What Is a Carbon Credit? How Carbon Markets Actually Work

A carbon credit is a tradable certificate representing one metric tonne of carbon dioxide equivalent (tCO2e) that has been avoided, reduced, or removed from the atmosphere. It is the base unit of carbon markets: a standardized, verifiable claim that a specific action prevented or reversed greenhouse gas emissions.

The concept is simple. A solar farm in India displaces coal-fired electricity, preventing emissions. A biochar project in Kenya pulls CO2 from the atmosphere and locks it in stable carbon for centuries. A forest conservation project in Brazil prevents logging that would release stored carbon. Each of these actions can generate carbon credits, verified by independent auditors and issued by registries like CDR.fyi, Verra, or Gold Standard.

The buyer of that credit can use it to compensate for their own emissions. A company that emits 10,000 tonnes of CO2 and buys 10,000 carbon credits can claim carbon neutrality. Whether that claim is credible depends entirely on the quality of the credits purchased. That distinction between quality and quantity is the central tension in carbon markets today.

Carbon credits exist because of a fundamental economic insight: it does not matter where in the world a tonne of CO2 is reduced. The atmosphere is a single system. If it is cheaper to prevent a tonne of emissions in Vietnam than in Germany, a market mechanism that directs capital to the cheapest reduction delivers the same climate outcome at lower cost. That is the economic logic behind carbon trading. The practical challenge is making sure the credits represent real reductions.

Carbon markets split into two fundamentally different systems. Understanding the difference is essential because they operate under different rules, at different scales, and with different levels of integrity.

Compliance markets are created by governments. A regulator sets a cap on total emissions, distributes or auctions emission allowances, and requires companies to surrender enough allowances to cover their actual emissions. If a company emits more than its allowances, it must buy more on the market. If it emits less, it can sell the surplus. The cap tightens over time, forcing aggregate emissions down.

The largest compliance market is the EU Emissions Trading System (EU ETS), which covers roughly 40% of EU greenhouse gas emissions across power generation, manufacturing, and aviation within Europe. Other major systems include California's Cap-and-Trade Program, China's national ETS (the world's largest by coverage, launched in 2021), South Korea's K-ETS, and the UK ETS (post-Brexit replacement).

Voluntary markets allow companies and individuals to purchase carbon credits without a legal mandate. Buyers are typically corporations pursuing net-zero pledges, airlines offering passengers the option to offset flights, or individuals compensating for personal emissions. Credits on voluntary markets are issued by independent registries, primarily Verra (Verified Carbon Standard), Gold Standard, American Carbon Registry, and Climate Action Reserve.

The scale difference is striking. The compliance market is over 1,300 times larger than the voluntary market. When policymakers discuss carbon pricing as a tool for decarbonization, they are discussing compliance markets. The voluntary market, despite receiving most of the media attention, represents a fraction of a percent of global carbon trading.

This matters because the mechanisms are different. In a compliance market, a credit is an allowance: a permission slip to emit. If the cap is tight, allowances are scarce and expensive, creating genuine economic pressure to decarbonize. In a voluntary market, a credit is an offset: a claim that emissions elsewhere were avoided or removed to compensate for the buyer's emissions. The quality of that claim depends entirely on the methodology and verification behind the credit.

Not all carbon credits are the same. The most important distinction is between credits that prevent emissions from happening (avoidance) and credits that pull CO2 out of the atmosphere (removal). The difference in price, permanence, and credibility is enormous.

Avoidance credits represent emissions that would have occurred in a baseline scenario but were prevented. Protecting a forest from logging avoids the carbon release. Building a wind farm that displaces coal generation avoids the coal plant's emissions. The credit value depends on a counterfactual: what would have happened without the project? This counterfactual is inherently uncertain, which is why avoidance credits face greater scrutiny on additionality (would the action have happened anyway?) and baseline accuracy (how much would have been emitted without intervention?).

Removal credits represent CO2 actively extracted from the atmosphere and stored. Methods include biochar production (pyrolysis locks carbon in stable form for 500+ years at temperatures above 550°C), direct air capture with geological storage (DAC+S), enhanced rock weathering, and afforestation (growing new forests that absorb CO2 as they grow). Removal credits are more expensive because they require energy-intensive processes or decades of biological growth, but they deliver a measurable, verifiable reduction in atmospheric CO2.

The price gap reflects the quality gap. A $3 renewable energy credit is cheap because building renewables is now the default economic choice in most markets, raising serious questions about whether the credit represents additional climate benefit. A $150 biochar credit is expensive because it verifiably removes carbon from the atmospheric cycle and locks it in a geologically stable form.

Major corporate buyers like Microsoft, Stripe, and Shopify have shifted their procurement toward removal credits specifically because of the integrity advantage. Microsoft's 2024 carbon removal portfolio paid an average of $120+ per tonne. Stripe's Frontier fund has committed over $1 billion to advance purchase agreements for carbon removal at prices ranging from $50 to over $600 per tonne. This buyer shift is repricing the market from the demand side.

Carbon credits do not exist until a registry issues them. Registries define the methodologies (rules for calculating how many credits a project earns), manage the verification process (independent third-party audits), and maintain the public ledger that tracks issuance, ownership, and retirement of credits. When a company claims to have "retired" a carbon credit, that retirement is recorded on a registry's public database.

Verra dominates the voluntary market with approximately 65% of all credits ever issued. Its REDD+ forestry credits have also been the most scrutinized, with a 2023 Guardian/Die Zeit/SourceMaterial investigation finding that more than 90% of Verra's rainforest offset credits were likely "phantom credits" that did not represent genuine emission reductions. Verra disputed the findings and has since revised its methodologies, but the reputational damage contributed to a broader crisis of confidence in voluntary offsets.

Gold Standard, founded by WWF, differentiates itself by requiring projects to deliver measurable sustainable development co-benefits alongside carbon reductions. Its credits typically command a price premium over Verra credits for the same project type.

For carbon removal specifically, Puro.earth has become the leading registry, with standardized methodologies for biochar, enhanced weathering, carbonated building materials, and bio-based construction. The CDR.fyi tracker aggregates removal credit data across all registries, showing that carbon removal credit volume grew 166% year-over-year in 2024.

In 2023, the Integrity Council for the Voluntary Carbon Market (ICVCM) launched its Core Carbon Principles (CCPs), a new quality benchmark designed to distinguish high-integrity credits. Credits that meet CCPs must demonstrate additionality, permanence, robust quantification, and no net harm. The ICVCM framework aims to be the voluntary market's quality floor, though adoption is still in early stages.

Carbon credit prices span a 300x range depending on which market and which credit type. That range is not a flaw. It reflects the enormous gap in quality, permanence, and regulatory rigor between the cheapest voluntary avoidance credits and the most expensive compliance allowances or removal credits.

The EU ETS allowance price peaked at approximately €100/tCO2 in early 2023 before settling in the €60-70 range through 2024-2025. The decline reflected increased renewable energy deployment reducing power sector emissions (lower demand for allowances), surplus allowances accumulated during COVID lockdowns, and political uncertainty about the pace of cap tightening. The EU's Carbon Border Adjustment Mechanism (CBAM), which began its transitional phase in October 2023, is designed to prevent carbon leakage by requiring importers to pay for the carbon content of goods entering the EU.

In the voluntary market, average credit prices fell from approximately $10/tCO2e in 2022 to $5-8/tCO2e in 2024. The decline reflects both the integrity crisis (buyers becoming more cautious) and oversupply. Total voluntary market value dropped from approximately $2 billion in 2022 to $723 million in 2023.

Carbon removal credits move in the opposite direction. As demand from major corporate buyers grows and methodologies mature, removal credit volumes are increasing while prices remain high. CDR.fyi data shows biochar credits at $131-164/tCO2e through Puro.earth. Microsoft's 2024 carbon removal procurement averaged over $120/tCO2e. Stripe's Frontier fund has contracted removal at $50-600+/tCO2e depending on method and maturity.

The voluntary carbon market accumulated nearly 1 billion tCO2e in unretired credits by 2024. That surplus represents a structural oversupply: more credits were issued than buyers wanted to purchase and retire. The oversupply itself is a symptom of deeper problems.

The additionality problem. Many credits were issued for projects that would have happened regardless of carbon finance. A wind farm in a country where wind is already the cheapest energy source does not need carbon credit revenue to be built. If it would have been built anyway, the credit represents no additional climate benefit. The buyer pays for a tonne of CO2 reduction that was going to occur regardless.

The baseline problem. REDD+ forestry credits depend on estimates of how much deforestation would have occurred without the project. If a project assumes aggressive deforestation in its baseline scenario, it generates more credits. The 2023 Guardian investigation found that baseline scenarios for Verra-certified REDD+ projects systematically overstated deforestation risk. Credits were being issued for "avoided" deforestation in forests that faced little realistic threat of being cut.

The permanence problem. Forestry credits carry reversal risk. A forest fire, drought, pest outbreak, or policy change can release the stored carbon back to the atmosphere, invalidating the credit. Buffer pools (a percentage of credits held in reserve against future reversals) partially mitigate this, but the 2020-2023 wildfire seasons demonstrated that buffer pools were insufficient for the scale of reversal events.

The double-counting problem. Under Article 6 of the Paris Agreement, countries have agreed that emission reductions cannot be counted by both the country where the reduction occurs and the country (or company) that purchased the credit. Corresponding adjustments are required to prevent double counting. As of 2025, implementation remains patchy. Many voluntary market credits have not been subject to corresponding adjustments, meaning the same emission reduction may be counted toward both a country's nationally determined contribution and a corporation's net-zero claim.

These are not theoretical concerns. They explain why the voluntary market contracted by over 60% in value from 2022 to 2023, and why sophisticated corporate buyers are shifting procurement to removal credits where these problems are structurally less severe.

Carbon markets are not disappearing. They are splitting into two distinct trajectories.

Compliance markets are expanding. The EU CBAM is extending carbon pricing to imported goods, effectively exporting the EU ETS carbon price to trading partners. China's ETS is gradually tightening and may add additional sectors beyond power generation. Brazil launched its regulated carbon market framework in 2024. India is developing its own system. The trend is toward more sectors, tighter caps, and higher prices. By 2030, compliance carbon markets could cover over 60% of global GDP, up from approximately 23% in 2024.

Voluntary markets are bifurcating. The low end of the voluntary market, cheap avoidance credits with questionable additionality, is contracting. The high end, verified removal credits with durable storage and transparent methodologies, is growing. The ICVCM's Core Carbon Principles are designed to formalize this split by creating a quality label that distinguishes credible credits.

Article 6 implementation is slowly building infrastructure. The Paris Agreement's Article 6 creates a framework for international carbon trading between countries, potentially connecting compliance markets across borders. Article 6.4 establishes a new UN-supervised crediting mechanism to replace the CDM (Clean Development Mechanism) from the Kyoto Protocol. Full implementation is proceeding slowly, but the architecture for a global carbon market is being laid.

Removal is becoming the floor, not the ceiling. Corporate science-based targets increasingly require that companies reduce their own emissions first and only use credits for residual emissions that cannot be eliminated. For those residual emissions, the expectation is shifting from offsetting (avoidance) to neutralizing (removal). The Science Based Targets initiative (SBTi) requires companies to neutralize residual emissions with permanent carbon removal. This demand signal is driving investment in removal capacity.

The direction is clear. Carbon pricing is expanding geographically through compliance markets. Credit quality standards are tightening through the ICVCM and registry reforms. Buyer expectations are rising toward removal. The question is no longer whether carbon has a price, but how fast that price rises and how effectively it directs capital to genuine emission reductions.