Back to insights

Quality isn’t binary in the voluntary carbon market

  • Ted Christie-Miller
    Director of Carbon Removal
  • Cara Howse
    Research Associate



Throughout the history of the voluntary carbon market (VCM), actors have tried to shape it into a commodity market. Historically, this was done through initiatives such as the Clean Development Mechanism (CDM) and through the voluntary registries. In practice, these initiatives sought to create a ‘pass’ or ‘fail’ for carbon credits. Once a credit had passed the criteria necessary, it was defined as a commodified tonne of carbon avoided or removed.

There are clear reasons for people wanting the VCM to be a commodity market. It has worked brilliantly at scaling compliance markets, such as the European Union Emissions Trading System (EU ETS), where fungible ‘allowances’ are equal to a tonne of carbon emissions. Actors in the VCM are, by definition, not subject to the same compulsory demand. 

However, the attempted commodification of this market has left carbon projects open to criticisms, because it is incredibly hard in practice to prove that each credit represents exactly one tonne of avoided or removed carbon dioxide equivalent. Most recently, this was exemplified by the Guardian investigation into avoided deforestation credits, which identified a large spectrum of quality in the voluntary carbon market.

Many remain hopeful that the VCM can establish itself as a commodity market. The logic of ‘a tonne is a tonne and all units are fungible’ certainly makes it simple for corporates seeking to retire credits to make net zero claims.

The reality of carbon credits is that the quality and impacts of various credits vary too widely for all credits to satisfy the same claims. In our view, we should move beyond relying solely on the commodity model for how the VCM operates to avoid repeating mistakes of the past. 

The main risk with continuing with the commodification status quo is that the VCM steers, once again, into reputational difficulties as people prove that a certified carbon credit does not necessarily represent a tonne of carbon removed or avoided. Total commodification also risks stifling innovation in new technologies and creating a race to the bottom for the cheapest credits possible.

This paper will set out why quality in the VCM isn’t binary and how to move beyond a market where all carbon credits are treated equally. 

1. Carbon credits are not commodities

A commodity, such as oil or grain, is a raw material or primary agricultural product. They typically trade on deep, liquid, and global markets, giving a variety of participants access to the tools to buy, sell, hedge, or speculate on the underlying goods.  

This can only be achieved by creating fungible instruments, meaning commodities deemed to be of the same grade or quality can be standardised and priced based on a given quantity. For example, a barrel of light crude oil will be priced, via supply and demand, at a set price on the global market. 

While significant amounts of the trading of commodity instruments is done to speculate on or hedge against price movements, buyers can still take physical delivery of the underlying commodity from sellers. This enables ongoing validation that the underlying commodity meets the quality specified in the contract. 

Carbon credits are often referred to as commodities, or the carbon credit market as a commodity market. However, unlike commodities - which are fungible, deliverable products with uniform pricing - carbon credits are intangible assets that cannot be physically delivered and have a wide spectrum of quality. Perhaps most importantly, carbon credits are not raw materials, but rather are private instruments designed to combat global climate change and protect a public good. 

Carbon credits have many similarities to bonds in the financial markets. Bonds represent tradable debt in the form of a loan from the buyer to the issuer. Bond markets offer a means for governments and corporates to raise capital from a variety of investors. These markets are often referred to as fixed income markets, as the financial instruments they are made up of typically pay interest at regular intervals. The buyer is repaid at maturity of the bond. 

Unlike commodities, there is no underlying good or commodity to a bond. It is a purely monetary instrument enabling the seller to raise capital from a variety of sources and repay that capital in a standardised way over a given period of time. 

Analogously, carbon credits are a contract, typically certifying the reduction or removal of a given amount of GHGs and the commitment for it to remain so for the contract's duration. There is no underlying good to be delivered; you can neither take delivery of the GHGs nor measure them perfectly to ensure they have been achieved through the activity. So, carbon credits are much more like bonds than commodities.

While the value of a bond is linked to the perceived creditworthiness of the issuer (i.e. their ability to repay) a carbon credit’s value is derived from the assessed likelihood that a project’s activity achieves the carbon it is issuing credits against.

That said, one key difference between a bond and carbon credit is that carbon credits are not financially settled. The analogy is not perfect, but drawing on the lessons of the world’s biggest and most liquid financial instrument is very useful for thinking about how to create a more robust and scalable VCM.

Commodities marketVoluntary carbon market
Nature of marketPhysical and derivative marketEnvironmental market for carbon credits
Assets tradedRaw materials, primary goodsCarbon credits
Asset typeTangibleIntangible
Measure of qualityStandardised grading system for each commodity type¹Carbon credit rating
Price determinationSupply and demand, global factorsCarbon credit ratings, market demand
Purpose and ImpactHedging, speculation, price discovery, underlying goodsRetiring to make carbon-based claims, raising funding for projects, carbon risk management

Table 1: Comparison of commodity and voluntary carbon markets.

2. Quality isn’t binary in the VCM

The theory of carbon markets assumes that all credits represent emissions avoidance or removal equal to 1 tCOe. The evidence increasingly shows that this is not the case, which has led to market failure, uncertainty, and skittishness on the part of corporate buyers. If carbon credits don’t trade and get retired, then they are not helping the climate.

At a carbon project performance level, we believe all carbon credits are not created equal. At BeZero Carbon, to date we have evaluated 355 carbon projects and offered our opinion on the likelihood of each of those carbon credits achieving a tonne of carbon removed or avoided. Of the credits we’ve evaluated, 51% have a low, very low, or lowest likelihood of achieving a tonne towards climate impact. Our analysis shows that 13% of credits have a high or very high likelihood of achieving a tonne of carbon removed or avoided.

Figure 1. Rating distribution of all-time issuance for BeZero-Rated Projects (as of September 2023), which represents over 42% of total credit issuance across the four major standard bodies

There is a major coordinated effort underway to create a new threshold for quality standards and methods for developing carbon credits. That will send a very important signal to buyers that the market cannot tolerate or accept junk. However, it has become clear that even with well-thought-out methodologies and standards, there are lots of ways for on-the-ground project implementation to fail to pass muster for creating high-quality tonnes. Even once a threshold of standard-level quality is established and proliferates in the market, there will still be huge variations in quality, with only the best actually representing 1 tCOe.

As an example of why project-level analysis is critical for assessing quality, in the following analysis we have raised the bar to credits rated ‘BBB’ and above, which represents the top 22% of projects rated by BeZero Carbon. This pool includes a wide range of credit types. For example, 74% are nature-based credits, 17% are industrial processes, 7% are waste, 2% are energy and 1% are household devices. 

Digging deeper into this upper level of credits, using data from the BeZero Carbon Markets platform, we can see the variations in the risk of individual projects across different credit types. The level and type of credit risk fluctuates significantly. For example:

  • Additionality risk increases as carbon credit quality decreases. All ‘AA’-rated projects have little risk of additionality, whereas, 4% of ‘BBB’ projects face ‘little’ risk, 85% have ‘some’ risk, and 11% have ‘notable’ risk. As additionality is a primary driver of quality in the VCM, this disparity shows the complications when commodifying these credits.

  • Over-crediting risk increases as BeZero Carbon Rating decreases. Over half - 57% of projects rated ‘AA’ have ‘little’ over-crediting risk. Projects rated ‘BBB’ display all levels of over-crediting risk from ‘very low’ (making up 2% of ‘BBB’ projects) to ‘significant’ risk (making up 7% of ‘BBB’ projects).

  • Non-permanence risk varies between upper-level ratings. The majority of upper-level projects  have ‘some’ or ‘notable’ risk of non-permanence. Projects rated ‘BBB’ have a wide spectrum of  risk, from ‘very low’ risk at 2% to ‘significant’ risk, also at 2%.

Figure 2: Rating distribution of additionality risk by projects (as of September 2023).

Deep dive: side by side comparison of a ‘AA’ and ‘BBB’ project

For this case study, we have picked out two projects rated under the BeZero Carbon Rating. One is a mangrove project in Myanmar, issuing ‘AA’-rated removal credits. The other is a renewables project in the Republic of Korea issuing ‘BBB’-rated avoidance credits. 

Both credits are in the top third of all BeZero Carbon-rated projects and would be seen as the ‘upper-level’ of credits on the VCM. When evaluating the different risk factors associated with these credits, large differences emerge - particularly with regards to additionality risk.

Figure 3. Rating distribution of a mangrove project and renewables project by level of risk for BeZero Carbon risk factors.

The mangrove project has faced persistent financial barriers and absence of regulation for the protection of mangrove forest. Financing mangrove reforestation projects in Myanmar has relied upon public grants and funding, since the mangrove forest will not provide revenue from timber sales. Thus, mangrove reforestation in Myanmar relies on carbon finance, reducing additionality risk. 

This is not the case for the renewables project, where government incentives have contributed to the development of onshore wind farms. In 2002, feed-in-tariffs (FITs) were introduced to increase the use of renewable energy, raising the target share from 2.5% in 2013 to 10% in 2022. This directly promotes renewable energy use in South Korea, increasing the risk of additionality for this project.

This case study is just one example of how no two credits are created equal, even at the very top level of the market.

BeZero Carbon’s white paper, Making Credible Claims, makes the point that not all credits are created equal, and introduces a concept for discounting the value of credits with lower ratings. The paper identifies proposed discount rates of 5% for ‘AAA’-rated credits to reflect that there is risk even among the very best credits, going down to 99% for ‘D’-rated credits. BeZero Carbon analysis finds that in an analytical sample from the five major registries, an extra 123% of credits would need to be retired to justify an offset claim, if our risk-adjusted approach was taken into account. Figure 4 outlines how this risk-adjusted approach impacts the claims that could be made for outstanding issuances.

Figure 4: Outstanding credits from BeZero-rated projects from the four major registries with and without this risk-adjusted approach.

These analyses show not only that there is a large array of credit types within the ‘upper-level’ credits, but more importantly the levels of risk are varied and nuanced even amongst these top quality credits on the market. In short, quality isn’t binary in the voluntary carbon market. 

3. Binary labels for quality miss project-level risks

Ongoing market initiatives, such as the Integrity Council for the Voluntary Carbon Market, are seeking to establish new quality labels for the VCM. These proposals are welcome. They raise the bar for what standards are acceptable in the market. They serve as a timely and coordinated attempt to drive threshold standards, the quality of methodologies, and help advise organisations on how to make net zero claims using carbon credits. 

However, we believe that solely relying on a binary quality label risks returning to the expectation that ‘a tonne is a tonne’. We also need a risk-based framework to reflect variations in quality at the project and credit levels. 

As this paper has shown, no credit is 100% risk free, or exactly equivalent to a tonne of carbon removed or avoided. Risk-based analysis at a project level, as provided by ratings, are a necessary and complementary tool to better understand carbon credit quality and build confidence in the market.

Let’s look at another label system that categorises credits as in or out. Over the last few years we have seen the emergence of the Carbon Offsetting and Reduction Scheme for International Aviation (CORSIA) label for carbon credits being used in the aviation industry. CORSIA is a carbon market for reducing emissions from the aviation sector. It will be open for voluntary contribution from 2021 to 2024 and will be mandatory 2027 onwards. 

Our analysis shows that CORSIA credits represent a spectrum of quality. BeZero has rated a total of 27 projects that are CORSIA eligible. These projects fall between ‘BBB’ and ‘C’, with 44% of projects rated ‘B’. The highest-rated CORSIA-eligible project is rated ‘BBB’, which has just a moderate likelihood of achieving a tonne of carbon removed or avoided.

Our analysis finds that many CORSIA eligible credits have some specific additionality issues. 67% of CORSIA eligible projects have ‘significant’ risk of additionality, primarily driven by these projects having a limited reliance on carbon finance. Note, it’s not the CORSIA criteria that caused these projects to be high or low risk, but rather the actual project implementation, where the assessment should be focused.

Figure 5: Rating distribution of CORSIA-eligible projects by BeZero Carbon Rating (as of September 2023).

Figure 6: Rating distribution of additionality risk of CORSIA-eligible projects (as of September 2023).

4. Risks of a binary quality label only

Climate risk

Carbon markets exist to accelerate action to meet net zero. Hundreds of billions, if not trillions, of dollars of private-sector capital could be unlocked for tackling climate change if these markets perform.

Limiting warming to below two degrees requires us to understand how effective measures are at reducing carbon. Solely relying on binary labels or accreditation, instead of coupling them with risk-based analysis, would fail to appreciate the complexity and nuance of these carbon reduction measures. It could restrict the flow of capital to quality because the price mechanism isn’t as effective as it should be.

Companies want to know how much carbon credits can contribute to their sustainability plans and net zero targets. Governments need certainty on how carbon credits can help meet their NDCs. A more nuanced approach is laid in our recent white paper - Making Credible Claims.

Reputational risk

By oversimplifying the evaluation of carbon offset projects into a ‘pass’ or ‘fail’ criterion based on what standard or methodology they were created under, companies still face high risk that they may inadvertently invest in projects and credits that lack comprehensive climate impact and integrity. 

This oversimplification could re-stoke the same reputational challenges that have plagued the VCM to date. Negative exposés in the media have not focused solely on which carbon credit standards have failed; they look in depth at individual projects to show how those projects have failed to deliver the climate impact they promised. Corporates intending to use carbon credits to offset emissions should engage in similar project-level scrutiny before buying credits.

Embracing a more nuanced approach - through mechanisms such as credit-level ratings - that considers various dimensions of a carbon credit project, such as additionality, over-crediting, and non-permanence not only safeguards a company's reputation but also reinforces genuine efforts to drive positive environmental change. 

A multifaceted and granular assessment better reflects the complexities of carbon reduction initiatives and ensures that companies are making well-informed, impactful choices that align with their sustainability goals and resonate with stakeholders.

The risk of not scaling the market

Climate change is complicated and demands a nuanced approach, accommodating diverse and innovative solutions. Oversimplification risks stifling potential emissions reduction avenues and discouraging companies with unique circumstances from participating.

The risk of solely relying on a binary measure of quality is that it could cause a race to the bottom. If every credit above a certain level is going to ‘pass’ as a tonne of carbon removed or avoided, why would buyers seek to invest in innovative, less certain and more expensive credits? The perverse outcome of this commodification of the market will be that buyers would go for the cheapest credit possible which has the ‘quality’ label attached.

To foster a thriving voluntary carbon market that attracts diverse participants and fosters innovation, it is imperative to adopt a more risk-based and comprehensive evaluation system that embraces the complexity of climate solutions and encourages a dynamic ecosystem of carbon reduction efforts. At BeZero, we are eager to work with fellow actors in the VCM to integrate both high-quality standards and project-level risk assessments into the market.

5. Conclusion

The voluntary carbon market is at an inflection point. Two paths lie ahead. 

It could continue operating like a commodity market, which assumes that all credits will be created perfectly equal. The problem with this approach is that confidence in the market can be seriously undermined by any evidence that proves an individual project isn’t delivering on its climate impacts. 

The alternative path is one where the market embraces project-level risk frameworks.This would be a market which can account for the spectrum of quality of carbon credits and develops a system to price in the inherent risks. If we go down this route, the VCM can be a vital tool for harnessing private capital for the climate transition.

¹Lauter and Prokopczuk, 2022