Record high milk growth vs. record high losses, dissected
By Sherry Bunting, both parts of a two-part series in Farmshine, July 2021
The dairy industry continues to wait for USDA to provide details on three areas of dairy assistance already approved by Congress or mentioned as “on the way” by Ag Secretary Tom Vilsack.
The fly in the ointment, however, is the record-high 2021 milk production (Table 1) and accelerated growth in cow numbers (Table 2) at a pace the recent USDA World Agriculture Supply and Demand Estimates (WASDE) expect to continue into 2022.
USDA is reportedly looking at production reports — up vs. year ago by 1.9% in March, 3.5% in April, 4.6% in May — to determine how to assist without adding fuel to expansion that could threaten late 2021 milk prices in the face of rising feed costs and a worsening western drought. (The latter two challenges could temper those forecasts in future WASDEs.)
May milk production a stunner
U.S. milk production totaled 19.9 billion pounds in May. This is a whopping 4.6% increase above 2020 and 2018 and a 4.1% increase over May 2019.
Let’s look at year-to-date. For the first five months of 2021, milk totaled 96 billion pounds, up 2.3% vs. the 93.8 billion pounds for Jan-May of 2020, and it is 4.4% greater than the 91.9 billion pounds of Jan-May milk produced in pre-pandemic 2018 and 2019. Of the four years, only 2020 had the extra production day as a Leap Year.
Milk per cow was up 3% over year ago in May. Compared with 2019, output per cow is up 2.2%, according to USDA.
Cow numbers vs. 2018 tell the story
Milk cows on U.S. dairies in May 2021 totaled 9.5 million head, up 145,000 from May 2020’s 9.36 million, up 172,000 from 2019’s 9.33 million, and up 83,000 head from 2018’s 9.42 million.
Counter to the national trend, Pennsylvania had 48,000 fewer milk cows than May 2018 — dropping 30,000 into 2019; 10,000 into 2020, and 8,000 into 2021.
Elsewhere in the Northeast and Southeast milksheds, among the 24 major monthly-reported states, New York had 4000 more milk cows in May 2021 than 2018, Vermont 8000 fewer. Georgia dropped 1000, Florida 12,000, and Virginia 11,000. In the Central states, Illinois was down 10,000 head.
The total decline in cow numbers for the 24 lesser quarterly-reported states, the collective loss in cow numbers is 59,000 head from May 2018 to May 2021
Accelerated growth is coming from three key areas where major new processing assets have been built or expanded.
In the Mideast, where the new Glanbia-DFA-Select plant became fully operational in Michigan this spring, there is a net gain of 32,000 cows for 2021 vs. 2018, Ohio’s cow numbers that had been declining 2018-19, began recovering in 2020-21. Indiana had 18 months of substantial growth, and Michigan returned to its growth pattern in 2020. Taken together, the Indiana-Ohio-Michigan region had a loss of 8,000 cows heading into 2020, but gained a whopping 40,000 cows over the past year.
In the Central Plains, where new plant capacity is starting up this spring and summer — Minnesota, South Dakota and Iowa, combined, added 40,000 cows May 2018 to May 2021.
In the Southern Plains, where joint-venture processing capacity continues to grow, Texas has continued full-steam-ahead, gaining 87,000 cows from 2018 to 2021, along with 29,000 added in Colorado and 17,000 in Kansas. New Mexico regained earlier losses to be 2000-head shy of 2018.
The growth patterns in these regions somewhat mirrored dairy exits from other areas — until Jan. 2020 (Table 2). The past 17 consecutive months of year-over-year increases in cow numbers leave the U.S. herd at its largest number in 26 years (1995).
However, the assumption that ‘dairy producers are okay because the industry is expanding’ ignores several essential factors. The playing field has become more complicated and inequitable. There are four main factors at play. We’ll look at them one at a time.

Factor #1 — Milk dumping and base programs
A year ago in April and May 2020 — at the height of the Coronavirus pandemic disruptions — the dairy industry saw dumping of milk, stricter base programs and bigger milk check deductions. Producers culled cows, dried cows off early, changed their feeding programs, even fed milk in dairy rations.
But milk production still grew, according to the USDA data.
Some cooperatives and milk buyers, like Land O’Lakes, had base programs already in place and triggered them. Others made changes to prior programs or implemented new ones.
Dairy Farmers of America — the nation’s largest milk cooperative, largest North American dairy processor and third-ranked globally by Rabobank — quickly implemented a new base program in May 2020, seeking 10 to 15% in production cuts from members, varying by region, with overage priced on ‘market conditions.’
It is difficult to assess the ‘equity’ in these base programs and the cross-layers among producers between and within regions, or to know how these ‘bases’ are being handled presently. When questioned, spokespersons say base decisions are set by regional boards.
Meanwhile, product inventory and pricing schemes affect all regions, and milk rides between FMMOs in tankers and packages — with ease.

According to USDA, the 11 FMMOs dumped and diverted 541 million pounds of milk pooled as ‘other use’, priced at Class IV, during the first five months of 2020, of which 350 million pounds were in April alone. This is more than three times the ‘other use’ milk reported by FMMOs during the first five months of pre-pandemic 2019 (171.4 million pounds). By June, the amounts were double previous years.
Of this, the largest amount, by far, was the 181 million pounds of ‘other use’ milk in the Northeast FMMO 1 during Jan-May 2020, comprising one-third of all the dumped and diverted milk pooled across all 11 FMMOs in that 5-month period.
In the Southeast milkshed, the Appalachian, Florida and Southeast FMMOs 5, 6 and 7, together pooled 88 million pounds of ‘other use’ milk in the first five months of 2020. The Southwest FMMO 126 had 106.2 million pounds of ‘other use’ milk; Upper Midwest FMMO 30 had 46.1 million pounds; Central FMMO 32 had 36.7 million pounds; Mideast FMMO 33 had 30.7 million pounds; California FMMO 51 had 28.9 million pounds; Arizona FMMO 131 had 21.7 million pounds; and Pacific Northwest FMMO 124 had 1.3 million pounds.
The dumping had begun the last week of March 2020 and was heaviest in the month of April. Producers also saw deductions as high as $2/cwt. for balancing costs, lost quality premiums, and increased milk hauling costs. Unaccounted for, were the pounds of milk that had reportedly been dumped on farms without being pooled on FMMOs.
All of this against a backdrop of pandemic bottlenecks and record-high March-through-August imports of butter, butteroil, milkfat powder, and blends — adding to record-high U.S. butter inventories and contributing to the plunging Class IV, II and I prices vs. Class III (PPD).
Meanwhile, not only did production growth in key areas move ahead, so did strategic global partnerships. Just one puzzling example in October 2020, after eight months of deflated producer milk checks, depressed butterfat value, burdensome butter inventory, record butterfat imports, and a plunging Class IV milk price that contributed to negative producer price differential (PPD) losses, Land O’Lakes inked a deal to market and distribute cooking creams and cream cheeses — Class II and IV products that use butterfat — from New Zealand’s Fonterra into United States foodservice accounts.
The New Zealand press reports were gleeful, citing this as a big breakthrough that could be followed by other of their cheeses entering the “huge” U.S. foodservice market through the Land O’Lakes distribution.
Factor #2 — Class price wars and de-pooling
As reported in Farmshine last summer, dairy farmers found themselves in uncharted waters. As Class IV prices tumbled from the get-go with all of the ‘other use’ dumping and diverting, butter inventory building as butter/powder plants tried to keep up with diverted loads at a disruptive time, the USDA Food Box program started drawing products in the second half of May, and really got going by July 2020.
Cheese, a Class III product, was a big Food Box winner. The cheese-driven Class III milk price rallied $7 to $10 above Class IV, and massive volumes of milk were de-pooled by Class III handlers, which has continued through May 2021.
Reviewing the class utilization reports, an estimated 80 billion pounds of Class III milk normally associated with FMMOs has been de-pooled over the past 26 months.
At the start of this ‘inequitable’ situation, academic webinars sought to explain it.
“We’re seeing milk class wars,” said economist Dan Basse of AgResource Company, a domestic and international ag research firm in Chicago, during a PDPW Dairy Signal webinar a year ago.
He noted that under the current four-class pricing system, and the new way of calculating the Class I Mover, dairy farmers found themselves “living on the edge, not knowing what the PPD (Producer Price Differential) will be” (and wondering where that market revenue goes).
“A $7.00 per hundredweight discount is a lot of capital, a lot of income and a lot of margin to lose with no way to hedge for it, no way to protect it, when the losses are not being made up at home as reflected in the PPD,” Basse said in that summer 2020 webinar.
What does this have to do with year-over-year milk production comparisons?
Two words: Winners. Losers.
Some handlers, and producers won, others lost — between and within regions.
Here’s why all of this matters from a production comparison standpoint: Dairy economists — Dr. Mark Stephenson, University of Wisconsin, and Dr. Marin Bozic, University of Minnesota — are both on record acknowledging that USDA NASS uses FMMO settlement data, along with producer surveys, to benchmark monthly milk production.
So, on the one hand: How accurate are these data for comparison over the past 26 months, given the inconsistent FMMO data from dumping, diverting and de-pooling?
On the other hand: Did the negative PPDs and de-pooling, resulting in part from the 2018 Farm Bill change in the Class I Mover, allow Class III handlers to capture all of that additional market value and use it to fuel the 2020-21 accelerated milk growth for regions and entities connected to the new Class III processing assets?
Factor #3 — New dual-processing concentrates growth
Accelerated growth in cow numbers is fueling record production in 2021. It is patterned around ‘waves’ of major new processing investments in some areas, while other areas — largely fluid milk regions — are withering on the vine or growing by smaller margins with fewer cows.
In the 24 major milk states, production growth was even greater than the All-U.S. total — up 4.9% vs. year ago. In part one, the breakdown was shown vs. 2018.
Here’s the breakdown for just the 12 months from May 2020 to May 2021 — a time in which the industry dealt divergences that created steep losses for some and big gains for others, while FMMOs became dysfunctional.
In just one year, over 40,000 cows were added in Indiana, Ohio, and Michigan, combined, and milk production was up in May 2021 by 12.6, 3.2 and 5.1%, respectively. The draw is the massive new Glanbia-DFA-Select joint-venture cheese and ingredient plant that began operations late last year in St. Johns, Michigan. Sources indicate it reached full capacity this spring. Add to this the 2018 Walmart fluid milk plant in Fort Wayne, Indiana and other expansions in Ohio and Michigan.
Ditto for the Central Plains, where new cheese and ingredient line capacity became operational this spring and summer. Supplying these investments, Minnesota grew production 6%, South Dakota 14.6%, and Iowa 6.2% over year ago.
Number two Wisconsin grew by 5.6% in May 2021 vs. year ago.
Milk production was up 5% in number one California, even though cow numbers were down by 1000 head, and dairy farmers in a referendum voted recently by a slim margin to keep their quota system. They are also dealing with a devastating drought that news reports indicate is now impacting both the dairies and the almond growers.
Then there’s Texas, where growth continues to be a double-digit steamroller, up 10.8% in May 2021 vs. 2020 — pushing New York (up 4.2%) to fifth rank.
The Southern Plains has had several strategic investments, starting in Texas and New Mexico (up 6% vs. year ago).
In Colorado, where production was up 5.3% in May, DFA’s joint ventures and strategic partnerships with Leprino, Kroger and others have fueled growth.
Kansas grew milk production 7.3% vs. year ago. In 2018, a state-of-the-art whole milk powder and ingredient plant became fully operational in Garden City, Kansas. The plant was to be a joint-venture between DFA and the Chinese company Yili but ended up as a joint-venture between DFA and 12 of its member farms that are among the 21 Kansas dairies shipping milk to it.
DFA’s Ed Gallagher gave some insights on this during a May 2021 Hoards webinar. He said, “We went through a period of investing in powder plants in the U.S. It seems like there is a follow-the-leader approach when deciding on investments, and it goes in waves. The industry just completed a wave of a lot of investment in Class IV manufacturing plants, and now… it’s flipping to Class III.”
Looking back on the Class IV ‘wave’ 2013 through 2018, there were several times in those years that Class IV beat Class III, leading to FMMO de-pooling, but not to the extreme extent seen in the past 12 months as Class III now beats all other classes, including Class I, leading to negative producer price differentials (PPDs).
Gallagher sees Class III and IV prices “coming together” in the “next period of years” because the ‘wave’ of capacity investment has flipped from Class IV to III. He predicted more Class III capacity will be added.
Are these past 26 months of PPD net losses for producers the industry’s answer to, in effect, increasing processor ‘make allowances’ without a hearing?
The average PPD value loss (see chart) across the seven multiple component pricing FMMOs was $2.57 per hundredweight for 26 months, which began with implementation of the new Class I pricing method May 2019 through the most recent uniform price announcements for June 2021 milk.
Applying a conservative 5-year average PPD (prior to Class I change) for each FMMO, only the few gray blocks on the chart represent ‘normal.’
This means even positive-PPDs show margin loss for farm milk pooled on FMMOs. In fact, the CME futures markets as of July 14 show August through December divergence between Class III and IV above the $1.48 mark, indicating Class I value loss and negative PPDs or smaller positive PPDs could return after barely a two-month reprieve.
Many handlers that don’t pool on FMMOs also use the uniform prices as a benchmark.
This $2.57 net loss for seven MCP FMMOs across 26 months represents almost a doubling of the current make allowance levels.
Current USDA make allowances and yield factors add up to a processor credit of $3.17 per hundredweight on Class III and $2.17 on Class IV. This already represents 11 to 25% of farm milk value, according to 2018 analysis by John Newton, when he was Farm Bureau’s chief economist.
Why is this important? Because we are already seeing additional margin transfer from Class I to Class IV as the industry moves to blended beverages that mostly use ultrafiltered (UF) milk solids. Blends using whey would fall under Class III.
Looking ahead, DFA now owns most of the former Dean Foods’ Class I fluid milk plants since May 2020. New manufacturing synergies are undeniable, considering the direction of dairy checkoff’s fluid milk revitalization plan emphasizing these dairy-based-and-blended beverages and ‘dual-purpose’ processing facilities.

As low-fat UF milk solids are blended with other ingredients in a manufacturing process to make new combined beverages, the result is a competing beverage, and the milk in the beverage drops from Class I to Class IV.
Meanwhile, these beverages cost more at the grocery store, and the ingredients are not part of the USDA end-product pricing ‘circle’. Therefore, no new make allowances should be requested because processors are already getting a reduced class value, and a higher margin.
DMI’ vice president of global innovation partnerships, Paul Ziemnisky, gave some insights into this “future of dairy beverages” — and how it ties into new processing plants investments during the virtual Pennsylvania Dairy Summit in February.
Ziemnisky went so far as to say new processing facilities will “need to be built as beverage plants able to handle all kinds of ingredients” for the blended products of the future. In essence, he said, the future of fluid milk is “dual purpose” processing plants.

While 11 of the top 24 states had milk production increases of 5% or more in May, the 13 states with increases below 5%, or negative, are mainly located within traditional Class I fluid milk marketing areas: Florida, up 0.5%, Georgia up 2%, Virginia down 2.3%, Illinois up 1.9%, Arizona, down 0.5%, Washington, down 0.9%, Pennsylvania and Vermont both up 1.8%, and New York up 4.2%.
Idaho and Utah, up 2% and unchanged, are outliers and largely unregulated by FMMOs. Some beverage assets are coming to that region in the form of ultra-filtration and aseptic packaging, including a plant renovation to make Darigold’s FIT beverage. Additionally, a new Fairlife filtration membrane plant was opened near Phoenix, Arizona in March, and Kroger is doing filtration and aseptic packaging in Colorado.
Meanwhile, Pennsylvania is often described as a ‘fluid milk state’ with a Milk Marketing Board setting minimum prices for fluid milk, and a string of independent milk bottlers that figure prominently in their communities.
Ranked fourth in milk production in 2006, Pennsylvania was passed by Idaho in 2007. By 2016, Michigan had pushed Pennsylvania to sixth. The very next year, in 2017, Texas leapfrogged both Pennsylvania and Michigan. Now, Minnesota has pushed the Keystone State to eighth.
How does the future of dairy affect traditionally ‘fluid milk’ states like Pennsylvania, or the Southeast for that matter?
New dairy-‘based’ beverage innovations can be made anywhere and delivered anywhere, often as shelf-stable products. Most are not Class I products unless they meet the strict FMMO definition which was last spelled out in the USDA AMS 2010 final rule.
For now, this also includes the Pa. Milk Marketing Board. Executive secretary Carol Hardbarger confirms that the 50/50 drinks are not regulated under PMMB, which generally uses federal classification, but that a legal interpretation of the Milk Marketing Law with regard to blends may be in order.
The 50/50 blends are already in some Pennsylvania stores and elsewhere in the Northeast, which is the second phase of the ‘undeniably, purely perfect’ marketing plan for fluid milk revitalization.
Factor #4 — USDA, industry coalesce around climate
Ag Secretary Tom Vilsack has been outspoken from the outset about using and aiming every available USDA program dollar in a way that also addresses the Biden administration’s strategies for equity, supply chain resiliency, and climate action.
Speculating a bit as to why USDA is taking so long to announce details about already funded dairy assistance, it could be that Sec. Vilsack is looking at the fit for ‘climate impact.’
Paid around a million a year in dairy checkoff funds to serve 4 four years as CEO of the U.S. Dairy Export Council — between prior and current Ag Secretary posts — Vilsack understands the future plans of the dairy industry’s checkoff-funded proprietary precompetitive alliances on a global scale.
Vilsack has been privy to the DMI Innovation Center’s discussions of fluid milk revitalization through ‘dual purpose’ plants and blended beverages. He is no doubt looking at the accelerating growth in milk production that is occurring right now for ways to tie dairy assistance to measured climate impacts in the net-zero file.
Producers on the coasts and fringes of identified growth areas have a target — fresh fluid milk and other dairy products produced in regional food systems for consumers who have a renewed zeal for ‘local.’ Fresh fluid milk will have to find a path outside of the consolidating system and cut through the global climate-marketing to directly communicate fresh, local, sustainable messages about a region’s farms, animals, environments, businesses, economies, jobs and community fabric.
-30-
Pingback: ‘Carbon-negative milk?’ Northeast, Southeast milksheds can already claim it | Ag Moos
Pingback: Are we moving toward cow islands and milk deserts? | Ag Moos