U.S. milk output:
more supply,
not a shelf-price promise.
The route from the farm to the grocery aisle is more complicated.
September 2026 vs. September 2025
September U.S. milk output rose 2.6%. USDA data show where growth came from and why more farm milk does not guarantee cheaper groceries.
One report, two production headlines
The United States produced an estimated 19.443 billion pounds of milk in September, 2.6% more than a year earlier, according to the USDA National Agricultural Statistics Service report released October 22. The increase was 501 million pounds. It is a substantial addition to the country's raw milk supply, but it is not a measurement of supermarket milk sales, a forecast for the price of a gallon, or a statement that every dairy region expanded.
More cows mattered more than a dramatic jump in yield
In the 24-state group, USDA counted 9.304 million milk cows in September, up 162,000 from a year earlier and 24,000 from August. Average production per cow was 2,012 pounds for the month, 18 pounds above September 2025. Both the size of the herd and production per cow contributed to the increase in milk, but they did not contribute equally.
Calculated from the table, cow numbers rose about 1.8% while milk per cow rose about 0.9%. Multiplying a larger herd by a somewhat higher yield produces the roughly 2.7% increase in output. The relationship is multiplicative, and the published components are rounded, so adding the two percentages is an approximation rather than an exact reconstruction of the official total. Even so, it makes the central point clear: this was not solely a story of each cow suddenly producing much more milk.
For the United States as a whole, the monthly table reports 9.743 million cows, compared with 9.574 million a year earlier. National milk per cow increased from 1,978 to 1,996 pounds. These are national values and should not be mixed with the selected-state totals in the same calculation. A production-per-cow comparison becomes misleading if its numerator uses all U.S. milk and its denominator uses only cows in the 24-state group.
The definition of the herd also matters. The report includes dry cows, which are temporarily not producing milk, and excludes heifers that have not yet freshened. Reported production excludes milk suckled by calves. The average is consequently a statistical measure of the specified herd and production coverage, not a guarantee of the output of every animal being milked on a particular day. Herd composition, management and other conditions can affect that average, but this release does not isolate the causal contribution of each factor.
The biggest percentage gain was not the biggest addition
Kansas recorded one of the clearest expansions in the September state table. Output rose 14.7%, from 429 million to 492 million pounds, while its cow count increased from 219,000 to 248,000. Production per cow rose from 1,960 to 1,985 pounds. The table therefore shows an expansion strongly associated with a larger herd, rather than an equally large percentage jump in yield.
Wisconsin's percentage increase was smaller, at 3.3%, but its addition to output was larger: 88 million pounds, from 2.694 billion to 2.782 billion. Texas added 62 million pounds, New York 57 million and Idaho 54 million. Together with Kansas, these five states supplied 324 million pounds of the 494-million-pound increase in the selected-state total, approximately 65.6% using the rounded published quantities. This is a contribution calculation, not an official USDA attribution of causes.
The contrast between Kansas and Wisconsin explains why rankings can mislead. A high growth rate can occur on a smaller base, while a modest rate in a large producing state adds more physical product to the national supply chain. For a reader interested in total milk availability, the change in pounds matters alongside the percentage. For a reader examining structural change within a state, the percentage and cow numbers may be more informative.
California remained the largest producer in the table at 3.290 billion pounds, compared with 3.289 billion a year earlier. Its change was small enough for the percentage column to use the report's notation for less than half the unit shown. Cow numbers edged up, while milk per cow slipped from 1,920 to 1,915 pounds. Arizona, Illinois, New Mexico and Utah recorded declines in output. National growth therefore does not establish that every regional processor faced the same supply conditions or that every dairy farm experienced a stronger month.
A shorter month can disguise the daily pace
September's selected-state production was lower than August's total: 18.724 billion pounds against 19.304 billion. A raw month-to-month comparison would show a decline of about 3.0%. But September had 30 days and August had 31. Before interpreting that difference as a sudden weakening in the industry, the calendar needs to be put back into the calculation.
Dividing the monthly totals by the number of days gives approximately 624.1 million pounds per day in September and 622.7 million in August for the 24-state group. This simple daily average removes one obvious calendar difference. It is not a seasonally adjusted series, and it does not remove changes in weather, herd composition or the seasonal pattern of milk output. It should not be relabeled as USDA's measure of an underlying monthly trend.
Quarterly growth continued, at a slower year-over-year rate
U.S. production during July through September totaled 59.529 billion pounds, up 2.3% from 58.208 billion in the same quarter of 2025. The average herd was 9.715 million cows, 172,000 more than a year earlier. Milk per cow rose from 6,100 to 6,128 pounds over the quarter. These quarterly figures cover the United States, not only the selected states in the release's opening monthly headline.
The table reports year-over-year production growth of 3.1% in the first quarter, 2.6% in the second and 2.3% in the third. That sequence shows a slowing annual growth rate across the three quarters. It does not mean output fell 0.3% from the second quarter to the third, and it does not mean the industry contracted. The rate of growth remained positive while becoming smaller relative to the corresponding quarters of the previous year.
Nor does the table establish a full-year 2026 outcome. Its fourth-quarter cells for 2026 are blank because that estimation period had not yet been reported. Filling them with the third-quarter growth rate would introduce a forecast that is not in the release. The next question is whether the larger herd and the observed yield pattern persist, not whether an incomplete annual column can be presented as a finished result.
Farm milk and a supermarket gallon are different products
A pound of raw farm milk is an input into a broader production and distribution system. It may become beverage milk, cheese, butter, yogurt, ice cream, milk powder or whey products. Processing, packaging, transportation and retailing add services and costs between the farm and the store. More input supply can matter for prices, but the production report does not measure all the other forces acting along that chain.
USDA's Economic Research Service makes this distinction explicit in its farm-to-consumer price-spread work. The farm value of a retail food is the value of the agricultural commodities embodied in it. The farm-to-retail spread is the difference between that farm value and the retail value. It is not, by itself, a measure of retailer profit. Treating the entire difference as profit would erase processing, distribution and other economic activities that the spread includes.
A historical example shows why a one-for-one price assumption is too simple. In its June 2025 analysis of whole milk, ERS reported a 2024 average retail price of $3.98 per gallon and a farm value of $1.97, giving a farm share of about 49%. Those are 2024 national averages, not current October 2026 shelf prices. ERS also reported that the farm share fluctuated between 47% and 56% from 2015 through 2024. The relationship changes over time and cannot be treated as a fixed markup on today's farm price.
As a purely illustrative calculation, if a farm input represented half a retail price and that input became 10% cheaper, passing through only that change while holding every other cost constant would reduce the retail price by about 5%, not 10%. That is arithmetic under stated assumptions, not a prediction for milk this month. The actual change would depend on what happens to the other components and how prices adjust. September's output report does not supply those missing conditions.
Butter and milk powder were already moving in different directions
A separate USDA release provides a timely warning against speaking of one universal dairy price. The Agricultural Marketing Service's October 21 National Dairy Products Sales Report covers qualifying manufacturers' sales for the week ending October 17. It reports different movements for butter, cheddar cheese, dry whey and nonfat dry milk.
Butter averaged $1.3547 per pound, down from a revised $1.3659 in the preceding week. Forty-pound cheddar blocks averaged $1.3856, down from $1.4309. Dry whey averaged $0.7260, up from a revised $0.7196, and nonfat dry milk averaged $1.9959, up from a revised $1.9661. In rounded terms, butter and cheddar were lower while whey and nonfat dry milk were higher. The same broad dairy sector contained both directions.
These are not supermarket prices. The butter specification covers salted butter in 25-kilogram and 68-pound boxes meeting USDA Grade AA standards. Cheddar is reported in 40-pound blocks, and the powder products are sold in specified bulk packages, totes or tankers. A household cannot take the wholesale butter number and assume it is the price of a branded one-pound retail package. Product specifications, transaction stage and unit all belong beside the quoted price.
A cheese plant does not use every part of milk in the same way
ERS explains that the value of co-products must be considered when estimating the farm value attributable to a retail product. Without that adjustment, the same farm input could be assigned too heavily to one product even though part of its value emerges in another output. This is why a simple comparison between the farm price of milk and the retail price of cheese is not enough to calculate a meaningful margin.
The agency's dairy basket also represents more than fluid milk. Its current methodology uses the mix of dairy foods purchased by U.S. households in 2022, including milk, several kinds of cheese, ice cream, butter, cream and yogurt. That basket is a framework for comparing farm and retail values over time, not a claim that every household buys the same quantities. A family's grocery bill depends on the products it actually purchases, their package sizes and the stores it uses.
For the October production news, the practical implication is that additional pounds of milk do not map automatically into a particular number of discounted gallons. The destination of that milk and the demand for its components matter. A report about the quantity produced is indispensable, but it is only one part of an explanation of what households eventually pay.
More production is not the same as more farm profit
The production report measures cows, milk per cow and pounds of milk. It does not contain a full farm income statement. A producer can sell more milk while facing a lower price, higher costs or both. Conversely, a farm with little change in output can experience a different financial result if its prices and costs change. Inferring profitability from volume alone skips the variables needed to answer the question.
ERS's dairy research distinguishes operating costs such as feed, veterinary services, energy, repairs and hired labor from broader measures that also account for owned resources. The distinction is important when comparing farms or studies: a figure that covers cash operating costs is not interchangeable with a measure that assigns costs to unpaid labor, land or capital. This article does not use September's production growth as a claim that those costs were covered.
The industry's structure adds another layer. ERS describes a long-running decline in the number of dairy operations alongside growth in average herd size and total production. A larger national herd does not necessarily mean more farms opened. The monthly cow total and the number of dairy businesses are different measures. Neither should be substituted for the other when describing whether the industry's expansion is broadly shared.
Revisions are part of the evidence, not a footnote to hide
NASS explains that it combines producer surveys with administrative information and historical patterns. Generally, all large producers and a sample of smaller producers are included in the survey. The agency reviews the evidence for consistency and reasonableness. Because not every dairy operation is sampled, the estimates are subject to sampling variability as well as reporting and processing errors.
What shoppers can take from the release
The immediate finding is that U.S. milk supply expanded in September compared with a year earlier, with a larger herd doing much of the work. Growth was uneven across states, and the third-quarter annual growth rate was slower than in the first two quarters. Those are observations from the production tables. They do not establish a shortage, a retail-price cut, a farmer-income boom or a guaranteed direction for the next month.
NEWJINMEDIA prepared this article by analyzing USDA's October 22 Milk Production release, the October 21 National Dairy Products Sales Report, and ERS methodology and background research. It is document-based reporting, not a farm visit or an interview with producers. Photographs are separately dated archival images of the dairy supply chain; they are not presented as evidence of conditions at a particular farm during September 2026.
Comments
Post a Comment