When a standard reprices a vintage, it reprices your book overnight.

A carbon credit is not a fixed thing. It is a claim that holds only as long as the criteria that issued it still hold. When a standard-setter moves those criteria, the credit does not change. The judgement about it does. And in this market, that judgement can move for an entire methodology and vintage at once, on a single published date, with no warning to the holders. Credit markets built a whole apparatus to manage this kind of event. They call it a ratings migration, and they wrapped it in watchlists, outlooks and gradual notches. The carbon market has the migration without any of the apparatus.

A ratings migration without the watchlist

When a corporate bond is downgraded, almost nobody is surprised. The issuer has usually been on negative outlook for months. A rating agency publishes a watchlist, the spread widens gradually, and holders have time to decide whether to sell, hedge or hold. The downgrade, when it lands, confirms a repricing the market already started.

Carbon credit reclassification works nothing like that. A standard-setter assesses a methodology, and on the day it publishes, every credit issued under that methodology is either eligible or it is not. There is no notch between approved and excluded. There is no published outlook telling you a methodology is one review away from losing its label. The first many holders hear of it is the announcement itself. The information arrives at the same moment for everyone, which is exactly the condition under which an orderly repricing becomes a cliff.

The second difference is correlation. A bond downgrade hits one issuer. A methodology decision hits every project and every vintage built on that methodology at the same time. The exposures that look diversified across dozens of projects in a book can in fact share a single point of failure: the criteria the credits were issued against. When that point moves, it moves the whole cluster.

When a target standard reprices demand

The clearest recent example sits on the demand side. On 11 June 2026 the Science Based Targets initiative published its Corporate Net-Zero Standard V2.0, effective from 1 February 2027. Two provisions matter for anyone holding credits against expected corporate demand. First, the standard does not allow credits to be counted toward a company's Scope 1, 2 or 3 emission reductions. A company cannot claim its value-chain footprint fell because it bought credits. Second, from 2035 large companies are required to cover a rising share of ongoing emissions with carbon removals, starting at 1 per cent in 2035 and climbing toward 100 per cent by the net-zero year, with a minimum slice of that obligation met by long-lived removals rising over the same path. The provisions are summarised by removal supplier Climeworks.

A standard like this does not retire a single credit. It re-sorts the entire demand curve. Categories that buyers were treating as fungible against a net-zero claim are split into counts toward the target, counts only as a contribution outside the target, and increasingly mandatory. A holder whose thesis rested on broad corporate demand for avoidance credits, and a holder sitting on long-lived removals, wake up on the same morning facing very different forward books. Neither credit moved. The criterion that values them did.

The credit does not change when a standard moves. The judgement about it does, and that judgement can move for a whole vintage at once.

The supply side does it too

Supply-side bodies reprice the same way. The Integrity Council for the Voluntary Carbon Market assesses methodologies against its Core Carbon Principles and awards, or withholds, a CCP label. In November 2024 it approved three newer REDD+ methodologies, including Verra's VM0048 and the ART TREES standard. The decision that mattered for existing holders was the one that did not make headlines: the older avoided-deforestation methodologies that issued the bulk of REDD+ credits in circulation, including VM0007, VM0009 and VM0015, were not submitted for assessment and so cannot carry the label. Those older methodologies accounted for around a quarter of all carbon credits retired in 2023, per analysis flagged by Jones Day. One assessment cycle drew a line straight through the existing REDD+ stock.

Aviation drew the same kind of line. Under CORSIA, the ICAO Council decides which programmes and vintages may be used for airline offsetting obligations. The Phase 1 eligibility rules require credits from approved programmes within defined crediting-period start dates and emission-reduction vintages, and crucially a host-country authorisation confirming a corresponding adjustment under Article 6 of the Paris Agreement. A credit that is perfectly valid in the voluntary market can be ineligible for CORSIA simply because it lacks that letter. Same tonne, different door, and the door is shut by an administrative criterion the holder may not control.

Consolidation is reclassification with a delay

Methodology consolidation is the quieter version of the same event. When Verra launched VM0048 on 27 November 2023, it set existing unplanned-deforestation projects on a path to transition off the older self-baselined methodologies and onto jurisdictional risk-based baselines. The change is not cosmetic. Projects transitioning are required to requantify, and the baseline reform was designed specifically to reduce the over-crediting that earlier methodologies produced. A project that issued comfortably under the old rules can emerge from requantification with a materially smaller credit stream. The reclassification is real. It is simply spread across a transition window rather than landing on one date, which makes it easier to miss and no less consequential when it arrives.

Across all four cases the pattern is identical. A criterion that the market treated as settled turns out to be live. When it moves, it moves a methodology or a vintage as a block, the holders learn from the announcement, and the exposures that looked uncorrelated turn out to have shared a single dependency all along.

The exposure has to be visible before the ruling

The defence against a ratings migration was never to predict the downgrade. It was to know, continuously, where each holding sat relative to the criteria that could move it, so the exposure was legible before the event rather than reconstructed after it. Bond books have that. Carbon books mostly do not. The criteria live across the target standards, the integrity council, the aviation scheme and the registries, they change on their own schedules, and most holders only map their book against them when a decision is already on the wire.

An independent, continuous read closes that gap. It means holding a book-level view of where each credit sits against the criteria currently in force and the ones visibly in motion, so that when a standard reprices a methodology or a vintage, the affected slice of the book is already identified rather than discovered from a press release. The decision will still land without a watchlist. The holder does not have to.