supply therefore does not signal a large domestic harvest. It reflects rice already in storage. That distinction matters for growers, millers and buyers because stored inventories can cushion an immediate shortfall, but they do not replace acreage or establish the production base for a following season.

USDA expects total use to hold at 226.0 million hundredweight. With use unchanged, the additional carry-in inventory flows directly into a higher ending-stock estimate of 40.4 million hundredweight. Even so, those ending stocks would be a three-year low and 31 percent below the previous season, indicating that the buffer is expected to contract materially over the marketing year.

The price signal is consistent with that tighter balance. USDA left the all-rice season-average farm price at $14.90 per hundredweight, roughly 20 percent above the revised $12.40 estimate for 2025/26. This is a forecast rather than a guaranteed producer price, but it shows how the agency expects a smaller crop and declining inventories to outweigh the apparent comfort of unusually large beginning stocks.

For producers, the critical constraint is harvested area rather than yield. Higher yields can improve output per acre, but they cannot fully compensate when fewer acres reach harvest. The September revision is therefore also a reminder that national productivity statistics and national production can move in opposite directions.

For processors and food businesses, the numbers imply a market that is supplied but less forgiving. A 40.4-million-hundredweight ending-stock forecast is not a shortage forecast; it is a smaller margin for absorbing further production losses, import disruption or stronger-than-expected use. Buyers will need to watch the composition of inventories by rice class as well as the aggregate total, because not every stored pound is interchangeable in milling and food applications.

The figures cover the marketing year from August 2026 through July 2027 and will change as harvest information, trade flows and use become clearer. The strongest conclusion available now is narrow but important: record-high carry-in stocks are masking the scale of the production retreat, while the projected drawdown and higher farm price point to a substantially tighter U.S. rice balance.