The party that rates an asset cannot be the party that sells it.

Every mature financial market eventually arrives at the same rule, usually the hard way, after a crisis has made the lesson expensive: the party that grades an asset cannot be the party that sells it. Bonds learned it. Audited accounts learned it. Structured credit had to relearn it in 2008. Nature finance has not learned it yet. In the carbon and natural-capital markets, the originator still grades its own pool, documents its own additionality and hires its own verifier. The market has imported the certificate, the rating and the registry from older finance. It has not yet imported the one rule that makes any of them trustworthy.

Independence began with a subscription

The modern credit rating was born independent because of how it was paid. In 1909 John Moody published the first volume of Moody's Analyses of Railroad Investments, assigning letter grades to railroad bonds, and he sold those analyses by subscription to investors. The reader of the rating was the payer of the rating. The grader had no stake in whether any particular bond sold, only in whether investors kept trusting the grade enough to keep subscribing. That alignment, the analyst paid by the side that loses money if the grade is wrong, is the original definition of independence in capital markets.

It did not last. During the early 1970s the major agencies shifted from investor-pays to issuer-pays, so that the entity issuing the bond now paid to have it rated. The justification was practical, photocopiers made it hard to charge subscribers for something easily shared, and the conflict was tolerable for as long as the instruments were simple and the agencies' reputational caution held. The arrangement survived for three decades. Then it met an instrument complex enough to hide inside.

2008 was an independence failure, not a maths failure

The structured-credit collapse is often told as a story of models that underestimated correlation. The deeper failure was about who paid for the opinion. Issuers paid the agencies to rate their securities and, in many cases, paid for guidance on how to structure deals to capture the maximum triple-A tranche. The Financial Crisis Inquiry Commission found the agencies' work had been shaped by "the pressure from financial firms that paid for the ratings" and "the relentless drive for market share", and called the agencies "essential cogs in the wheel of financial destruction". A grade bought by the seller had been treated by the whole market as if it were a grade bought by the buyer.

Independence is not a credential you display. It is a question of who writes your cheque, and what happens to them when you are wrong.

Enron taught the same lesson on the audit side

The audit profession had its own version of this reckoning, one year earlier and from a different direction. Enron's auditor, Arthur Andersen, earned roughly 25 million dollars in audit fees and about 27 million in consulting fees from the same client in a single year. The auditor whose job was to challenge Enron's accounts was also selling Enron lucrative advisory work, which meant the watchdog had become a vendor to the firm it was meant to police. When Enron failed, Andersen failed with it.

The legislative response was the Sarbanes-Oxley Act of 2002, signed on 30 July 2002, which prohibited audit firms from providing a list of non-audit consulting services to their audit clients. The reform did not try to make auditors more honest by exhortation. It attacked the conflict directly, by separating the cash flows. That is the pattern every time: a market discovers that good intentions and professional codes do not survive contact with the wrong payment structure, and the fix is always structural, not moral.

Nature finance is grading its own pool

Now hold carbon up against that history. The developer of a project documents its own additionality, the central claim on which the credit's value rests. The same developer selects and pays the validation and verification body that is supposed to test that claim, because across the major standards the verification fee is paid by the project developer. Even where methodologies are data-driven and satellite-based, the baseline is chosen and interpreted by the credited party or its retained agent. The originator grades its own pool, and then hires the examiner who certifies the grade.

This is the issuer-pays model and the Andersen model combined into one chain, before any crisis has forced a correction. The peer-reviewed record already shows the predictable result: claimed avoided deforestation running roughly an order of magnitude above independent estimates, and a 2025 follow-up finding only 19 per cent of projects met their reported targets. Nature finance is at the stage structured credit was at in 2005: the conflict is documented, the early evidence is in, and the market is still pretending the grader is independent because it has not yet had its 2008.

Independence is defined by who pays for the read

The lesson of a century of financial plumbing is narrow and unsentimental. Independence is not conferred by a methodology, an accreditation badge or a conflict-of-interest declaration. It is conferred by the payment structure. Moody's was independent when investors paid. The agencies stopped being independent when issuers paid. Andersen stopped being independent when it sold consulting to the firm it audited. In every case the cure was to route payment for the opinion to the party who bears the loss if the opinion is wrong. Nature finance can wait for its own crisis to teach it this, or it can build the independent layer first.