Walk the aisles of a large grocery store and you will find a word that was barely there five years ago. Regenerative. It turns up on flour, eggs, chocolate, coffee and cotton T-shirts.
Here is what sits behind that word in American law. Nothing.
That is not an opinion. The Congressional Research Service, the non-partisan agency that answers research questions for members of Congress, published a report on regenerative agriculture claims on 29 July 2025. Its finding on the definition question was blunt: according to the Congressional Research Service, “neither term has been formally codified in either federal or state statutes or regulations.” No statute defines it. No regulation defines it. The report also says plainly that foods labeled this way are not certified or governed by any federal or state program.
One number from that report is worth sitting with. An investor network called the FAIRR Initiative looked at 79 publicly listed agriculture and food companies with roughly $3 trillion in combined annual revenue, about a third of the global agri-food sector. Fifty of them mentioned regenerative agriculture in their financial disclosures. Four had set a baseline against which progress could be measured.
That gap, between how often the word is used and how often anything is counted, is what this article is about.
What the congressional researchers found
The Congressional Research Service works for Congress, not for an administration or a party. Its job is to describe a policy question accurately enough that a member of either party can rely on it. In July 2025 it took up regenerative agriculture and regenerative organic agriculture as marketing terms and asked a narrow question: what law governs them?
The answer ran through several agencies and came out the same way each time. The Food and Drug Administration regulates most food labeling under the Federal Food, Drug, and Cosmetic Act, and it does not define either term. It also does not pre-approve food labels, so nobody at FDA reads the word before it reaches a shelf. The agency can act against a claim it finds false or misleading, which is a different thing from having a standard the claim must meet.
The Department of Agriculture has not defined the terms for labeling either. The report quotes a USDA scientist saying regenerative agriculture “seems to be defined differently by different people,” with most of those definitions built around the goal of regenerating soil. USDA’s 2022 climate adaptation plan did not reference the terms at all.
Congress has looked at the question and not finished it. In the 118th Congress, one bill would have identified 32 existing conservation practices as regenerative. Another would have defined a regeneratively produced product by reference to soil health principles, and a regeneratively organic product by reference to the Organic Foods Production Act of 1990. Neither became law.
California went furthest. Its State Board of Food and Agriculture finished recommendations for a state definition in January 2025, and the agriculture department drafted language built on soil health and target outcomes. The recommendations were never written into state law, and the draft says it is not meant for certification. No other state has tried.
The practices people usually mean
Strip away the marketing and a fairly consistent set of farm practices sits underneath.
Reduced tillage comes first. Tillage is the mechanical breaking and turning of soil before planting, which kills weeds and prepares a seedbed but also exposes soil to wind and rain and burns through the organic matter holding it together. No-till means planting directly into undisturbed ground. Reduced till means fewer or shallower passes.
Then cover crops: a crop planted to protect and feed the soil rather than to harvest, usually in the window between cash crops. Then diverse rotations, meaning more than a two-crop cycle. Then integrating livestock onto cropland, so manure and grazing become part of the system. Many definitions also push toward fewer purchased inputs, and some exclude synthetic pesticides and fertilizers entirely. What those practices measurably do to soil, as opposed to what they are claimed to do, is article 737’s subject.
The Congressional Research Service draws a useful distinction between two kinds of definition. A process-based definition lists the practices a farm must use. An outcome-based definition lists the results a farm must achieve, such as soil health, water quality or carbon storage, and allows any method that gets there. The two are not interchangeable. A farm can follow every listed practice and show no measured improvement, and a farm can improve measurably using methods a process list would reject.
Here is a correction that rarely makes the marketing. Critics quoted in the report point out that several of these practices are already standard on well-run conventional farms. The federal conservation practice standards NRCS publishes include no-till and reduced till, and the report calls those standards broadly applicable to most conventionally produced agriculture. These are not a separate category of farming. They are choices on a spectrum most American farms sit somewhere on already.
Who sets the rules when the government does not
A private seal stacked on a federal one
The most structured thing in this field is Regenerative Organic Certified, run by the Regenerative Organic Alliance, a nonprofit. Its own description covers three areas: soil health and land management, pasture-based animal welfare, and fairness for farmers and farmworkers. It awards three levels, Bronze, Silver and Gold.
The structurally interesting part is what it requires underneath. According to the Congressional Research Service, the certification requires USDA certified organic status as its base, with the regenerative requirements layered on top. That nesting matters, because the two layers behave completely differently.
The bottom layer is federal regulation. The National Organic Program is administered by USDA’s Agricultural Marketing Service under the Organic Foods Production Act. AMS accredits third-party certifying organizations, those certifiers inspect and certify farms, and certification status is published in a federal database anyone can search. Violate the organic rules and there is a federal complaint process behind it. What those rules require of a farm is article 735’s subject.
The top layer is a private contract. The regenerative requirements above the organic base are written by a nonprofit, audited under its own program manual, and enforceable the way any private standard is, which is to say by the certifier withdrawing the certificate. A problem with the organic part of a nested claim goes to USDA. A problem with the regenerative part goes to the certifying body, or to the Federal Trade Commission as an advertising question.
It is also one scheme among many. The report’s footnotes list a crowd of others, including Certified Regenerative by A Greener World, FoodChain ID’s Regenerative Agriculture Standard, Regenified, and Ecological Outcome Verified under Land to Market. They do not share a definition, a threshold or an audit method.
Underneath all of them, verification comes in three grades, which the Congressional Research Service describes without ranking. A claim can be self-enforced by the producer, meaning the farm attests that it followed the practices. It can be verified by the group the producer belongs to, which is an industry body checking its own members. Or it can be verified by an independent third party with no stake in the result. All three can sit behind the same word on the same shelf, and nothing on the package tells them apart.
The company programs
Most regenerative acreage sits inside large company and retailer programs, and the report names them because naming them is the only way to describe the field.
Walmart’s initiative requires suppliers to follow production principles covering soil and nutrient management, pest management, and water and land protection. Whole Foods Market’s requires practices that improve soil health, including year-round living roots and rotational grazing. ADM’s is built on five land management principles addressing soil disturbance, living roots, soil cover, diversity and inputs. Cargill’s supports farmers adopting soil health practices. General Mills runs its program inside a wider greenhouse gas and water effort. Nestlé, Danone, Unilever and PepsiCo released a joint framework in 2023 covering water, soil, biodiversity and climate.
Describing those programs is not the same as assessing the farms inside them. On that the report is restrained and specific: most labeling schemes do not tend to monitor or evaluate the environmental and social benefits their programs describe. The FAIRR audit fills in the shape. Of the 50 companies that mentioned regenerative agriculture in disclosures, 18 quantified a company-wide target, 8 discussed metrics and data, and 4 set a baseline.
The only federal rule that reaches the word
One federal instrument touches a regenerative claim, and it is not a definition. It is the Federal Trade Commission’s Guides for the Use of Environmental Marketing Claims, at Title 16 of the Code of Federal Regulations, Part 260, usually called the Green Guides.
Read the first section and you find an unusual admission. The guides state that they confer no rights on anyone, do not bind the FTC or the public, and do not preempt federal, state or local law. They are the Commission’s stated view of when an environmental claim becomes deceptive under Section 5 of the FTC Act, which prohibits false advertising that is misleading in any respect. The FTC can still act if a claim is inconsistent with them.
Two sections do real work here. Section 260.4 covers general environmental benefit claims, the green and eco-friendly family of phrasings. Its reasoning is that an unqualified claim of that kind gets read many different ways by different shoppers, and a marketer is unlikely to be able to support every reasonable reading. Its examples are sharp: calling a gas mower eco-smart after a 0.1 percent fuel efficiency gain is misleading even though the gain is real.
Section 260.6 covers seals. Its key move is that a seal which does not tell you what it is based on probably reads as a general environmental benefit claim, with all the substantiation problems that carries. It also says third-party certification does not remove the seller’s own obligation to support everything the seal communicates, and treats a self-awarded seal, or a trade association seal presented as independent, as deceptive unless the relationship is disclosed.
Here is the gap. The Green Guides were last revised in 2012, and the Congressional Research Service reports that they do not address regenerative or regenerative organic claims. The hook exists, but it runs through a general deception standard rather than any definition. A company does not have to meet a regenerative standard. It has to avoid saying something misleading about whatever it is doing.
USDA now has a working definition, for payments rather than packages
Something changed after the report was written, and it is worth reporting precisely because it is easy to overstate.
In December 2025, NRCS launched a Regenerative Pilot Program. The agency calls it a farmer-first, outcomes-based approach to conservation, and the program page carries a working description of regenerative agriculture as a conservation management approach emphasizing natural resources through improved soil health, water management and natural vitality.
The program bundles practices into a single application across the Environmental Quality Incentives Program and the Conservation Stewardship Program. A participating farm completes a whole-farm assessment, adopts at least one of 15 named primary practices, and agrees to soil health testing in the first and last year of the contract at a minimum, which sets a baseline and records what changed. USDA has put $1 billion behind it for fiscal 2027, $650 million through EQIP and $350 million through CSP, raised from an earlier $700 million.
The 15 practices come from the existing conservation catalogue and carry their ordinary code numbers: cover crop is 340, conservation crop rotation is 328, no-till is 329, reduced till is 345, nutrient management is 590, grazing management is 528, and the rest run through mulching, stripcropping, contour farming and irrigation water management. NRCS publishes a national standard for each one, setting minimum science-based criteria, then tells you not to use the national standard directly. Each state adapts it into a Field Office Technical Guide that can be stricter than the national version but never looser.
Now the limit. This is a conservation funding program. It decides who receives federal payments and how results get measured. It is not a labeling standard, and nothing about it travels onto a package. A farm can satisfy it completely and never print the word. A company can print the word without any farm in its supply chain having gone near it.
Carbon storage: contested in both directions, settled in neither
Soil carbon is the claim carrying the most weight and the least agreement. The report lists improved soil carbon and climate mitigation near the top of what advocates claim, notes that critics dispute the scientific basis, and reports that soil health and carbon were the outcomes companies cited most often. The peer-reviewed picture says a measurable effect exists and its size is genuinely unresolved.
A 2019 meta-analysis in Global Change Biology pooled 3,049 paired measurements from 417 published papers. Conservation tillage raised soil organic carbon by roughly 5 percent and cover crops by roughly 6 percent, and the authors open by saying existing measurements diverge widely. A 2020 analysis in Science of the Total Environment, working from 1,928 data pairs across 243 studies, found no-till raised soil organic carbon stocks about 11 percent in the top 30 centimetres, with a larger effect in the upper half of that profile than the lower half.
That depth pattern explains much of the disagreement. Two careful teams sampling the same field to different depths can reach different conclusions, because carbon accumulating near the surface under no-till may be redistributed rather than added.
A 2025 paper in Scientific Reports ran a multi-model ensemble across 46 million hectares of US Midwest cropland. No-till combined with cover crops raised soil carbon stocks by 0.36 megagrams of carbon per hectare per year against a business-as-usual baseline, with that gain roughly halved when either practice was used alone, and fully realized only where the starting carbon stock was low. The part worth quoting back at anyone selling certainty is what happened when the authors added nitrous oxide emissions: the net climate benefit ranged from zero to 3 megagrams of carbon dioxide equivalent per hectare per year. The bottom of the range is nothing.
Scale up and the spread grows. A 2021 Global Change Biology study put a realistic global figure for improved cropland management at 0.28 to 0.43 gigatonnes of carbon a year, against published estimates running from 0.1 to 2 gigatonnes. That is a twentyfold range among serious researchers.
So what is established? That these practices usually add carbon to the upper soil in most settings. What is not: how much, how fast, how deep, how durably, whether it survives a return to tillage, and whether it offsets the nitrous oxide that comes with nitrogen. A 2025 review in Frontiers in Nutrition arguing strongly for wider adoption concedes the same structural problems, listing an unresolved operational definition, the lack of standardized certification, and limited supporting research among the barriers.
Six questions to put to the seal
If you want to know what a package is telling you, the word is the wrong place to look. The seal beside it is the right place. These six will get you there.
- What is the certification called, by name, and who issues it?
- Is anything federally enforceable underneath it, such as USDA organic?
- Who verifies compliance: the farm itself, a group the farm belongs to, or an independent auditor?
- What specific outcome is claimed, measured against what baseline, and when was the baseline set?
- Does the seal say what it covers, or only suggest that something good happened?
- Does the claim on the package match the claim on the company’s own standards page?
If you think a claim crosses into false advertising, the Federal Trade Commission takes consumer complaints about it. If the organic part of a nested claim looks wrong, that goes to USDA’s Agricultural Marketing Service, which runs the National Organic Program and publishes certification status.
If you farm, the advice runs the other direction. Before signing a program contract, talk to your local NRCS service center or a land-grant extension agronomist about the transition specifically. The soil carbon and water benefits take years to appear, several studies show a yield penalty in the early years of no-till, and the economics of those first seasons are the part marketing material skips.
Read past the word to the certification
The absence of a definition does not mean the practices are worthless. That is the inference most people draw and it is the wrong one. The measured effects on water movement, erosion resistance and topsoil carbon are real, documented in federal literature reviews and peer-reviewed meta-analyses, and in many settings substantial.
What the absence means is narrower. Nobody has to earn the word before printing it. There is no threshold, no audit requirement and no agency checking. Two products carrying it side by side may be separated by an independent annual inspection on one and a supplier questionnaire on the other.
So read past it. Find the certification name, find who audits it, and find whether anything enforceable sits underneath. Two minutes on a phone in the aisle covers it, and it is the only part of the exercise that returns information.
Sources
- Congressional Research Service. “Regenerative Agriculture and Related Food Product Labeling and Marketing Claims (R48610), 29 July 2025.” everycrsreport.com
- U.S. Government Publishing Office (eCFR). “16 CFR Part 260, Guides for the Use of Environmental Marketing Claims.” ecfr.gov
- U.S. Government Publishing Office (eCFR). “16 CFR 260.4, General environmental benefit claims.” ecfr.gov
- U.S. Government Publishing Office (eCFR). “16 CFR 260.6, Certifications and seals of approval.” ecfr.gov
- U.S. Department of Agriculture, Agricultural Marketing Service. “National Organic Program.” ams.usda.gov
- U.S. Department of Agriculture, Natural Resources Conservation Service. “NRCS Regenerative Pilot Program.” nrcs.usda.gov
- U.S. Department of Agriculture, Natural Resources Conservation Service. “Conservation Practice Standards.” nrcs.usda.gov
- Regenerative Organic Alliance. “Regenerative Organic Certified.” regenorganic.org
- PubMed. “Bai X et al. "Responses of soil carbon sequestration to climate-smart agriculture practices: A meta-analysis." Global Change Biology, 2019. PubMed. DOI: 10.1111/gcb.14658.” doi.org
- PubMed. “Li Y et al. "Residue retention promotes soil carbon accumulation in minimum tillage systems: Implications for conservation agriculture." Science of the Total Environment, 2020. PubMed. DOI: 10.1016/j.scitotenv.2020.140147.” doi.org
- PubMed. “Basso B et al. "A multi model ensemble reveals net climate benefits from regenerative practices in US Midwest croplands." Scientific Reports, 2025. PubMed. DOI: 10.1038/s41598-025-08419-y.” doi.org
- PubMed. “Lessmann M et al. "Global variation in soil carbon sequestration potential through improved cropland management." Global Change Biology, 2021. PubMed. DOI: 10.1111/gcb.15954.” doi.org
- PubMed. “Rosier CL et al. "From soil to health: advancing regenerative agriculture for improved food quality and nutrition security." Frontiers in Nutrition, 2025. PubMed. DOI: 10.3389/fnut.2025.1638507.” doi.org