Companies often have a two-step approach to deciding what carbon credits to buy. First, they assess whether a credit has the “right quality”, and then they decide what price they are willing to pay. The question to ask is not simply, “is this good quality” or “is this the right price”. Rather, we have to ask: “Is this the right quality for the right price?”
In other words, where do you get the most climate impact for your money?
Our CEO Mikkel Larsen shares his thoughts below on how we can approach this question.
The Price-Quality Nexus today
One key challenge when pricing for “quality” is that there is still no strong correlation between “price” and “quality” today. A recent study by carbon ratings agency BeZero suggests that “on average, the price difference between credits with a BeZero rating of ‘A’ or higher compared to those with a BeZero rating of ‘BBB’ or lower has been 79%, with a standard error of +/- 8%”. While this may seem like a strong correlation, the analysis shows that the correlation changes over time and is not linear over the rating bands. The correlation is nevertheless markedly improved since the introduction of their publicly available headline ratings in 2022. To add to this, while carbon data provider Sylvera’s carbon credit pricing tool, which compares prices and quality, also suggests that the correlation is not always strong, it is a step in the right direction to getting clarity alongside BeZero’s whitepaper on risk-adjusted returns.
These results are consistent with studies published by Ecosystem Marketplace, which found that credits with specific Sustainable Development Goals (SDGs) attract a premium of up to 86%, further demonstrating that we do not yet have a uniform view on what quality is.
Such inconsistencies in pricing strategies go to show that companies are either basing their buying decision on imperfect information, and/or that there is simply imperfect logic in decision-making.
Prices do not necessarily reflect positive climate impact
Prices today often reflect both the demand-supply dynamics and the cost to produce the credits.
When pricing based on the former, the demand curve is often heavily impacted by regulation and the claims that can be made. Removals are priced higher, in part because they can be used to claim Net Zero under the Science Based Target initiative (SBTi) or compliance with CORSIA; but also because of their acceptance under regulations such as the EU Carbon Removal Certification Framework (CRCF) and the Singapore Carbon Tax scheme.
In general, prices also reflect the cost to produce the credits, especially for those that do not fit under regulatory frameworks. It is more expensive to generate credits from an afforestation project than a REDD+ avoidance project – the price differential between these projects can be 10 times. Similarly, sequestering a tonne of carbon using Direct Air Capture (DAC) is significantly more expensive than afforestation. The former can be priced anywhere from US$700-1,200, while the latter at US$5-20 depending on quality – a difference of nearly 100 times.
The challenge is that these price differentials do not necessarily reflect the full impact of a project’s positive impact on climate change mitigation.
Binary rules are helpful but have their limitations
We often evaluate quality as binary – it is either good enough or not. For example, a methodology is in the ICVCM’s Core Carbon Principles, or it is not. In a nascent and complex market with thousands of projects to choose from, it is understandable that users of carbon credits would gravitate towards binary rules that set out what “good enough quality” is. But this binary approach by necessity also masks the complexity of evaluating a project. Rarely is it as simple as saying that there is a 0% chance of a full (100%) permanent climate change impact.
Rating agencies help to supplement binary rules by introducing “probabilities”. In considering and scoring elements such as leakage and permanence with a risk rating of high, medium or low, they are essentially attempting to put a measure on the probability of these impacts to reward projects that have higher climate impacts with higher prices.
Nevertheless, thinking in terms of probabilities rarely carries over into the evaluation of price. If a project has a “good enough” assessment or rating – and meets company specific requirements – the cheapest credit type is often purchased.
The math of Probability Weighted impact
A more nuanced decision would look at the “probability weighted” positive impact of different credit types and projects, compare prices, and then determine where a company will get the most positive impact for its money. This may sound complex, but let me try and illustrate here with some examples.
- Removal versus avoidance. Conventional and current “wisdom” seem to be that we will only reach Net Zero by focusing on removal credits with long storage duration (ideally 1,000 years, but at least 100 years). This is perhaps true – but only in a simplified world where avoiding further emissions without the incentives that carbon credits provide is possible.
Using the example of reforestation versus avoiding unplanned deforestation (e.g. illegal logging), it is evident that both are subject to the risk of not being permanent and could lead to leakage. If we accept that we need to incentivise both avoidance (stop deforestation) and removal (start reforestation), then there is evidence that avoiding the damage is far superior economically (and better for biodiversity). Restored forests in moist tropical regions sequester up to 11 tonnes per hectare per year (above ground). By contrast, protecting standing rainforests immediately preserves 300-400 tonnes per hectare per year, a 30 times difference. Considering that the price for nature-based removal credits is often seven times that of avoidance credits like REDD+, you get a rough efficiency factor of over 200.
The question is whether reforestation is really 200 times more impactful than avoidance?
- Probability Weighted Impact. Next, let’s consider as an example the probability of a project delivering less than a fully permanent solution. If we take it that DAC has a 100% certainty of emissions being permanently stored (which it does not), and that reforestation does not offer the same level of guarantee – and map that against the price difference of over 200 times (see above). The reforestation project has some mitigants to ensure longer storage duration or permanence, such as legal rights, community benefits and risk buffers – not perfect but directionally valuable. For instance, monitoring periods under Verra are now 40 years.
- With these in mind, I wonder if our climate would be better off if a company bought 200 credits from a reforestation project or one credit from a DAC project? If we considered this on a portfolio basis, would only 1/200 = 0.5% of the trees planted be standing in 100 years? This takes us into the topic of discounting and over-compensation”. A topic for another day.
Final thoughts
To be clear – the price-quality nexus is complex, and I am not suggesting that all buyers can afford or should even undertake scientific studies with precision to determine the relative value of carbon credits. Perhaps one day, but we are not there yet.
What I do suggest is that the next time you buy carbon credits, consider whether you get more value for money by buying a project that has a lower perceived quality and, if in doubt… buy a portfolio of different credits.
