U.S. textile mills enter November with a split picture of fiber demand
USDA reports lower manufactured-fiber use and a small cotton-only spindle gain as mills face a tighter global cotton outlook.
The Cotton System Consumption and Stocks report, released November 2 by USDA's National Agricultural Statistics Service, puts those changes inside a manufacturing business whose raw materials, machinery and final customers do not move in lockstep. A cotton-system mill can process synthetic staple fiber or blends as well as cotton. Its monthly consumption therefore measures a particular stage of textile production, not all purchases of cotton clothing in the United States.
September's clearest change came in manufactured fiber
NASS reported 16.3 million pounds of manmade staple fiber consumed on the cotton system during September, down 5 percent from August and 11 percent from September 2025. Polyester accounted for 92 percent of that total. The category is therefore heavily influenced by one fiber, rather than reflecting equal movements across every synthetic or manufactured alternative.
Polyester staple consumption was 15.028 million pounds, compared with 15.985 million in August and 16.809 million in September a year earlier. Rayon and acetate moved the other way: their combined consumption reached 605,000 pounds, against 564,000 in August and 448,000 a year before. Other manmade staple fiber totaled 649,000 pounds, up from August's 579,000 but below the previous September's 1.040 million.
These differences matter for interpreting the headline decline. The report describes reduced aggregate manmade-fiber use, but it does not show every component retreating. Nor does it identify the business decisions behind the change. Orders, fiber substitution and production schedules are possible subjects for further investigation, not explanations established by this release.
Published inventories also differed by fiber. Polyester staple stocks were 7.223 million pounds on September's last Saturday, compared with 7.729 million in August and 16.398 million a year earlier. Rayon and acetate stocks were 4.075 million pounds, up from 3.955 million in August and 3.759 million the previous September. Other manmade staple stocks reached 1.706 million pounds, above August's 1.100 million but below the year-earlier 1.799 million. These are stocks at mills, not retailers. The changes do not by themselves establish which combination of purchases, consumption or other movements caused the inventory balance to rise or fall.
The calendar affects the monthly comparison
September had 30 calendar days, one fewer than August. NASS places that information prominently in the release because a monthly volume combines both the pace of activity and the length of the reporting period. It does not publish a seasonally adjusted monthly growth rate for the figures cited here.
Dividing the reported totals by calendar days can provide an illustration, but it cannot establish factory utilization. Mills do not necessarily operate every day, and maintenance, holidays and shifts can differ. September's lower total manufactured-fiber consumption therefore should not automatically be read as a matching reduction in daily production capacity or worker hours.
A small cotton segment grew from a small base
Extra-long-staple cotton consumption reached 151,000 pounds in September, up 22 percent from August's 124,000 pounds and 24 percent from the previous September's 122,000 pounds. The corresponding number of running bales increased to 302, compared with 251 in August and 244 a year earlier.
The percentage gain is notable within that category, but its scale needs to remain visible. Extra-long-staple consumption is not a measure of all cotton consumed by U.S. mills. It should not be described as a 24 percent increase in total U.S. cotton demand. The November report's consumption table presents this particular cotton segment alongside manmade staple fibers.
There is also an important difference between running bales and the standardized bales used in USDA's supply-and-demand outlook. The October outlook specifies 480 pounds per bale for its cotton projections. Running-bale counts describe physical bales in the monthly survey. Comparisons across the two publications should retain the original unit or use reported weights, rather than treating every bale figure as interchangeable.
Machinery counts give a second view
The cotton system had 545,000 spindles in place in September. That total comprised 379,000 ring spindles, 20,000 airjet spindles and 146,000 open-end spindles. All three counts were unchanged from August. The open-end count was below September 2025's 174,000, while the other two types were unchanged from that year-earlier comparison.
Active spindles consuming only cotton numbered 281,000, up from 279,000 in August and 278,000 a year earlier. NASS rounded both changes to 1 percent. Active spindles consuming only manmade fiber totaled 11,000, the same as August but below the previous September's 18,000. The mixed-fiber or blend category remained at 186,000, compared with 213,000 a year earlier.
Missing values are a disclosure safeguard
Several cells in the report carry a “D,” including September extra-long-staple stocks and the latest cotton-only spindle-hour observations. NASS explains that the symbol means information is withheld to avoid disclosing data for individual operations. It does not mean zero consumption, no inventories or a failure to collect information.
The published blend category supplies a useful example of why counts and hours need separate attention. Active blend spindles stayed at 186,000 between August and September, while their reported operating hours fell from 90 million to 82 million. The unchanged equipment count therefore did not mean unchanged activity. For manmade-only spindles, published hours decreased from 6 million to 5 million over the same period.
The monthly survey has a defined scope
The report belongs to NASS's Current Agricultural Industrial Reports program. Its methodology says data are collected from all known mills that consume cotton. Questionnaires ask for the previous calendar month's cotton and manmade-fiber consumption, inventories and spindle activity. The agency follows up on missing responses and uses historical information and current relationships to estimate nonresponding operations.
NASS says approximately 50 reports arrive each month, representing about 95 percent of total spindles. It revises monthly data when late or corrected reports arrive and publishes final figures in the following year's annual summary. Changing respondents and errors in reporting or processing can affect the estimates; the agency checks submissions for consistency.
Inventories also have a specific timing convention. The stocks column refers to the last Saturday of the month, while consumption covers the calendar month. Those measures can illuminate supply-chain conditions, but they are not observed on precisely the same time basis. The distinction becomes especially important when someone tries to turn a stock figure into an estimate of how many days of production a mill could sustain.
The farm outlook points to a smaller crop
The latest verified Cotton and Wool Outlook, issued October 14 by USDA's Economic Research Service, forecast U.S. cotton production of approximately 13.4 million 480-pound bales for the August–July 2026/27 marketing year. That was 241,000 bales above the September forecast but 455,000 below the preceding crop. The report characterized the projected crop as the second smallest since 2015/16.
Harvested area was projected at nearly 8.2 million acres, compared with 7.8 million a year earlier, but the national yield was expected to fall to 790 pounds per harvested acre from 852. More acreage reaching harvest therefore did not imply more cotton overall. The lower yield more than offset the acreage gain in the national forecast.
Regional differences remain substantial
USDA's October outlook projected Southwest upland cotton production at 4.6 million bales, down 23 percent from the preceding season. It cited worsening drought conditions and estimated harvested area of 4.3 million acres from 6.5 million planted acres. The implied abandonment rate was about 34 percent, compared with roughly 24 percent a year earlier.
The Delta and Southeast had a different outlook. Delta production was projected at 4.0 million bales, with a record regional yield of 1,267 pounds per harvested acre. The Southeast forecast was 4.1 million bales, 500,000 above the preceding crop, as additional acreage more than offset a lower yield. Western upland production was projected at 359,000 bales.
Such differences make a single national production headline incomplete for a local business. A gin's incoming volume depends on nearby harvested cotton, while a mill's purchasing exposure may span several regions or countries. The government projections do not supply individual companies' procurement plans. They do show why a national decline can coexist with increased production in parts of the Cotton Belt.
Exports dominate the U.S. demand balance
ERS projected U.S. cotton demand, consisting of exports and domestic mill use, at 13.8 million bales for 2026/27. Exports accounted for 12.3 million bales and domestic mill use for 1.5 million, both unchanged from the previous season's figures cited in the October outlook. Calculated from those forecasts, exports represent about 89 percent of the combined total.
That export share helps explain why changes in foreign spinning markets matter to American cotton producers even when domestic machinery indicators barely move. USDA projected a U.S. share of world cotton trade of 27.5 percent. Brazil remained the leading exporter in the outlook, with shipments forecast at 16.0 million bales and roughly 36 percent of global trade.
Tighter stocks do not set a store's price tag
USDA projected world cotton production at 118.0 million bales and mill use at 123.1 million for 2026/27. Global ending stocks were forecast to fall to 70.7 million bales, down 5.0 million from the prior season and the lowest level since 2011/12. The world stocks-to-use ratio was projected at 57 percent, compared with 63 percent in the preceding two seasons.
For the United States, ending stocks were forecast at 3.8 million bales. USDA's projected season-average upland farm price was 77 cents per pound, compared with a final 62.2 cents in 2025/26. That forecast refers to the farm commodity, not the retail price of a shirt, a towel or a pair of socks.
A finished item's price cannot be calculated from the cotton outlook alone. The official reports cited here do not quantify the retail pass-through of fiber costs, nor do they forecast holiday discounts. Buyers can use the balance sheet to understand raw-material exposure while keeping separate the additional processing, product design, distribution and selling decisions embedded in a consumer price.
Quality is another part of the commercial equation
Cotton is traded by more than weight. The USDA Agricultural Marketing Service's classing program measures properties including color, trash content, fiber length, length uniformity, strength and micronaire, a combined indication of fineness and maturity. The agency says it tests a sample from every bale produced in its grading operation.
Those measurements help connect an agricultural product with manufacturing requirements. Two lots with the same weight need not offer the same characteristics for a given production process. This is why a national bale count, though essential, cannot describe the entire purchasing decision. The characteristics of the available fiber and its intended use remain relevant.
What businesses can reasonably take from the release
The most defensible reading is a three-part one. September brought lower aggregate manmade-fiber consumption on the cotton system; extra-long-staple cotton consumption rose from a small base; and cotton-only active spindle counts increased slightly while several operating measures remained confidential. Those observations are useful together, with none standing in for an industry-wide verdict.
Businesses tracking demand also need to distinguish raw cotton exports from textiles traded after processing. ERS's raw-fiber-equivalent trade data convert finished and intermediate textile products into estimates of the fiber used to make them. The agency accounts for losses during production because the amount of raw fiber used can exceed the weight of the final textile product. That is a different measure from bales shipped directly overseas.
The same distinction between a forecast and a measured outcome matters across agriculture. Our analysis of USDA’s corrected corn outlook explains how a revised comparison can change the apparent movement without changing the latest forecast. Cotton readers should likewise preserve each report’s date, unit and measurement stage before drawing a conclusion about the whole supply chain.
Comments
Post a Comment