A carbon credit vintage, the calendar year in which the underlying emission reductions or removals actually occurred, now moves prices as much as the project label does. Two credits from the same registry, the same methodology and sometimes the same project routinely clear at very different prices for one reason: the year attached to the serial number. This tutorial explains what a vintage is, where the hard vintage rules come from, shows verified spread data from public market sources, and sets out how to build vintage discipline into a procurement policy.

What a Carbon Credit Vintage Actually Is

The vintage is the year the tonne was reduced or removed, not the year the credit was issued or sold. A 2017 vintage REDD+ credit means the avoided deforestation was measured for the 2017 monitoring period, even if the registry issued the credit years later after a delayed verification cycle. “Newly issued” and “new vintage” are different things, and conflating them is a common procurement error.

Vintage is also not a direct quality score. The Singapore government guidance on carbon credits in corporate decarbonisation (National Climate Change Secretariat, Ministry of Trade and Industry and Enterprise Singapore, updated 28 October 2025) states that vintage “is not a direct indication of quality”, but recommends that companies purchase and retire credits issued within their commitment periods, because newer vintages generally reflect more up-to-date methodologies and baselines.

Where the Hard Vintage Rules Come From

Buyer preferences for recent vintages are not fashion. They are anchored in eligibility windows written into compliance schemes and exchange contracts.

CORSIA is the clearest case. According to Verra’s CORSIA eligibility page, VCUs were eligible for the pilot phase (2021-2023) if the reductions occurred between 1 January 2016 and 31 December 2023. For the first phase (2024-2026), eligibility tightens to reductions occurring between 1 January 2021 and 31 December 2026, and every VCU with a 2021 or later vintage additionally needs an Article 6 “International Mitigation Purposes” label, backed by a corresponding adjustment evidenced in the host country’s Biennial Transparency Report or by an approved insurance product. When the first phase began in 2024, international aviation, the largest compliance demand pool for voluntary-style credits, closed its doors to all pre-2021 vintages. That is a documented, date-bound devaluation event for old stock.

Standardized exchange contracts encode the same logic on a rolling basis. Xpansiv’s CBL environmental markets update for December 2024 specifies that the N-GEO contract prices a basket of Verra AFOLU credits with CCB certification from vintages 2019-2024, with the range rolling forward every year on 1 July. The C-GEO follows the same rolling logic. Credits that age out of the front window migrate to “Trailing” contracts: N-GEO Trailing covers 2016-2018 vintages and C-GEO Trailing covers 2013-2018, with ranges that expand rather than roll. The practical consequence is that a credit does not depreciate smoothly. On a known calendar date it falls out of the benchmark deliverable basket and starts trading against a cheaper contract. A treasury holding credits across a 1 July roll can see the benchmark eligibility of its position change without anything changing in the underlying project.

Verified Price Spreads: What Vintage Does to Price

The figures below come from the CBL order book and settlement data published in Xpansiv’s December 2024 update, the most detailed public snapshot of vintage-by-vintage pricing we could verify. They are historical, November 2024 data, not current quotes.

The cleanest signal is same-project pricing. On the CBL order book at the end of November 2024, credits from the Keo Seima Wildlife REDD project in Cambodia (VCS project 1650) were bid at $2.50 for the 2017 vintage, while the 2016 vintage was offered at $1.50 and the 2015 vintage at $1.25. Same project, same methodology, same standard: a two-year vintage gap doubled the price. The Rio Anapu-Pacaja REDD+ project in Brazil showed the same pattern in miniature, with the 2018 vintage offered at $1.20 against $1.10 for 2017.

At the contract level the pattern is visible but mix-dependent. The front N-GEO (2019-2024 vintages) settled November 2024 block trades at a volume-weighted average of $0.76, against $1.31 for the N-GEO Trailing, a reversal Xpansiv attributes to the composition of credits settled through each contract rather than to a vintage preference inversion. The structural premium sits elsewhere in the same report: over 60,000 recent-vintage Asian reforestation credits transacted between $25.00 and $42.00 per tonne, while bids for older-vintage Mai Ndombe, Southern Cardamom and Kasigau AFOLU credits sat at $0.25-$0.30. That two-order-of-magnitude spread mixes quality and vintage effects, which is exactly why the same-project ladders above are the purer evidence.

Why Vintage Matters More in 2025-2026

Three developments sharpen vintage discrimination. First, oversupply: CORE Markets reported in its May 2025 global report that the pool of non-retired credits had reached 817 million tonnes, while demand “focused on newer vintages (V22, V23)”, partly on expectations of Article 6 usability. In a glut, buyers can afford to be selective, and selectivity concentrates on the young end of the stock.

Second, thin liquidity. CORE Markets’ June 2026 update describes the voluntary market as relatively subdued, with activity concentrating around higher-integrity credits and broader liquidity remaining thin, while CORSIA supply grows only modestly under corresponding-adjustment constraints. In thin markets, vintage discounts are not arbitraged away; they persist.

Third, methodology turnover. Standards periodically revise baselines and quantification tools, and older vintages were quantified under older rules. That is precisely the rationale Singapore’s guidance gives for its commitment-period recommendation, and it is why vintage works as a cheap proxy for methodological currency.

How to Build Vintage Discipline Into Procurement

  • Match vintage to the claim period. Following the Singapore guidance, a company with a 2030 target against a 2020 baseline should source credits issued for 2021-2030. Older vintages remain usable if they are high-quality and consistent with current methodologies, but they carry an extra evidentiary burden.
  • Check eligibility windows before banking inventory. CORSIA phase windows and the CBL contract roll on 1 July are fixed dates. If you buy credits to hold for future claims or resale, model what happens to eligibility when the window moves.
  • Price vintage discounts with same-project comparables. Use registry serial ranges from the same project across vintages, as in the Keo Seima example, instead of comparing across projects where quality differences swamp the vintage effect.
  • Track Article 6 labels for 2021 and later vintages. For voluntary claims a corresponding adjustment is not required, as the Singapore guidance confirms, but CORSIA eligibility now depends on it, and label status increasingly separates premium demand from the rest.
  • Disclose vintages. The same guidance recommends disclosing the vintage of credits used, alongside project ID, programme and methodology, consistent with ISSB-based reporting.

What This Means for Buyers and Sellers

For buyers, an old vintage is not automatically a bad credit, but it is structurally locked out of CORSIA first-phase demand and out of the front exchange baskets, and it must clear a higher diligence bar on methodology currency. For developers and traders holding unsold old-vintage stock, the devaluation events are predictable: phase start dates and the annual 1 July roll are published in advance. Pricing those cliffs into inventory decisions is now a basic competence of carbon market participation, and as of August 2026 the spread between young and old stock shows no sign of closing on its own.