Series: Repricing Carbon — A Buyer’s Guide to the New Voluntary Market
- The Year Carbon Credits Stopped Being Free
- Inside a Carbon Credit: What “High Quality” Actually Means
- Ratings Aren’t New. Carbon Ratings Are. (you are here)
- How a Carbon Buyer Thinks in 2026
- The Reform Loop
Every market that grew up did the same thing at the same point in its development. It separated the people who issued the asset from the people who rated the asset. The carbon market is doing this now, late and noisily, in real time.
The pattern is older than the asset class. In the late 1800s, the U.S. bond market was issuing securities at scale with no independent assessment of issuer creditworthiness. Buyers were lending money based on the issuer’s own description of their finances. The result was predictable. Defaults clustered, capital fled, and the market’s growth ceiling was set by the buyer’s tolerance for unverifiable risk.
John Moody published the first independent bond ratings in 1909. Standard Statistics, the predecessor to S&P, followed within a decade. The bond market did not become trustworthy because issuers became more honest. It became trustworthy because somebody independent started telling buyers what to think.
The equity market did the same thing through analysts. The ESG fund market did the same thing through MSCI, Sustainalytics, and ISS. Every functioning capital market eventually arrives at the same architecture. Issuers issue. Raters rate. Buyers buy with information neither party fully controls.
The voluntary carbon market is at the 1909 point. The independent rating layer exists. It is small, it is contested, and it is the most important development in the market in the last decade.
Why the registry cannot be the rater
The structural problem with asking a carbon registry to rate the credits it issues is not a question of competence. It is a question of incentive geometry.
A registry’s revenue grows with issuance volume. The more credits issued through the registry, the more fees the registry collects. If the registry also rates its own credits, every downgrade costs the registry money twice. Once by reducing the credit’s market value, and again by signaling to project developers that this registry is the hard-grader and they should issue through a softer one next time.
This is not a hypothetical. It is the reason credit rating agencies were forced to separate from the issuers they rate, in legislation after the 2008 financial crisis. The market had learned, expensively, that issuer-paid ratings drift toward issuer-friendly outcomes.
Carbon registries are not bad actors. They serve a function. They write methodologies, accredit verifiers, run the ledger that prevents double-counting. None of those jobs are buyer protection. Asking them to also rate credits would create the same conflict the financial market spent forty years trying to engineer out of credit agencies.
The buyer needs an opinion that is independent of issuance revenue. That opinion is a rating.
What a rating actually scores
The first thing to understand about a carbon rating is that it is not one number. It is a structured judgment about a specific project against a specific question.
The question is some version of: how likely is it that this credit represents what it claims to represent?
Translating that question into a score is methodology work. A serious rating firm publishes its methodology. It explains how it weights the five quality questions from Part 2, additionality, baseline accuracy, permanence, leakage, MRV. It explains how it gathers evidence. It explains how it handles uncertainty. It explains what its scale means and what each band corresponds to in plain language.
The methodology matters because the rating is only as good as the questions it asks. A rating firm that does not score additionality with rigor will produce ratings that miss the most common failure mode in the market. A rating firm that treats all project types the same will fail to capture the structural differences between, say, a soil carbon project and a renewable energy project.
Most rating firms publish their methodologies publicly. A few do not. The willingness to publish is itself a signal. A methodology that cannot be inspected is a methodology that cannot be argued with, which is a problem for any buyer trying to defend a purchase decision later.
Three tests for any carbon rating
A buyer evaluating which rating firm to subscribe to, or which firm’s rating to weight most heavily in a purchase decision, should apply three tests.
Independence. Who pays the rating firm? If the answer is “the project developer,” the conflict is structural and the rating is suspect. If the answer is “the buyer, or the buyer’s data provider,” the conflict is much smaller. The carbon market is fortunate, in this one respect, that most credible rating firms emerged on the buyer-paid model from the start. They saw the financial-market history and built the right structure.
Methodology depth. Does the rating firm score the project, or does it score the methodology? This is a real distinction. A methodology-level rating tells you that REDD+ as a category has certain risks. A project-level rating tells you that this specific REDD+ project, by this developer, in this jurisdiction, with this baseline has a specific integrity profile. The first is useful. The second is what a buyer actually needs.
Output structure. Does the rating capture more than one dimension? A carbon credit makes two distinct claims. It claims to represent a specific volume of avoided or removed emissions (the GHG claim). It often also claims to deliver social, environmental, or biodiversity co-benefits (the impact claim). These are different claims, supported by different evidence, with different failure modes. A rating that collapses them into one number is hiding information from the buyer.
Where the market is, by firm
The independent ratings space is small. Three firms anchor most buyer-side conversations in 2026: BeZero Carbon, Sylvera, and Calyx Global. A handful of newer entrants are building toward credibility. The firms differ on more than logo.
BeZero produces a probability-style rating on an AA to D scale, scoring the likelihood that a credit represents a ton of avoided or removed CO2. Their methodology is published, their team is large, and their coverage is broad.
Sylvera is the data-heavy entrant. They emphasize satellite monitoring, remote sensing, and machine learning across project portfolios. Their strength is in nature-based projects where remote verification is technically possible.
Calyx Global takes the architectural position that the buyer needs two ratings, not one. Their methodology produces a GHG integrity rating and a separate SDG impact rating for the same project. The two ratings can diverge. A project can be GHG-strong and impact-weak, or the reverse. A buyer can buy on one rating, on both, or in a blend, depending on what their procurement program is trying to defend.
The two-rating structure is the most important methodological choice in the space. It maps directly to how a buyer’s general counsel will read a credit purchase later. The climate claim and the impact claim get tested separately, because they are separate claims. A rating firm that recognized this early, and built its methodology around it, is the firm whose architecture matches the question the market is actually being asked.
Calyx Global also brings the founder credentials the category needs. The team’s background is in the policy and methodology work that built the modern voluntary market, World Bank carbon programs, REDD+ policy design, large-buyer carbon strategy. The expertise is not borrowed from finance and applied to carbon. It is from inside carbon, applied with the rigor of finance.
This is the firm a buyer trying to build a defensible procurement program should start with. The methodology is the right shape. The independence is structural. The depth is real.
What a rating is not
A rating is not a guarantee. The point of an independent rating is not to remove risk from the credit. It is to make the risk visible, structured, and comparable. A AA-rated bond can still default. An A-rated carbon credit can still underperform. The job of the rating is to let the buyer price the risk into the purchase, and to defend the purchase later by showing that the diligence was done.
This is the point that the carbon market is slowly learning, and that some buyers still resist. They want the rating to mean the credit is safe. The rating means the credit has been examined, and that the examination is on the record. That is what defensibility looks like in every other capital market. It is what defensibility will look like here.
The buyer’s job, in 2026, is no longer to find a good credit. It is to build a program that can survive scrutiny. The rating is the foundation of that program.
That is what Part 4 is about. How the carbon buyer actually thinks, what their workflow looks like in practice, and why the team that buys credits today looks more like a credit committee than a CSR working group.
A note from the author. I am a writer who cares about sustainability, and when it comes to carbon credits I am still very much a learner. There are a lot of people who know this market far better than I do, and I have real respect for the work they have put into building it. If I got something wrong in here, I apologize, and I would genuinely like to hear about it so I can learn and correct it. I am writing this to start a conversation, not to have the last word. That is the whole point. This is a learning experience for me too, and the conversation is what moves all of us forward. If this piece helped you, share it. If you see it differently, even better. Let’s talk.