Carbon credits have a fundamental credibility problem: the majority of them don’t deliver the climate benefits they promise. A 2024 study in Nature Communications found that 87% of carbon credits purchased by major companies carry a high risk of not providing real emissions reductions. The system is plagued by flawed accounting, perverse incentives, and a pricing structure that makes it cheaper to buy offsets than to actually cut pollution.
Most Credits Aren’t “Additional”
The core promise of a carbon credit is additionality: the idea that the emissions reduction wouldn’t have happened without the money from selling the credit. A company pays for a wind farm in Brazil, and the claim is that the wind farm only exists because of that payment. But in practice, most of these projects would have been built anyway. Renewable energy in Brazil, China, and India already enjoys favorable policy conditions, subsidies, and market demand that make projects financially viable on their own. The offset revenue is a bonus, not the deciding factor.
This isn’t a minor technicality. If a project would have happened regardless, the credit attached to it represents zero additional climate benefit. The company buying it gets to claim it “offset” a ton of CO2, but that ton still sits in the atmosphere. The Nature Communications analysis found that forest conservation and renewable energy projects were the worst offenders, and these are also the most commonly purchased credit types.
The Phantom Credits Problem
Forest-based carbon credits have attracted the most damning scrutiny. A joint investigation by The Guardian and academic researchers examined Verra, the world’s leading carbon credit certifier for the voluntary market. The findings were stark: more than 90% of Verra’s rainforest offset credits were likely “phantom credits” that did not represent genuine carbon reductions. A further analysis put the number at 94% having no measurable benefit to the climate. Only a handful of projects showed real evidence of reduced deforestation.
These credits work by estimating how much deforestation would have occurred without the project, then issuing credits for the “avoided” emissions. The problem is that the baselines are inflated. Projects routinely overestimate the threat of deforestation in their area, making it look like they’re saving more forest than they actually are. The result is millions of credits entering the market that correspond to trees that were never in danger of being cut down.
Leakage: Deforestation Just Moves
Even when a forest protection project genuinely prevents logging in one area, the economic pressure to harvest wood doesn’t disappear. It shifts. This is called leakage: protecting forest in one region causes compensatory harvesting in another, unprotected region. When wood supply drops in one area, markets respond by sourcing it from somewhere else, often from forests with no monitoring or protection at all.
Leakage is notoriously difficult to measure and almost impossible to prevent within a voluntary credit system. A project can show real results within its own boundaries while contributing to deforestation just across the border. The net climate benefit, once leakage is factored in, can be close to zero.
Double Counting Across Borders
Carbon credits create a peculiar accounting challenge. When a company in Europe buys a credit from a forest project in Indonesia, both the company and Indonesia have an incentive to count that reduction on their own ledger. The company claims it toward its net-zero pledge. Indonesia could claim it toward its national climate targets under the Paris Agreement. The same ton of CO2 gets counted twice, but only one ton was actually reduced.
The Paris Agreement’s Article 6, finalized at COP26 in 2021, attempted to fix this by requiring “corresponding adjustments.” When a credit is transferred between countries, the selling country is supposed to add those emissions back to its own inventory so the buyer can subtract them. But full implementation has been slow. Negotiations at COP29 added provisions to flag credits that are “inconsistent” with national targets, but the system still relies on countries voluntarily adjusting their books, and many host countries are reluctant to do so because it makes their own climate pledges harder to meet.
Credits Are Priced Far Below Real Costs
The average voluntary carbon credit sells for roughly $10 to $25 per ton of CO2, with some as cheap as $2.55. Meanwhile, the actual cost of permanently removing a ton of CO2 from the atmosphere using technologies like direct air capture runs between $400 and $1,000 per ton. That gap reveals the core economic dysfunction of the market: the cheapest credits are the ones most likely to be worthless, and companies have strong financial incentives to buy them.
The Nature Communications study confirmed this pattern. Companies have predominantly sourced low-quality, cheap offsets rather than investing in higher-cost options with genuine climate impact. When a company can “neutralize” its emissions for a few dollars per ton instead of hundreds, the offset becomes a PR tool rather than a climate solution. It costs far less to buy credits than to redesign supply chains, switch energy sources, or invest in real removal technology.
The Permission-to-Pollute Effect
Beyond the technical failures, carbon credits create a psychological and strategic problem. They allow companies to continue emitting greenhouse gases while claiming climate responsibility. A company that buys offsets can delay the harder, more expensive work of actually reducing its own emissions. This is sometimes called “mitigation deterrence,” and it may be the most damaging effect of the credit system: not just that the credits themselves are flawed, but that their existence slows down real decarbonization.
When an airline offers you a $5 offset at checkout, the implicit message is that your flight’s climate impact can be erased for the price of a coffee. It can’t. But the availability of that option makes it easier to avoid confronting the actual scale of the problem.
Reform Efforts and Their Limits
The Integrity Council for the Voluntary Carbon Market (ICVCM) has developed a set of Core Carbon Principles meant to raise quality standards. These include stricter additionality requirements, permanence guarantees, independent third-party verification, transparent public registries, and explicit rules against double counting. Credits that meet all ten principles would theoretically represent real, lasting emissions reductions.
The challenge is adoption. These standards are voluntary. No company is required to buy only high-integrity credits, and the market still rewards the cheapest option. Until regulations mandate minimum quality standards or impose penalties for using phantom credits in climate claims, the same incentive structure that produced the current mess will remain in place. The principles exist on paper. Whether they’ll reshape a market built on convenience and low prices is a different question entirely.