Pande in Science Magazine: Offsets, carbon markets, and climate and economic justice

This article first appeared in Science on September 12, 2024

Today, 682 million people, or 8.5% of the world’s population, live in extreme poverty, which the World Bank defines as subsisting on less than $2.15 per day. The majority live in low-income and lower-middle-income countries. The traditional pathway for economic development available to these countries has historically required a lot of energy; the world’s richest countries frequently have the highest per capita historical greenhouse gas emissions and are also primarily responsible for anthropogenic climate breakdown. Lower-income countries today confront very reduced emissions budgets if catastrophic warming is to be prevented, and they are more vulnerable to climate change–related damage.

As Rachel Glennerster and Seema Jayachandran have pointed out, many of the most cost-effective carbon-reduction investments are in low- and lower-middle-income countries. Some of these investments, such as those in renewable energy, more energy-efficient buildings, and electrifying transportation, promote economic development while lowering emissions. Others, however, such as preventing deforestation, may have long-term costs for development and poverty reduction.

Carbon emissions reduction strategies vary hugely in terms of cost per ton. Given a limited income stream and an eventual global net-zero target, funding the least costly mitigation opportunities first, and the most costly last, is usually the most effective way to reduce long-term damage. This is because damage increases with CO2 concentration in the atmosphere, and enacting the least costly mitigations first reduces the overall rate of emissions faster, resulting in a lower peak concentration when the emissions rate hits zero.

In this context, how do international carbon markets, particularly voluntary carbon markets, interact with economic development, political conceptions of climate justice, and the global objective of eradicating extreme poverty?

A well-functioning international carbon market can substantially benefit low- and lower-middle-income countries. It can provide investment for efficient, low-carbon growth paths that these countries cannot otherwise afford, provide subsistence to low-income communities to prevent them from engaging in environmentally and ecologically destructive practices that would otherwise be their sole income source, and preserve ecosystems with substantial future value. And, of course, it can assist these countries to avoid the economic damage of unmitigated anthropogenic climate breakdown. However, such a market must meet certain conditions.

First, an international voluntary carbon market based on carbon offsetting is almost certain to fail to achieve climate justice. By “offsetting,” I mean a process where a Global North company aiming to construct a narrative that it is not contributing to climate breakdown—but facing high costs to eliminate its own emissions—chooses to achieve “carbon neutrality” or a “net-zero” target by buying carbon credits from a Global South developer working with a Global South community to sell a carbon abatement project through the unregulated voluntary market. The intention is to match its own emissions one for one. By buying credits, it is, in theory, preventing emissions elsewhere that are equal to its own emissions. This process typically entails buying the cheapest available credits and thus spending far less to achieve “net zero” with the CO2 reduction bought through the credits than the company would have had to spend to lower its own emissions.

The problem is that to prevent climate breakdown, we need to achieve global net-zero emissions. Even if the credits genuinely represent the reductions claimed by the developers, the Global North company, in the absence of any emissions targets other than its own “net-zero” target, is essentially absolving itself of responsibility for removing all of its own expensive-to-reduce emissions by purchasing offsets. This means that, although the company may now consider itself carbon-neutral, viewed from the global perspective, only a portion of the Global North company and Global South community’s combined cross-country emissions have been removed, and no one is willing to take responsibility for the remaining, more expensive, emissions that the Global North company continues to emit and profit from. To achieve global net zero, these remaining emissions must be removed.

To be truly good climate citizens, Global North companies that benefit from national infrastructure and wealth created through decades of intensive energy use and greenhouse gas emissions must work with Global South project developers to finance reduction pathways that lead to eventual global net zero. This requires a commitment to both finance less costly reductions in Global South countries and eventually eliminate their own, more costly to reduce, emissions.

Such a commitment is best structured as participation in a novel compliance, or cap-and-trade, market, as I described in a previous Science piece. In a cap-and-trade system, a regulator moves a group of emitters toward a specified emissions target by gradually lowering an emissions cap and reducing the number of permits (or credits) allocated to each emitter to stay at or below the cap. Emitters that reduce their emissions faster can trade permits with those who do not reduce them fast enough. Those with insufficient permits to cover their emissions incur substantial regulatory penalties. Market rules guarantee that credits accurately represent the claimed emissions reductions. This structure ensures that emissions targets, such as a genuine net-zero target across emitters, are achieved. Outside of such a structure, which, inter alia, also guarantees the quality of credits, those who decry offsetting and carbon markets as a greenwashing tool for Global North companies are likely correct.

Second, additionality—the requirement that a credit’s stated emissions reduction would not have occurred without the credit payment—remains essential. Some ecosystem-based carbon credit proponents argue that additionality becomes less important in a low-income country with considerable extreme poverty and where any income is valuable; even if a project fails to reduce emissions, the revenue from selling credits can still reduce poverty and support economic growth.

However, this perspective ignores the damage to the market and the planet caused by credits that do not represent the CO2 reductions that they claim. Because buyers wish to be seen as “good citizens” on climate, public revelations that some credits do not offer credible CO2 reductions cause confidence and prices to drop; in turn, this constricts the market’s potential to enable real carbon reductions, which will require buyers to pay the higher prices that genuinely additional credits will cost. Further, when companies use carbon credit purchases to fulfill climate commitments while continuing to produce substantial emissions, credits that do not provide the promised CO2 reductions can increase rather than decrease emissions, exacerbating the economic damage to low-income countries caused by climate breakdown. Additionality, therefore, is essential regardless of the context.

Third, even in a compliance market, carbon trading predicated on ecosystem-based carbon credits may impose substantial costs on lower-income countries. This carbon trading requires discontinuing less-profitable economic activity in the Global South in order to continue more profitable economic activity in the Global North, while adhering to a specific emissions reduction trajectory. While a Global South landowner would only accept a credit price that makes up for the direct income lost by ceasing her economic activity, supplying the credit may impose costs on her community for which the credit price alone is inadequate compensation.

For example, if deforestation enables agriculture, a landowner may be willing to halt deforestation and agriculturalization if her profits from this business can be replaced like for like by revenue from carbon credit sales. However, her business may have provided income to farm workers, which could have been spent on products and services, and taxes to the local government, which could have funded infrastructure and education. So the price the landowner is willing to accept may be lower than the cost faced by the community as a whole.

One solution could be to levy a local tax on carbon credit sales, so that the income realized by halting economic activity extends beyond the landowner who stops the work and includes the affected community, which loses out on the indirect benefits of the work. The effective price of credits to buyers would then rise to cover the real costs of providing them to the entire community.

Fourth, it is important to recognize that ecosystem-based carbon credits that maximize additionality will not necessarily benefit the poorest community members or those who currently act in ways that preserve the ecosystem. To achieve additionality, payments must alter landowners’ behavior—that is, they must flow to those who were at risk of destroying the ecosystems they owned. Paying “responsible” landowners—those who have not and do not intend to degrade the ecosystem—does not change behavior and yield additionality. Those with the means to exploit and degrade the ecosystem are sometimes more affluent—and may have increased their relative affluence through these destructive activities—so targeting additionality may mean directing payments to wealthier landowners. A carbon credit–funded income stream that targets these most affluent and destructive community members, but which is unavailable to those who have thus far conserved forests, may appear unjust and become politically difficult to administer. These political barriers may especially arise if local governments, who are under democratic pressure to ensure fairness, are enabling the development of the conservation project.

A potential solution to the political challenge of implementing apparently “unfair” carbon credit targeting could involve recognizing that proving additionality on average across multiple payment recipients is sufficient to justify the issuance of carbon credits. These credits would be determined by the overall average emissions averted, even if some recipients are not engaged in environmentally destructive activities. An identical “payment not to damage” can be given to a heterogeneous group of recipients in deforestation-vulnerable areas; some will accept it and cease damaging economic activities, while others will accept it as a “reward” for not doing so at all. In a payments-for-ecosystems-services experiment in Uganda, Seema Jayachandran shows that the poorest, whose economic activities cause minimal damage to the forest, in fact benefit most from such an initiative; this is because richer community members reduce deforestation and lose income in response to the payment, whereas poorer community members lose no income from ceasing activities and so receive payments without loss. Payments yield additionality on average, even if some recipients do not change their behavior. Of course, if a payment of this kind is not tightly focused on recipients who will provide additionality, credits must be sold at a higher price. However, if a scheme of payments appears unjust, it may be politically impossible to bring these credits to market; higher pricing then becomes essential if the credits are to be sold at all.

Finally, in a previous opinion piece, I made the case for a functional market that establishes a single price for carbon credits at any given time. This price would reflect the value to humanity of a specific reduction of CO2, regardless of where the reduction occurs. A thoughtful letter responding to that opinion piece suggested that the pricing of credits should also reflect the social benefits that accompany generation of the credits. For example, the letter writers argued, while two forest conservation projects—one excluding local communities and another promoting sustainable community management—might be equally effective in conservation, the latter has greater social value and thus should command a higher price. A cap-and-trade system, in their view, could disadvantage more-ambitious projects that also enhance biodiversity and local livelihoods.

In a cap-and-trade market, the regulator sets the cap, and the market then sets the price. The cheapest credits—representing the emissions that are least costly to remove—are bought first, and, as the emissions cap is lowered, the number of credits and, correspondingly, the supply of cheaper emission reduction opportunities are reduced, and the price goes up. Emitters can choose to sell credits at the price where it becomes financially more rewarding to reduce emissions and sell the credit than to continue emitting.

Because a cap-and-trade market ensures demand, creates scarcity, and assures quality, credits in well-regulated cap-and-trade markets command much higher prices than those in the existing unregulated voluntary market, and the prices for these credits increase over time. So, overall, there is more income available to communities and developers selling credits in these markets than to those selling in the unregulated market. In cases like those explored above, where there may be extra costs involved in producing credits that are considered fair and politically viable to implement—beyond those of compensating individuals for direct losses from ceasing ecosystem use—credit prices should include these costs.

The case the letter writers cite, where developers may choose to implement carbon-abatement schemes in ways that either bring more or less benefit to local communities, then becomes one example of a more generalized justice issue: how income from credits is split between developer and community, and how the community and developer decide on the final price that allows for conservation to happen and for credits to be sold. A commercially minded developer may want to share less of the credit-sale income and sell faster, for a lower price. And it is likely that a developer is both better informed and better funded than the community it is negotiating with, so it holds the upper hand.

This suggests, then, a final condition for constructing just international voluntary carbon markets. I have previously set out how cap-and-trade markets need to have mutually agreed-upon rules that ensure quality of credits in terms of guaranteed carbon reductions. The letter writers’ emphasis on the importance of paying attention to the social co-benefits of developing credits is justified; it should be possible to trade credits knowing that they were developed in a way that was ethical and economically just, and with a fair division of returns between community and developer. So voluntary carbon credit compliance markets should also have ethical rules and standards that developers are required to follow in order to participate in the market.