Tokenizing Carbon Credits

On October 2, 2024, three federal agencies filed actions arising out of the same carbon credit facts. The CFTC charged fraud in connection with voluntary carbon credits. The SEC settled charges over a $250 million offering fraud. The Justice Department indicted the former chief executive on wire fraud, commodities fraud, and securities fraud.

One company, one set of cookstove projects, one set of inflated credits. Three different characterizations, and all of them were right.

Sponsors ask me whether tokenized carbon is a commodity or a security. That case answers the question, though not in the form they want. The credit and the instrument you sell get separate answers, and both regulators can show up in the same week.

The rest of this post works through a single hypothetical program to show how that plays out in a structure you might actually build.

The deal we will follow

Assume a sponsor owns thirty-eight thousand acres of hardwood timberland in eastern Kentucky. It has registered an improved forest management project with Verra, committing to lengthen harvest rotations and carry more standing inventory than the regional baseline, and it earns credits on the difference. Call it the program. It needs eight million dollars to buy an adjoining tract and build out the measurement and verification work the methodology requires. Its plan has three parts.

First, raise the eight million by selling preferred units in the entity that owns the timberland to roughly forty accredited investors.

Second, once Verra issues credits against the first verification period, tokenize two hundred thousand of them. One token equals one credit. The credits sit in a registry account, and any holder can redeem a token and retire the underlying credit against their own emissions. Call this the credit token.

Third, launch a second token backed by a managed basket. The sponsor buys credits across projects and vintages, decides what enters and leaves the basket, and charges 1.5% a year. Call this the pool token.

The founder wants all three to trade on the same decentralized exchange, because that is where the buyers are.

Three instruments, one business. They do not get the same answer, and working out which is which determines the entire architecture.

What the three agencies actually charged

Read the 2024 filings side by side and the structure of the market becomes obvious.

The CFTC’s theory ran to the credits themselves. It alleged that the developer delivered false, misleading, or inaccurate reports to a credit certifier in order to receive an inflated quantity of voluntary carbon credits, and it settled with the company for a one-million-dollar civil penalty while litigating against the former chief executive. Those were the first CFTC fraud actions in the voluntary carbon market.

The SEC’s theory had nothing to do with credits. Its order describes a company that began marketing its equity to institutional investors in 2022 and made material misrepresentations about its business, culminating in a $250 million offering fraud. The company admitted the facts and acknowledged that its conduct violated the federal securities laws.

Map that onto our program. The credit token corresponds to what the CFTC went after. The preferred units correspond to what the SEC went after. They are different instruments with different regulators, and the same overstated project data would have supported both cases at once.

That last point is the one sponsors miss. One overstated growth measurement on the timber inventory is a commodities problem in the credit token and a securities disclosure problem in the preferred units. Tokenization doesn’t change this. It adds a third instrument to track, and each one gets its own analysis, because tokenizing an asset changes the format of the interest and not the obligations attached to it.

Carbon credits are commodities

The Commodity Exchange Act defines a commodity broadly enough to reach goods, articles, services, rights, and interests in which contracts for future delivery are dealt. Carbon credit futures have traded on CFTC-regulated exchanges for years, which settles the question as a practical matter.

The agency then went further. In September 2024, the CFTC approved final guidance for designated contract markets listing voluntary carbon credit derivative contracts, outlining factors exchanges should consider when addressing Core Principle requirements in contract design and listing.

Then it reversed course. On September 10, 2025, the CFTC withdrew that guidance entirely, saying Section 5c of the Commodity Exchange Act and Parts 38 and 40 already establish the framework for listing derivative contracts, and that singling out carbon contracts had created a disproportionate focus likely to cause confusion. Nothing replaced it.

Anyone still citing the 2024 guidance is citing a withdrawn document. That matters less than it sounds, since the guidance never governed spot credits anyway. It matters a great deal if the citation is sitting in your offering memorandum.

Our program never had a claim to it in the first place. The guidance applied to futures exchanges listing carbon derivatives. The program is selling spot credits in token form, which the guidance never touched.

The CFTC’s authority over the credit token is narrower than it looks

If you’re selling tokens that represent credits instead of futures on credits, no CFTC registration regime applies to you.

That sounds like freedom. It’s closer to a vacuum.

What the CFTC retains over spot commodity markets is anti-fraud and anti-manipulation authority. Return to the program. Nobody has to register the credit token, nobody reviews it, and no license exists to apply for. But if the monitoring data behind those two hundred thousand credits turns out to be overstated, the CFTC can bring exactly the action it brought in 2024. The sponsor has no registration obligation and no compliance regime to point to afterward.

For a sponsor who wants certainty that’s the worst of both. There’s nothing to comply with, and nothing to show a regulator that you did.

Two other CFTC issues reach the credit token. If the program lets retail buyers acquire tokens on margin, on leverage, or with financing from the seller, section 2(c)(2)(D) of the Commodity Exchange Act treats the transaction as if it were a futures contract unless actual delivery occurs within twenty-eight days. And if the sponsor hedges basket price exposure with carbon futures, commodity pool operator analysis follows for the pool token.

Running the three instruments through the securities analysis

Now take each piece of the program in turn.

The credit token

The buyer acquires specific identified credits held in a registry account. The token evidences that ownership and transfers with it. Nobody manages a portfolio, there is no promised return beyond what the credits are worth, and the holder can pull the credits out and retire them.

This has the strongest argument for sitting outside the securities laws. It’s also the hardest of the three to build, for registry reasons I’ll get to.

The pool token

The sponsor acquires credits across projects, vintages, and methodologies, manages quality, decides when to sell, and charges 1.5%. Holders own a claim on a portfolio somebody else assembles.

That’s a securities offering. The underlying being a commodity is irrelevant, exactly as it is for a managed vehicle holding metal or grain. What the buyer owns is an interest in a business someone else runs, and the management fee is a large part of what creates that conclusion.

Carbon makes the valuation problem worse than most. Credits differ by methodology, vintage, registry, and buyer perception of integrity, and there is no single reliable reference price for a heterogeneous basket. If the program commits to a net asset value in its documents before working out how it will strike that number, it inherits a recurring problem, because pricing illiquid tokenized interests is a documentation question long before it is an accounting one, and thin secondary trading makes mark-to-market pricing for a tokenized instrument harder still.

The preferred units

This is where the 2024 case lived, and it catches sponsors who think they’re in the carbon business and not the capital markets business. Raising eight million dollars to buy timberland and fund verification is a securities offering in any form, tokenized or not, and every statement about expected credit yield per acre is securities disclosure.

Notice the overlap. The projection the program shows a credit buyer is the same projection it shows a preferred unit investor. One document, two regulators, which is why the securities risk carried by a pitch deck is a live exposure for a carbon developer who has never thought of himself as an issuer.

Where each instrument can actually trade

The venue question is where these programs break, and it follows mechanically from characterization. Apply it to the founder’s plan to list all three on one decentralized exchange.

The pool token cannot go there. It’s a security, so secondary trading belongs on a national securities exchange or an SEC-registered alternative trading system, and whoever matches buyers and sellers needs broker-dealer registration. There is no carbon exception, and the SEC staff has been consistent that recording ownership on a ledger doesn’t change the analysis.

The preferred units cannot go there either, for the same reason, with resale restrictions on top.

The credit token can, in the sense that no federal venue requirement stops it. Then the state money transmission analysis arrives, along with the retail leverage rules, and the practical problem that no serious counterparty will trade an instrument whose legal status nobody has opined on.

So the founder’s plan produces one tradable token out of three, and it happens to be the one the registry may not permit. Carbon buyers are global even when the project is not. A Kentucky forestry credit gets bought by a European airline as readily as an American one, which means the analysis for non-US holders belongs in the structure from the beginning and not at the point an overseas buyer turns up in the whitelist.

The registry can veto the whole thing

This constraint catches the most technically sophisticated sponsors, and it has nothing to do with securities or commodities law.

In May 2022, Verra prohibited creating instruments or tokens based on retired credits, on the reasoning that retirement means the environmental benefit has been consumed. That landed directly on the bridging model, which had worked by retiring credits and minting tokens against them. Verra proposed an alternative, immobilization, under which a credit would be frozen in a registry subaccount while a token circulated, and ran a public consultation. A framework for tokenizing live credits has been slow to arrive. The American Carbon Registry has moved toward its own rules with a formal tokenized state sitting alongside live and retired.

Our program is registered with Verra. That single fact determines whether the credit token can exist at all, and no securities analysis reaches it.

Registry program rules are a gating item. If the registry won’t recognize your token, your token is backed by nothing the market will accept, whatever the legal opinion says.

Confirm the registry’s current policy in writing before you build. These policies have changed more than once and are still moving.

What actually decides characterization

I work through the same questions on every one of these.

What does the holder own: identified credits, an interest in an entity holding credits, or a contractual claim against the issuer? Can the holder redeem and take the underlying credits into their own registry account? Who selects which credits the program holds, and can holders replace that person? Is there a management fee, a performance fee, or a spread earned on redemption? Does the marketing discuss returns, appreciation, or a future trading market? Is the sponsor also raising capital to fund the projects generating the credits? Does the registry’s current policy permit the token to exist at all?

In practice the answers cluster. Most programs described to me as simple credit tokens turn out to have a manager, a fee, and a discretionary portfolio, which puts them where the pool token sits.

Each token structure gets its own regulatory answer

Go back to the program and count what it actually built. One commodity claim the CFTC can sue over but nobody registers. One security that cannot touch the venue the founder picked. One securities offering the founder never thought of as an offering. Three instruments, two regulators, and a registry that holds a veto over the only freely tradable piece.

That is the ordinary shape of a carbon program, not an unusual one. The commodity or security question is the wrong frame because it assumes one answer exists for the whole business.

The sequencing point follows. Characterization determines venue, and venue determines whether the liquidity you’re promising investors can exist at all. A managed carbon token that is a security cannot trade on a crypto venue, no matter how much demand sits there. A pure credit claim trades more freely and is much harder to build, because the registries haven’t finished deciding whether it may exist.

Decide the venue question before the token design freezes. Reverse that order and you end up with an instrument that works technically and can’t legally go anywhere.

If you are structuring a tokenized carbon program and need to determine how each instrument should be characterized before launch, book a call to review the token structure, trading venue, securities exposure, and registry constraints. 

Frequently Asked Questions

Are tokenized carbon credits a security or a commodity?

The credit is a commodity. The token may be either, depending on what it is a claim on. A token evidencing ownership of identified credits held for the buyer has a strong argument against security status. An interest in a managed basket of credits is a security, and both regulators can have jurisdiction over different pieces of the same program.

Does the CFTC’s carbon guidance apply to my token?

There is no carbon guidance left to apply. The CFTC withdrew its 2024 voluntary carbon credit guidance on September 10, 2025, and even before that it reached only designated contract markets listing carbon derivatives. It never governed spot sales of tokenized credits. The CFTC does retain anti-fraud and anti-manipulation authority over spot carbon markets, and has used it.

Where can a tokenized carbon credit legally trade?

If it is a security, on a national securities exchange or an SEC-registered alternative trading system, through registered broker-dealers. If it is a pure commodity claim, no federal venue requirement applies, but state money transmission rules, the retail leverage provisions of the Commodity Exchange Act, and counterparty diligence all constrain where it realistically trades.

Do I need registry permission to tokenize a carbon credit?

As a practical matter, yes. Verra prohibited tokenizing retired credits in 2022 and has explored immobilization instead. Registry program rules determine whether your token corresponds to anything the market recognizes. Confirm the registry’s current written policy before building, because these policies have changed repeatedly.

Can I raise capital for carbon projects and tokenize the credits at the same time?

You can, but you are running two offerings. The capital raise is a securities offering regardless of format, and the credits carry commodity exposure. Statements about project performance become both securities disclosure and potential CFTC exposure, which is precisely the overlap that produced parallel agency actions in 2024.

Is a tokenized carbon fund an investment company?

Usually not on the credits alone, since carbon credits are not securities for Investment Company Act purposes. Holding equity in project developers, interests in other funds, or notes changes that, because those are securities on the balance sheet and count toward the 40% threshold.