How Carbon Offsets Work

Buying a carbon offset sets off a chain of events that most buyers never see. Money moves to a project, a third party checks that the claimed reduction is real, a registry issues a credit and then retires it in the buyer's name. Each link in that chain can be strong or weak, and the value of the offset depends on all of them. This guide walks the chain from end to end so that a certificate means something specific rather than a vague promise.

Start with the activity being offset

Before any offset is bought, someone has to decide what is being balanced. That is the boundary question again: a flight, a year of a household's energy, a company's scope 1 and 2 emissions or the footprint of an event. The size of that boundary determines how many tonnes are needed, and the boundary should be stated in the claim. An offset without a stated boundary is impossible to check, because there is no way to know what it was supposed to cover. The offsets section explains how the boundary fits into a credible claim.

The project that generates the reduction

The money buys a share of a project. A wind farm that would not have been built without the extra revenue, a tract of forest protected from clearing, a landfill where methane is captured and burned instead of escaping: each of these reduces emissions relative to a baseline of what would otherwise have happened. The project is the physical source of the offset, and its quality is the foundation of everything downstream.

The critical test is additionality, which asks whether the reduction would have occurred anyway. A wind farm that was already profitable without carbon finance is a poor offset, because the credit did not cause the reduction. Establishing what would have happened in the absence of the project is genuinely difficult, which is why credible standards prescribe methods for it and require evidence rather than assertion.

Verification by an independent auditor

Once a project is running, its emissions reductions are measured and then checked by a third party. The auditor is accredited under one of the recognised standards and is independent of both the project developer and the buyer. The audit examines the project's monitoring data, its baseline assumptions and its compliance with the method, and issues a verification statement for a specific volume of tonnes. This step is what turns a claim into a verified claim, and it is repeated periodically because a project's performance can change over time.

Verification is not a one off stamp. Projects are monitored on a schedule, and credits are issued only for reductions that have already been verified. A project that underperforms issues fewer credits, which is exactly the discipline that a functioning system should apply.

The registry and the retired credit

Verified tonnes are issued as credits into a registry, a public database that tracks who holds what. Registries prevent double counting by ensuring a credit can be retired only once. When a buyer purchases an offset, the corresponding credit is transferred and then retired, and the retirement is recorded with a serial number and often with the name of the beneficiary.

This is the point that many buyers miss: the offset only counts once the credit is retired. A purchase that leaves a credit sitting in an account has not balanced anything. The registry entry is the evidence, and it is public. Anyone can look up a retirement and confirm that the credit exists, who issued it and that it is no longer available to be sold again.

What can go wrong along the way

Each link has a known failure mode. A project can be non additional, meaning the reduction would have happened regardless. A forest can burn or be cut years later, reversing the storage, which is the permanence problem. A credit can be sold to two buyers if a registry is not used properly. A project can overstate its baseline, inflating the tonnes. These are not hypothetical worries; they are the reasons the standards, the auditors and the registries exist, and the reasons a buyer should care which standard stands behind a credit rather than only what it costs.

A strong purchase therefore looks like this: a project that clears additionality, monitored and verified by an accredited auditor, issued by a recognised standard, and retired in a public registry with a serial number the buyer can cite. The verification guide shows how to check those details for a specific certificate, and the project types guide explains which families carry which risks.

The price of a credit

Credit prices vary enormously, and the spread is not simply a matter of quality. A methane capture project can be cheap per tonne because the reduction is large relative to the cost, while a removal project is expensive because taking carbon out of the air is harder than avoiding its release. Forestry sits in between. Price also reflects the standard, the vintage of the credit, the location and the demand from buyers with targets to meet. A very cheap credit is not automatically bad and an expensive one is not automatically good, but a price far below the market for a given project type is a reason to ask why. The project types guide explains the cost differences between families.

Where offsets fit in the wider plan

Offsetting is the last step, not the first. The order that holds up is to measure the footprint, reduce what can be reduced, and offset the remainder while saying how large that remainder is. A household might cut home energy and driving and then offset the flights it still takes. A business might improve efficiency and then offset what is left, being clear about which scopes are covered. The household calculator and the business guide give the starting measurement, and the questions page answers the practical doubts about cost, choice and double counting.