Close
Carbon Markets

Carbon markets can pay farmers if three conditions hold

(14 min read)

Make It Make Sense is a UNDP Youth4Climate session series, and its third session asked whether carbon markets can deliver for the climate and for the communities that host projects. Four voices shaped it: Leticia Guimaraes of UNDP on Article 6 and high integrity markets, Tom Hayes of Gold Standard on certification and verification, Vaibhav Jain of TerraTrac on satellite monitoring of rice methane, and Ege Ertugrul presenting questions from Youth4Climate members. This analysis tests their claims against 2025 market data, the first Article 6.4 issuance, a per hectare model of rice methane credits and Kenya’s community share rules, and it ends in a conditional yes.

Seven findings

  • Value up, volume down. 2025 retirements fell 4.5% to 168M, yet value rose to about US$1.04B on higher prices.
  • Three price tiers. Spot averages US$6.10, while forward offtakes imply about US$160. The market is pricing quality and durability.
  • A 40% haircut. The first Article 6.4 credits were about 40% fewer than old CDM rules would give, and are already contested.
  • A wide rice range. AWD methane cuts are reported from 14% to 80%. Measurement, not the practice, is the weak link.
  • About US$9 a hectare. At spot price and a mid case cut, our model gives roughly US$9 per hectare per season. Verification must cost far less.
  • Percent is not payment. Kenya’s 40% and 25% shares depend on the base they are applied to.
  • Conditional yes. Markets can finance verified cuts where credibility, verification and fair payment all hold.

What a credit really claims

The session brought three views to one question. Leticia Guimaraes of UNDP covered Article 6 and the principles of high integrity markets: additionality, no double counting and fair benefit sharing. Tom Hayes of Gold Standard explained certification, monitoring and verification. Vaibhav Jain of TerraTrac showed satellite monitoring and improved rice practices cutting methane with farming communities. Ege Ertugrul presented the pre session poll of Youth4Climate members’ questions.

“Carbon markets are not a silver bullet.”Central takeaway, session follow up note

To see why, start with what a credit is. A credit does not measure a tonne removed from the air. It is a claim about the difference between two worlds: the one with the project and the one that would have existed without it. Nobody can observe the second world, so every credit rests on an estimate. That single fact explains the whole integrity debate, and it yields five tests that every credit has to pass.

TestThe questionHow it fails
AdditionalityWould this reduction have happened without credit revenue?Credits paid for what was happening anyway
BaselineIs the “without project” world estimated honestly?Inflated baselines create credits from nothing
LeakageDid the problem simply move elsewhere?Emissions rise next door
PermanenceWill the benefit last?Stored carbon is released later
No double countingIs the tonne claimed once?Seller and buyer countries both count it

A sixth test sits beside these, and the poll was likely to press on it: do the people who host the project share fairly in the money? The session note names three conditions that cover all six: credible reductions, robust monitoring and verification, and fair benefits. The rest of this analysis tests each one against data.

A market splitting into tiers

In 2025 retirements fell 4.5% to 168 million credits, but buyers paid more per credit, an average of US$6.10, and market value rose to about US$1.04 billion. A bridge shows which forces mattered.

What moved the voluntary market in 2025

Change in market value from 2024 to 2025, US$ million

Fewer corporate retirementsHigher average price paidMore compliance demandNet change, 2024 to 2025−US$86M+US$101M+US$59M+US$74M

Price paid added more than lower volume took away. Publications start from slightly different 2024 bases (about US$966M here, about US$980M in the headline 6% figure).

Buyers are paying for something other than volume. Sylvera finds a quality premium emerging, especially for tree planting and avoided deforestation credits, with wide price spreads inside each project type. The forward market shows the same shift on a much larger scale: offtake agreements, promises to buy future credits, reached about US$12.25 billion in 2025.

Promises to buy versus credits actually retired

Forward offtakes announced and 2025 spot value

US$ billion. Different kinds of money, so do not add them

Forward offtakes, 2023Forward offtakes, 2024Forward offtakes, 2025Spot market value, 20252.564.5012.251.04

Price per credit

Log scale, US$

$1$10$100$1,000Spot, average price paidForward offtakes, impliedUS$6.10about US$160

A 26 fold gap between the two rungs. Our arithmetic from published averages.

Three tiers, three different bets

TierPrice signalWhat buyers are paying for
Spot, market averageUS$6.10Credits retired now, across all quality levels
Quality rated spotPremium, variesHigher rated tree planting and avoided deforestation credits
Forward durable removalsabout US$160Scarce, long lived removal supply contracted years ahead

Why the tiers matter to farmers and communities. The headline growth sits mostly at the top of the curve, in durable removals, which agricultural methane projects do not supply. A rice or smallholder project selling near the market average is on the bottom rung. Its route upward is the same route the data reward: demonstrable quality, which is what Article 6 principles and certification are meant to provide.

Sylvera’s own estimate is that value could triple without any volume growth if quality pricing spreads. That is an argument for investing in verification, not for issuing more credits.

Article 6: the accounting that stops double counting

Guimaraes’ second principle is the least visible and the most important. Two countries can easily both claim the same tonne: the host says it cut emissions, and the buyer says it bought the cut. Article 6 solves this with corresponding adjustments, a ledger rule that moves the tonne from one country’s books to the other’s.

How a corresponding adjustment stops double counting

One tonne, one claim. Stylised from the first Article 6.4 issuance

Host countryBuyer countryProject is hosted hereAdds the sold tonne back toits reported emissionsPays for the creditCounts the tonne towardits own climate target1 credit = 1 tCO2eOnly one country’s books carry the tonne. The accounting, not the credit, prevents double counting.

The first real test came in February 2026, when the Article 6.4 Supervisory Body approved the first issuance under the Paris Agreement Crediting Mechanism: 60,000 credits from a clean cooking programme in Myanmar. Credits serve both South Korea’s and Myanmar’s national targets through adjustments, and the reductions credited came out roughly 40% below what old Clean Development Mechanism rules would have given, thanks to updated values and more conservative calculations.

  • 60,000 credits in the first Article 6.4 issuance
  • 0.036% of 2025’s 168M retirements (our arithmetic)
  • 165+ host approved projects waiting to transition

Integrity is already contested in public. In October, ahead of the Supervisory Body’s Bonn meeting of 5 to 9 October, more than 130 civil society organisations asked it to suspend the project’s credits and investigate. They cited an analysis suggesting Article 6.4 credits could exceed real reductions by a factor of seven. We have not tested that analysis. What can be said is that conservative baselines and open challenge are now both visible in the system’s first output.

Reading the first issuance. A 40% haircut is evidence that the rule makers chose accuracy over volume, which supports the first condition. At the 2025 market average, 60,000 credits are worth about US$366,000 (our arithmetic; Article 6 units may trade higher). The architecture is large and the first output small.

Signals to watch: whether the second and third issuances keep the same conservatism, how the Supervisory Body answers the October letter, and how the transitioning pipeline of 165+ projects is treated. Agriculture and forestry methodologies face tougher baseline and permanence questions than cookstoves do.

Standards: what certification buys

Hayes’ argument is that a credit is only as good as the chain of custody behind it: project certification, monitoring against an approved method, independent verification, and a public registry entry that ends in retirement. The 2025 price data suggest buyers now pay for that chain.

The Gold Standard Impact Registry shows what transparency looks like for a reader, because anyone can inspect registered projects by sector and country. Three questions travel well to any credit: who verified it, against which baseline, and where is it retired? A credit that cannot answer all three belongs on the bottom rung.

Rice: a claim you can test, and a price you can compute

Jain’s example is useful because the science is old and the measurement is hard. Paddies emit methane because flooded soil starves decaying matter of oxygen. Rice is about 8% to 12% of human caused methane depending on source and year, and methane warms 27 to 34 times more than carbon dioxide over a century. In alternate wetting and drying (AWD), farmers flood to 5 to 7 cm, wait until water falls 10 to 15 cm below the surface, then reflood. The same practice can cut water use by up to 30%.

How much methane does AWD actually save?

Reported reduction versus continuous flooding, percent

On-station trials, AsiaOn-farm trialsADB summary of studiesIPCC 2006 defaultArkansas field, direct sensingAll gases per hectare, 16 studies14 to 80%, mean 43%20 to 70%30 to 50%48%64.5%35 to 41%0%25%50%75%100%

Methane rows are blue and green; the amber row counts nitrous oxide too. Sources differ in method, so compare the spread, not individual rows.

The spread matters more than the average. A credit priced on a default of 48% is a different claim from one measured at 14% or 80%. Field and satellite monitoring replace a regional average with a measurement, which is the case TerraTrac makes. Two further cautions apply: early AWD can raise nitrous oxide, so honest accounting reports both gases, and one review of 16 field studies, not yet peer reviewed, finds the System of Rice Intensification cuts emissions per kilogram of rice by more than AWD alone, because yields also rise.

The per hectare arithmetic

What is the credit worth to a farmer? Using the IPCC 2006 default of about 1.3 kg of methane per hectare per day for continuously flooded rice, a 100 day season and a warming factor of 27, one hectare emits about 3.5 tonnes of CO2 equivalent per season. Our model applies the three reductions from the chart.

What a hectare of AWD rice could earn per season

Gross credit revenue per hectare, US$. Our model, assumptions in the table below. The US$20 price is a hypothetical premium, not a market quote

Low case, 14% methane cutMid case, 43% cutHigh case, 80% cut$3.00$9.83$9.21$30.19$17.13$56.16At the 2025 average, US$6.10At a hypothetical US$20 premium
CaseCredits, tCO2e per haRevenue at US$6.10Revenue at US$20Community share at 20% effective
Low, 14% cut0.49US$3.00US$9.83US$0.60
Mid, 43% cut1.51US$9.21US$30.19US$1.84
High, 80% cut2.81US$17.13US$56.16US$3.43

Model assumptions and what they imply. The baseline rate, season length and warming factor are standard published values, but real baselines vary with region, water regime and organic inputs, so treat the table as an order of magnitude. The US$20 price is hypothetical.

The implication is blunt. At spot prices a mid case hectare earns roughly US$9 gross per season, so any monitoring, verification and registry cost above that figure per hectare erases the revenue. Projects survive only through aggregation across thousands of hectares, cheap satellite and sampled field measurement, and a move up the quality tiers. The practice works; the economics are about measurement cost.

Who gets paid: percent versus payment

Kenya’s Climate Change (Carbon Markets) Regulations, in force since 17 May 2024, require land based projects on public or community land to pay an annual social contribution of at least 40% to the community, and non land based projects at least 25%. A Community Development Agreement and a community committee govern spending. The open question is the base: one legal summary describes a share of aggregate earnings, while analysts quoted by DevelopmentAid warn it may apply after costs and wages, shrinking what arrives.

The headline share versus the effective share

Community share as a percent of gross credit revenue, if the percentage applies after project costs. Illustration, our arithmetic

Project costs 0% of revenueProject costs 30%Project costs 50%Project costs 70%40%28%20%12%25%17.5%12.5%7.5%Land-based, 40% headlineNon-land-based, 25% headline

Two ways percentages mislead. First, the base. If costs take half of revenue and the share applies afterwards, a 40% rule delivers 20 in every 100 earned. The cost ratios in the chart are assumptions, and the legal base decides which reading is right.

Second, the scale. Applying a 20% effective share to our mid case rice hectare gives about US$1.84 per hectare per season at spot prices. A fair share of a small pie is still small. Fairness therefore depends on three things together: the base of the percentage, the price tier the project reaches, and whether verification costs leave anything to share.

Auditing the speakers’ claims

Leticia Guimaraes, UNDP

Claim. High integrity markets rest on additionality, no double counting and fair benefit sharing.

Test. Article 6.4’s first issuance used a 40% haircut and adjustments, which is real evidence. More than 130 groups still contest it. The claim holds in design and is unproven in practice.

Tom Hayes, Gold Standard

Claim. Certification, monitoring and verification keep credits honest.

Test. Prices now reward quality and registries are public. What buyers cannot yet see is verification cost per smallholder field, the figure our rice model says decides viability.

Vaibhav Jain, TerraTrac

Claim. Satellites plus better rice practices cut methane and engage farmers.

Test. Methane cuts are well supported, from 14% to 80%. Field level measurement is what narrows the range. Nitrous oxide reporting and the amount reaching farmers remain open.

Youth4Climate poll

Claim. Young participants’ questions and concerns should shape the debate.

Test. Their sharpest concern, who benefits, is the least settled of the three conditions.

The verdict, and what would change it

ConditionSupported byUnproven
Credible reductionsConservative Article 6.4 baselines; a price premium for qualityResolution of integrity challenges; the 14% to 80% rice spread
Robust verificationPublic registries; satellite and field measurementVerification cost per smallholder hectare; nitrous oxide accounting
Fair benefitKenya’s legislated 40% and 25% sharesThe calculation base; money paid, and when

Our answer is a conditional yes. Carbon markets can finance verified reductions where all three conditions hold, and prices have begun to reward that. They do not replace cutting emissions, and they fail communities when measurement or payment is vague. Three observable signals would move us: a second Article 6.4 issuance with the same conservatism, published measured reductions for agricultural credits, and disclosed payments to communities.

What we would ask for

  1. Publish measured reductions, with ranges, for every agricultural credit, not regional defaults.
  2. Report nitrous oxide beside methane for water management credits.
  3. State whether community shares apply to gross or net earnings, and publish amounts and dates paid.
  4. Publish Community Development Agreements so communities can see what they signed.
  5. Report issuance haircuts and corresponding adjustments in one standard format.
  6. Treat verification as shared infrastructure, so its cost does not fall on the farmer.

Keep learning

The speakers recommend the Learning for Nature course on operationalizing Article 6.2, the UNDP High Integrity Carbon Markets Toolkit and the Gold Standard Impact Registry. The Youth4Climate community continues the discussion, and the next session covers geospatial technologies for climate action, the toolkit that makes field level rice measurement possible.

Glossary

  • Additionality. The reduction would not have happened without credit revenue.
  • Offtake. An agreement to buy future credits at an agreed price.
  • Retirement. Permanently using a credit so it cannot be sold again.
  • Corresponding adjustment. A ledger change that moves a tonne from the host country’s books to the buyer’s.
  • MRV. Measurement, reporting and verification of a reduction.
Sources & notes

SOURCES: UNDP Youth4Climate, Make It Make Sense Session 3 follow up note (supplied by the author); Sylvera, State of Carbon Credits 2025, and full report; Lloyds Bank, Voluntary Carbon Market newsletter, April 2026; Carbon Credits, The carbon credit market in 2025; UNFCCC, first Article 6.4 issuance, 26 February 2026; Global Forest Coalition, open letter, 6 October 2026; Down To Earth, 130 groups urge suspension; CGIAR, AWD practice brief; Runkle et al., AWD methane by eddy covariance; Dahlgreen and Parr, AWD and SRI emissions (preprint); Dhaka Tribune, How to stop rice fields producing so much methane; ADB, rice methane webinar; Phys.org, CIAT trial on nitrous oxide; Oraro & Company, Carbon Markets Regulations 2024; EY, Kenya publishes the Regulations; DevelopmentAid, Kenya’s carbon registry. The IPCC 2006 baseline rate of about 1.3 kg CH4 per hectare per day, the 100 day season and the per hectare revenues, price ratios, 0.036% share, US$366,000 value and effective community shares are our own model and arithmetic, to be verified against local data. Speaker positions are as summarised in the session note.