In the week of 10–14 August 2026, USDA quoted whey protein concentrate at two prices that differ by roughly a factor of six. WPC 34% traded in a mostly range of $2.0000 to $2.2400 a pound. Regular WPC 80%, in the same report, was “reported from $12 to $13.” Whey protein isolate ran $14 to the upper $14s.
That gap matters because the cheaper number is the one circulating as the commercial rationale for precision-fermented protein. The 4 August account of ADM’s agreement to make The EVERY Company’s OvoPro at Clinton, Iowa cites USDA figures of $1.52 a pound in January 2026, $1.74 by May, and as high as $2.18 by July — and frames that curve as what is “pushing formulators to look for a different route to protein fortification altogether.”
Those figures are real. They are also the WPC 34% series. A formulator adding 10 to 20 grams of protein per serving to a bakery product is not buying WPC 34%; they are buying WPC 80% or isolate, and paying five to seven times more for it.
The practical consequence: if you are building a cost-in-use model to decide whether to trial a precision-fermented egg protein, the whey line in that model should be $12 to $13, not $2.18. The switch looks far more defensible against the correct benchmark than against the one being quoted.
Identifying the series
USDA AMS publishes WPC prices under Dairy Market News as “Whey Protein Concentrate — Central and West U.S.” (MMN slug 1053, report MD_DA771), covering “Central and Western U.S., All First Sales, F.O.B., Extra Grade, Conventional, and Edible Whey Protein Concentrate” in dollars per pound. The headline series is WPC 34%. WPC 80% and WPI appear as commentary within the same report rather than as the quoted series.
The identification of the cited figures is close to certain. USDA’s own monthly-average file for the Central and West WPC 34% series averages the mostly range, and the 2026 monthly values reported from that series are $1.5175 for January and $1.7448 for May — which round to exactly the $1.52 and $1.74 quoted. No WPC 80% series produces those numbers.
One caveat on our own sourcing: the 2026 monthly averages come from an aggregator, IndexBox, explicitly citing the USDA AMS MyMarketNews release of 5 June 2026. USDA’s directly published monthly-average file was last released on 4 April 2025 and runs only to March 2025. Where the two overlap they agree to the cent — the aggregator’s January 2025 figure of $1.5707 matches the official file exactly — which is why we are willing to use them, and why we are saying where they came from.
Why the basis choice is not pedantry
USDA publishes both a full price range and a narrower “mostly” range, and the monthly average is built from the mostly prices. In the 13 August report the full WPC 34% range topped out at $2.7500 while the mostly range topped out at $2.2400 — a 22% spread between two numbers that a secondary source could each describe as “the price of whey protein concentrate.”
Single-point monthly figures quoted without a stated basis are not wrong, but they are not checkable either. That is a recurring problem in this sector, and one we have written about before in the context of modelled production costs.
What the correct benchmark actually shows
The stronger version of the argument is available and nobody is making it. USDA’s own commentary in the 13 August report reads: “Prices are supported by strong demand and very tight availability. Production remains constrained, with some manufacturers reporting limited to no spot supply.” The same report notes that a single load of WPC 80% can approach half a million dollars.
That is a supply story, not a price story, and it points at a constraint that is genuinely structural.
| Grade | USDA quote, week of 10–14 Aug 2026 | Typical role |
|---|---|---|
| WPC 34% | $1.8800–$2.7500 range; $2.0000–$2.2400 mostly | Commodity ingredient, low-inclusion |
| WPC 80% | $12 to $13 | Protein fortification, sports nutrition |
| WPI | $14 to the upper $14s | High-purity fortification, clear beverages |
The bottleneck is not milk. Global dairy production has been broadly stable and cheese output — whey’s origin as a by-product — has not slowed. The constraint sits downstream, in the specialised filtration and spray-drying capacity needed to turn liquid whey into food-grade concentrate or isolate. Those lines are already running flat out, and new ones take years and serious capital.
The capacity response is visible and dated:
| Company | Investment | Capacity effect | Date |
|---|---|---|---|
| DMK Group, Edewecht | €26m | New WPC 80% plant, 7,000 tonnes | May 2026 |
| Saputo, Ripon, Wisconsin | more than CA$180m | WPC 80% output up around 35% | June 2026 |
| FrieslandCampina, three Netherlands sites | more than $90m | WPC 80%, instantized whey proteins, microparticulated whey | May 2026 |
| FrieslandCampina, Borculo | not disclosed | Doubling WPI and MFGM capacity | not dated |
| FrieslandCampina | not disclosed | Acquired Wisconsin Whey Protein | December 2025 |
| Actus Nutrition with Darigold | not disclosed | Access to Darigold processing infrastructure | June 2026 |
The practical consequence: every one of these adds capacity in 2027 or later. A formulator with a 2026 or H1 2027 launch cannot buy their way out of the current tightness, which is exactly the window in which a second source has option value regardless of its headline price.
What ADM and EVERY actually announced
The joint release of 14 July 2026 says ADM and EVERY will “begin commercial-scale production of EVERY’s OvoPro at ADM’s Clinton, Iowa, facility.” ADM’s Kris Lutt framed it as demonstrating “how ADM’s industry-leading precision fermentation capabilities can complement existing production, expand supply and help customers innovate.”
What the release does not contain is any number. No tonnage, no capacity, no capex, no contract value. The single quantitative claim in it is a PDCAAS of 1.0.
The figures that exist come from interviews around it. EVERY founder and chief executive Arturo Elizondo told Food Ingredients First the partnership “will allow us to increase the available volumes of OvoPro by five times starting in the second half of next year” — H2 2027, not 2026. The “first US site with precision fermentation capacity at this scale” formulation is attributed to EVERY rather than ADM, and Green Queen reports that combined annual production volumes are “under wraps” and that the plant “is set to come online next year.”
Clinton is a corn processing complex — sweeteners, starches, beverage alcohol, corn oil, enzymes, ethanol, animal feed. Elizondo’s stated reason for going this route is the one that matters industrially: “Manufacturing has been a major bottleneck in our industry, largely because wholly owned facilities are often +$200M and have multi-year lead times.”
That is the same conclusion reached by the companies in our fermentation capacity map, and it is the second such arrangement EVERY has announced this year. On 15 June it said an agreement with Huvepharma’s Biovet AD would quadruple its capacity, at a Bulgarian site with “over 9,000,000L of installed fermentation capacity.” That release also states OvoPro is “shelf-stable for up to 24 months” with no cold-chain requirement, and that in the first four months of 2026 the company “secured annual orders worth 550% of its total 2025 order volume.”
Both figures are multiples off an undisclosed base. Four times an unknown number and five times an unknown number remain unknown.
The comparison that cannot be made
EVERY has never published a price per kilogram for OvoPro. We checked the ADM release in two verbatim reproductions, the Huvepharma release, the Food Ingredients First and Green Queen interviews, and targeted searches. There is no figure.
The company is explicit about why. Asked what decides mainstream adoption, Elizondo answered that economics would do the work — but specifically not the per-kilo cost of the ingredient, rather “total cost-in-use over the product lifecycle.” Chief revenue officer Mario Patiño made the parallel argument that “the barriers now are operational, not technical,” listing supply assurance, process integration and total cost-in-use.
Cost-in-use is a legitimate frame. It is also unauditable from outside. A buyer running a trial can compute it; a reader cannot check it; and a competitor cannot be undercut by a number nobody has seen.
The only public price point anywhere near this ingredient is a consumer proxy. Healthier Comforts sells an animal-free egg white protein powder confirmed as 100% OvoPro at the equivalent of $3.00 per ounce, against the same brand’s conventional egg white powder at $2.85 per ounce and its pea protein powder at $1.36 per ounce. That is a retail pack with brand margin, not an ingredient quote, and it should not be read as one. It is simply the only number that exists.
The practical consequence: treat “competitive with whey” as an untested claim until a supplier quote is in hand. Request pricing at the inclusion rate and volume you would actually buy, and ask whether it is indexed to anything — because the argument for switching is built on a commodity’s volatility, and an ingredient that reprices alongside that commodity solves less than it appears to.
Regulatory position
EVERY’s proteins are cleared in the US, and the notices are worth naming precisely because the two are different production strains. The notifier of record is Clara Foods Co. — listed under that name on the earlier notice, and as Clara Foods Co. DBA The EVERY Company on the later one.
| Notice | FDA substance description | Organism | Date of closure |
|---|---|---|---|
| GRN 967 | Soluble egg-white protein | Komagataella phaffii strain GSD-1209 | 9 September 2021 |
| GRN 1104 | Egg-white protein | Komagataella phaffii ATCC GSD-1235 | 17 October 2023 |
GRN 1104 is the later and broader notice. FDA’s response letter records that Clara describes the substance as “a phosphoglycoprotein containing predominantly recombinant ovalbumin,” and sets out specifications of protein greater than 75% w/w, with carbohydrate no more than 10%, fat below 0.4%, moisture no more than 10% and ash no more than 5%. Estimated eaters-only dietary exposure is 8.6 g per person per day at the mean and 16.3 g at the 90th percentile, based on 2017–March 2020 NHANES data.
Three constraints in that letter matter to a formulator. The intended uses exclude infant formula and any product under USDA jurisdiction. Use in a standardised food is permitted only where the applicable standard of identity allows it. And the ingredient requires allergen labelling under the FD&C Act because it contains protein derived from egg — that last is FDA’s own wording, and the notice records Clara’s position that the protein “will likely elicit an allergic response in egg-allergic consumers.”
None of that is a footnote. It constrains which products the ingredient can enter and what the pack has to say, regardless of how the economics come out. Our earlier survey of GRAS clearances for fermentation-derived protein sets out how narrowly these letters are worded.
The egg framing does not survive the data
Egg volatility is the hook in most coverage of this ingredient, and it is the weakest part of the case.
The BLS series for Grade A large eggs, US city average, peaked at $6.227 a dozen in March 2025 and had fallen to $2.191 by May 2026. The 4 August account states prices “have been climbing again since bird flu resurfaced.” The published series does not show that: June 2026 came in at $2.141 and July 2026 at $2.189, both below May, as of the 12 August release. The claim may hold for wholesale or spot markets, but not for the retail series in the same sentence.
The cull figure is similarly awkward. That account describes “more than 12 million laying hens culled since the spring.” The nearest figure we could verify — approximately 12.4 million commercial layer hens depopulated in 2026 to date, against 5.4 million in the whole of the second half of 2025 — was already published on 10 March 2026, before spring. Either the timeframe is wrong or there is a later figure we could not locate.
Meanwhile the US table egg flock stood at 307.9 million hens as of 1 February 2026, up 2.9% year on year, with 2026 production projected up 4.6% and wholesale demand estimated 40–50% below the 2010–2024 baseline.
EVERY’s own positioning is more honest than the coverage. Elizondo describes coexistence rather than replacement, calls the product an “adder,” and identifies protein fortification — not egg substitution — as what is driving a large share of demand. That is the whey story, which brings the argument back to a benchmark of $12 to $13.
The counter-argument
The case for quoting WPC 34% is not empty. It is the published, continuously quoted series with a public history; WPC 80% appears as a commentary range rather than a formal quote, so it is harder to cite and harder to chart. Directionally the two move together — when 34% is tight because production is constrained, 80% is tighter still. A journalist reaching for “the price of whey” will find the 34% series first, and it does describe something real.
There is also a reading in which the cheap grade is the right comparator: if the alternative protein is being weighed against total protein cost across a portfolio rather than against the specific fortification grade, a commodity index has some relevance.
We do not find that persuasive. The applications EVERY names — cakes, cookies, performance bakery, protein snacks, enriched doughs at 10 to 20 grams per serving — are fortification applications. The competing input in those formulations is WPC 80% or isolate. Using a six-times-cheaper grade as the comparator understates the case for the ingredient while making it look like the case is being overstated, which is the worst of both outcomes.
What we could not establish
- The July 2026 WPC 34% figure of $2.18. We could not read it on any USDA page. The 13 August report gives a mostly-range top of $2.2400; the value is plausible as a July monthly average or a late-July mostly top, but it is unconfirmed. USDA has published no 2026 monthly-average file we could retrieve.
- Whether the cited monthly figures are mostly-range averages. The identification is near-certain on the arithmetic, but the source does not state a basis.
- Any tonnage for OvoPro, at Clinton or Bulgaria. Green Queen states volumes are “under wraps.” The public record contains only multiples off an undisclosed base.
- Any capex or contract value for the Clinton arrangement. Not in the release. ADM separately announced a “multimillion-dollar investment to upgrade corn receiving capacity” at Clinton in May 2026, with no figure disclosed; we have no basis for connecting the two.
- A permalink on adm.com for the 14 July release. We verified its text through two independent verbatim reproductions but could not locate it in ADM’s own newsroom.
- The reported second partner. The 4 August article’s headline and URL name Rich’s, but the body never mentions Rich Products or any second partner, and we found no release from either company describing such an agreement. We are not asserting one exists.
- The “$11bn in announced US dairy expansion” figure, which appears in that article without attribution and does not come from the 30 June capacity piece it would naturally draw on.
- “Fermentation-derived rennet in over 90% of US cheese production” and “global egg consumption around 1.3 trillion annually” — both repeated without a primary source.
- OvoPro shelf life. Sources conflict: “roughly a year and a half” in the 4 August article versus “up to 24 months” in EVERY’s own June release. We have used the company’s figure and flagged the discrepancy.
What to watch
Three claims here are testable, and we intend to check them.
Does the H2 2027 date hold? Elizondo’s five-times volume increase starts in the second half of 2027, not on commissioning. Slippage in this sector is the norm rather than the exception, and this is a dated, attributable commitment.
Does WPC 80% ease when the new lines land? DMK’s 7,000 tonnes, Saputo’s 35% uplift at Ripon and FrieslandCampina’s Netherlands investments should show up in the USDA commentary during 2027. If $12 to $13 persists into 2028 with that capacity commissioned, the constraint is worse than a processing shortfall.
Does a price ever get published? The strongest possible move for a supplier confident of its economics is to publish an indicative ingredient price at a stated volume. Until one does, every comparison in this category is an assertion — and, as we found when looking at what it takes to switch to a fermentation-derived protein, assertions are not what gets a formulation through a stage gate.