Ask a sustainability lead about voluntary carbon credits and you will get one of two answers, delivered with equal certainty. Either the market is a reputational minefield that no serious company should go near, or it is the only mechanism currently moving private capital into climate projects at any scale. Both positions are defensible, which is a sign that the question is being asked at the wrong level of resolution.
The voluntary market is not one market. It is at least four, with different physics, different failure modes and different prices, and they are drifting further apart every year.
Avoided emissions from land use. Crediting a forest that was going to be cut down but was not. The methodological problem is the counterfactual: you are paid for something that did not happen, and proving it requires a baseline that nobody can observe. This is where most of the market’s credibility damage originated, and where price has fallen hardest.
Avoided emissions from technology. Cookstoves, efficient lighting, methane capture at landfills. Somewhat easier to verify because there is a device you can count, but still dependent on assumptions about what the user would otherwise have done.
Biological removals. Planting trees, restoring peatland, building soil carbon. The tonne is real and measurable. The problem is permanence: carbon stored in biomass can burn, rot or be harvested, and a hundred-year guarantee is only as good as the institution making it.
Engineered removals. Direct air capture with geological storage, enhanced weathering, biochar. Expensive, verifiable, and durable on a timescale that matches the problem. Supply is tiny.
Treating these as interchangeable — which a single price index encourages — is the root of most bad purchasing decisions. A buyer who wants to make a net zero claim needs durable removals. A buyer who wants to fund climate action in a supply chain region may reasonably want something else entirely, and should say so rather than dressing it up as neutralization.
Three things happened in quick succession. Investigative journalism found that a large fraction of forest-protection credits from one popular methodology did not represent real reductions. Standards bodies began issuing much stricter guidance on what can be labeled high integrity. And regulators in several jurisdictions started treating carbon neutral product claims as advertising claims subject to substantiation rules.
The net effect was not that the market died. It was that the market split. Prices for undifferentiated avoidance credits collapsed; prices for verified removals with credible permanence rose. Volume moved toward long-term offtake agreements between named buyers and named projects, away from anonymous spot purchases from a broker’s inventory.
The most reliable signal of a serious carbon buyer is not how many tonnes they retire. It is whether they can name the project, the vintage, and the reason they chose that project over the alternative.
Most of the legal risk in carbon credits sits not in the purchase but in the sentence written afterward. “Carbon neutral” applied to a product, on a package, without qualification, is now an enforcement target in several markets. “We reduced our operational emissions by 34 percent and purchased durable removals equivalent to the remaining 1,900 tonnes” is longer, less satisfying to a marketing team, and considerably harder to challenge.
Our advice to clients has converged on a simple rule: write the sentence first, then buy the credits that would make it true. It is remarkable how often that exercise reveals that the credits under consideration would not have supported the claim at all.
Compliance markets are expanding, and the boundary between voluntary and compliance is blurring as jurisdictions begin recognizing certain voluntary units. That will pull quality standards upward, because a regulator will accept a unit only if it can defend the acceptance.
The likely destination is a market where avoidance credits are priced as climate finance and removal credits are priced as an environmental commodity, with little confusion between them. That would be a smaller market by tonnage and a far more useful one. For companies planning a net zero pathway now, the sensible posture is to assume that future, buy accordingly, and treat any credit that only makes sense under today’s looser rules as a stranded asset in waiting.