Carbon offset markets are quietly outsourcing the work of climate science

The voluntary carbon market is, by funding volume, one of the largest sources of money flowing into forest-carbon measurement on Earth. It is also one of the least scientifically governed. The combination is producing a measurement substrate that is increasingly diverging from what working forest scientists believe to be true. The corrective is not what the market itself is offering.

The incentive mismatch

Registries that certify offset methodologies earn fees per credit issued. The result, predictably, is rule-making that favors high-volume, high-baseline scenarios. There is no countervailing entity whose budget grows when issuance is more conservative.

This is not a new observation. It has been the structural critique of voluntary carbon markets for at least a decade. What has changed is the volume — the markets are now large enough that the methodology decisions they make have systemic consequences for the underlying science. Land-use scientists who once participated in registry technical advisory groups increasingly do not, because the recommendations they make tend not to translate into rules that would reduce credit volume.

What the science says

Independent remote-sensing studies over the last five years have repeatedly found that issued credits over-count avoided emissions by factors of two to ten in specific project types. This is not a fringe critique published by skeptics. It is the modal finding of peer-reviewed evaluation by working land-use scientists, including teams that initially supported the methodologies they later evaluated.

The discrepancies are largest for avoided-deforestation credits, where baseline projections (what would have happened without the project) have proven systematically pessimistic — projecting forest loss that subsequent observation shows would not have occurred. Reforestation credits have fared better in evaluation but face their own concerns around permanence and additionality.

The pattern is not that registries are dishonest. It is that they operate within a structural incentive that makes some kinds of error much cheaper to commit than others.

What an honest correction looks like

The fix is not to pick a better registry. The fix is to fund the institutional infrastructure that the market does not, and cannot, fund itself: open monitoring platforms, calibration sites, independent verification that does not depend on the issuance pipeline.

Several countries have begun building this infrastructure publicly. The European Union's Land Monitoring System is one example; the proposed expansion of the U.S. Forest Inventory Analysis program for carbon-relevant variables is another. These are public goods. They are funded at a fraction of the volume that the voluntary market spends on certification.

Until that infrastructure exists at the scale needed, voluntary credits should be priced as discount credits rather than equivalents to direct emissions reductions. A credit issued under a methodology that has been independently evaluated as overstating its claim by a factor of three is not a one-for-one substitute for a tonne of CO₂ not emitted. Treating it as such is a pricing error that the market is currently unable to correct.

The argument is not against carbon markets. It is against the specific arrangement in which the entities certifying methodologies have a financial interest in the volume of credits those methodologies produce. That arrangement, applied to any other measurement-driven market, would be regarded as a structural problem. It is one here too.