This article is part of a Synergies series on Next generation trade arrangements for environment and sustainable development. Any views and opinions expressed are those of the author(s) and do not necessarily reflect those of TESS or any of its partner organizations or funders.

-----

While—contrary to many accounts—the World Trade Organization’s (WTO) Fourteenth Ministerial Conference (MC14) in March in Cameroon had its successes, one of the areas of disappointment was fossil fuel subsidy reform. The countries involved in the Fossil Fuel Subsidy Reform (FFSR) initiative at the WTO do not seem intent on moving forward to propose a binding agreement in the WTO legal framework; but instead in a Ministerial Statement simply affirmed the work plan in their MC13 statement two years ago at Abu Dhabi, a much more timid and vague agenda:

“We will advance work to identify harmful fossil fuel subsidies and to build broader recognition and support for the development of pathways to reform, reduce and eliminate these fossil fuel subsidies, considering the social impact, in particular on vulnerable groups.” 

There are already numerous studies, initiatives, and fora where the reduction or elimination of fossil fuel subsidies is advocated and/or attempted through voluntary international cooperation. These include:

G20 voluntary peer reviews of fossil fuel subsidies: Initiated to enhance transparency and accountability, members participate in voluntary peer reviews to identify and report on their fossil fuel subsidies, and to assess their subsidies and learn from others. Completed early reviews include China-US, Germany-Mexico, and Indonesia-Italy. Canada, among others, has committed to the peer review process.

Voluntary peer reviews under APEC: The Asia-Pacific Economic Cooperation (APEC) forum operates a similar peer review mechanism to identify and reform inefficient fossil fuel subsidies.

OECD/IEA reporting and inventory: The OECD Inventory of Support Measures for Fossil Fuels, in coordination with the International Energy Agency (IEA), provides ongoing monitoring and documentation of subsidies across G20 and OECD countries. This data informs G20 finance ministerials on progress towards reform.

G7-led initiatives: The G7 group of countries has set a more ambitious deadline to phase out inefficient fossil fuel subsidies by 2025 (although they are off track to meeting this target).

Despite these initiatives, G20 members have struggled to implement these commitments, with 2023 subsidies reaching $794 billion, the second-highest level ever recorded. Structural reforms are often hindered by political concerns over energy prices and the absence of a clear, universally agreed definition of "inefficient" subsidies. 

Given the lack of success of these voluntary or soft-law initiatives, the case for a binding international agreement on fossil fuel subsidies within the WTO framework seems stronger than ever. Such an agreement was concluded outside the WTO framework in 2024, the Agreement on Climate Change, Trade and Sustainability (ACCTS), signed by Costa Rica, Iceland, New Zealand, and Switzerland. 

As Grossman and Sykes note in a recent (and important) article, the economic case for international disciplines on subsidies generally is not strong, since most of the identifiable externalities of subsidies—positive and negative—are realized within the subsidizing state itself. And regarding the impacts on international competition, subsidies are difficult to separate from the many other government policies that burden or benefit firms and industries, affecting competitiveness.

There is, however, one case where Grossman and Sykes see the value of international cooperation to discipline subsides: this is where the subsidies create global negative externalities. Acting alone, individual states do not have an incentive—or not a sufficient incentive—to prevent these harmful effects beyond their borders. According to Grossman and Sykes:

“One common cross-border externality reflects the environmental impacts of expanded activity in a given sector. If production in a sector generates harmful emissions and if the adverse effects of this pollution are not confined to the area near the production facility, then the subsidy creates a negative externality for neighboring or even distant trade partners. If, on the other hand, a subsidy promotes activities that reduce pollution or otherwise protect the environment, then partners will experience positive spillovers. An ideal set of subsidy rules would encourage governments to take account of international environmental externalities, be they positive or negative.”

According to the IMF, removal of cash subsidies and tax expenditures (tax breaks) for fossil fuels is projected to reduce global emissions by 6% below baseline levels by 2035. This is a not insignificant contribution to the global public good of climate mitigation. So there is a positive international externality that benefits everyone.

In the past, the challenge of imagining a fully developed model for a binding agreement seemed daunting. However, we now have such an agreement in the form of the ACCTS. This agreement is a starting point for thinking about what might be possible at the WTO, plurilaterally as well as multilaterally. 

The ACCTS, Chapter IV and Annex IX, completely bans subsidies for coal-related fuels (HS 2701–2704, 2708, 2714) and for all stages of the oil and gas sector (HS 2709–2711), covering both production and consumption activities across the entire value chain. Commitments to limit or reduce other kinds of subsidies can be scheduled. Certain subsidies are explicitly permitted. These include, among others: measures for cutting fossil fuel emissions and transitioning to clean energy; incentivizing pollution reduction; supporting vulnerable groups; improving energy security; responding to crises; supplying energy for essential services; and enabling the energy transition. The ACCTS also permits parties to maintain subsidies to fulfil their existing international obligations of the Convention on International Civil Aviation, Vienna Convention on Diplomatic Relations, and Revised Rhine Navigation Act.

A highly innovative and significant feature of ACCTS is the introduction of the Standardized Carbon Rate Measurement (SCRM). A party can commit to a carbon price that reflects at a minimum its commitments to emissions reductions under the UN Paris Agreement. As long as fossil fuel subsidies do not lower this carbon price below the committed level they are permitted under the ACCTS. As Tipping et al. explain:

“The SCRM is the net total price applying to carbon dioxide (CO2) emissions from a given fossil fuel considering subsidies, carbon pricing, and some energy taxes. It mirrors the OECD’s Net Effective Carbon Rate and could include components like excise tax, carbon tax, and Emissions Trading Scheme price, minus any exemptions. For instance, a country might commit to an average SCRM of USD 50 per tonne of CO2 for gasoline. This is important because a lot of countries have carbon pricing (e.g. via the European Union Emissions Trading Scheme) but also provide exemptions for some sectors that can look like large subsidies in nominal terms. It is the net effect of policies that determine consumer prices (therefore demand), investment decisions, and government revenues, and this net effect is captured under the SCRM. For the ACCTS countries that choose to use the SCRM mechanism—currently Iceland and Switzerland—a subsidy only needs to be eliminated if it causes the SCRM to fall below the committed value.”

If we go back to the negative effects of fossil fuel subsidies identified above, the costs of fossil fuel subsidies in terms of local pollution (public health costs), the burden on public finances, and the distortion in the market for clean energy technologies are all felt locally (although the last effect could create a negative spillover for other states exporting clean energy technologies). The significant global externality is a rise in greenhouse gas emissions that accelerates climate change.

Despite the possibility that an agreement can achieve a global public good, the transaction costs of arriving at such an agreement may be prohibitively high, especially where states disagree on how much each must carry in terms of transitional costs (including political costs) to achieve a good that all will share in. The first international agreement to discipline subsidies to achieve a global good (biologically sustainable fish stocks in the ocean commons), the WTO Agreement on Fisheries Subsidies, took decades to negotiate and is only now in force for members that have ratified. It only covers a limited percentage of total subsidies and some major subsidizers have not joined. The different political economies of different states surrounding the fishing industry created a daunting challenge to reaching an agreement, even if all can benefit from the global good at which it aims.

The Agreement on Fisheries Subsidies and the ACCTS both show that negotiating disciplines on subsidies that create global negative externalities is not impossible. And economic analysis (at least as presented by Grossman and Sykes) explains why it is desirable. Further, there are examples where states have altered their domestic political economies (even outside the exercise of an international agreement) to capture the large domestic benefits from eliminating fossil fuel subsidies: Uruguay, India (with respect to fuel subsidies), and the ACCTS members. These states exploited the availability of affordable energy alternatives, whether hydropower, electric vehicles, or solar. Other states may need to think in terms of different alternatives that may require a longer transition period. Moreover, there are a number of cases where the goal of ensuring affordable energy supply to the poor, or the economically precarious, has been achieved through switching to more targeted programmes of economic assistance.

Until the ACCTS, the nature of the challenge of an international agreement to remove the global negative externalities created by fossil fuel subsidies was, arguably, never properly conceptualized.

In the ACCTS one finds a clue as to what is required for this conceptualization, namely in the SCRM. The SCRM—as well as to some extent the distinction between implicit and explicit subsidies—adumbrate the correct way of thinking about the problem of fossil fuels subsidies: they result in the price of carbon being set too low for purposes of achieving the global public good of climate change mitigation. 

Once we recognize the fundamental issue with fossil fuel subsidies is they lower the cost of carbon and that is how they contribute to raising emissions—thereby increasing a global negative externality—the definitional debates about what policy instrument qualifies or not as a fossil fuel subsidy, or which are more or less harmful, recede into insignificance. These debates among experts and officials have significantly delayed and frustrated concerted action towards a negotiated agreement. As long as a subsidy, all things being equal, contributes to lowering the cost of carbon, it produces a negative global externality. This means that in principle eliminating it through an international agreement is justified.

If all subsidies that contribute to lowering the price of carbon below what is required to meet a WTO member’s Paris commitments were in principle to be included in an international agreement to eliminate fossil fuel subsidies (as is logical), what would we expect the extent or shape of commitments to be in practice?

This will be determined by each state’s domestic political economy. Different states will be willing to bargain for elimination or reduction of different types of subsidy programmes. At the same time, we recognize there could be fairness or equity concerns if the shape of commitments is too asymmetrical or appears to be. Here, in determining whether the overall balance suggests that everyone is doing their part, and no one is bearing an unequitable burden to achieve the global common good, the negotiators would simply have in mind the way in which states’ commitments overall to that good have been defined in the Paris Agreement, including the concept of differential responsibilities.

However, in order to justify the transaction costs of arriving at an agreement, the collective ambition must be great enough to make a difference in terms of achieving the global public good in question (albeit some states may also seek the agreement with a view to reducing “unfair” or distorted competition that negatively impacts domestic producers). 

One approach would be to follow the example of the WTO Agreement on Agriculture, determining an aggregate measure of support (AMS) in the case of fossil fuel subsidies—this would be the total basket of subsidies that each member wishes to submit to international commitments for reduction or elimination—and an expected pace at which this AMS is reduced (e.g. 10% per year). 

Should all members be expected to reduce by the same annual percentage? This goes to a fundamental question: how have states changed their political economy to eliminate fossil fuel subsidies? While by now there is considerable literature about that, it is rarely brought into discussions on the design of an international agreement disciplining fossil fuel subsidies, although all too often those discussions end with an idealistic call for political will or a lament about the lack of it.

There is a basic, if fairly obvious truth: states that have successfully transitioned out of fossil fuel subsidies have done so—consistent with political mandates including for consumer energy security—because they have been able to shift consumption significantly towards green energy; this was dramatically the case for Uruguay for instance. If consumers are not able to shift to affordable alternative energy sources, and must continue to rely on fossil fuels, then eliminating fossil fuel subsidies will largely have a domestic redistributive impact, possibly regressive: people will still use fossil fuels but pay more. Perhaps there will be some emissions reductions, but they are not apt to be large since a good deal of such consumption is necessary for personal or business and industrial uses. 

It is therefore a recipe for failure not to link expectations for reduction in fossil fuel subsidies to transitioning to a green economy. Here, there continue to be technological and logistical limits to the extent and pace at which such transitioning to low-cost green energy is possible in particular states. 

The European Union began as the modest-sounding Steel and Coal Community. It may be advisable to begin with an agreement that addresses fossil fuel subsidies in a particular context, for example gasoline subsidies for ground transport. The areas chosen might reflect those where there is already domestic experience with how to transition from fossil fuels to alternative energy sources (e.g. EVs, electric and hybrid buses, rail). 

In all known and relevant cases, government support for this transition has been essential, usually through the subsidization or pricing of renewables. The existing WTO rules on subsidies and countervailing duties do not, as Grossman and Sykes suggest, map on very well to the main economic rationale of disciplining subsidies in order to eliminate a global negative externality or achieve a global public good (if one wants to articulate the goal in positive terms). On the other hand, they threaten the political viability of support to renewable energy that is essential for the transition. 

This is because the subsidies rules, as well as Article III of the GATT (National Treatment) and the Agreement on Trade-Related Investment Measures all prohibit local content requirements, which have been crucial to creating effective coalitions to support government expenditures for the transition to renewables. Even if there are legitimate energy security interests in developing a local supply chain for renewables, these local content requirements are seen within the WTO as inefficient and protectionist. Thus, India, which made climate and energy security arguments under the GATT exceptions for its local content requirements, lost at both the panel and Appellate Body levels in the India-Solar dispute.

In brief, there needs to be some kind of WTO waiver or peace clause on local content requirements for parties that are implementing a fossil fuel agreement. 

Further, the close link between transitioning away from fossil fuel subsidies and towards alternative energy sources suggests, as noted, that required reduction levels and pace for fossil fuel subsidies might be different in the agreement for different WTO members, depending on their capacity—fiscal and technical—to manage such a transition in a particular time frame. This would involve grouping WTO members and assessing capacity and possible needs for international assistance, financial or otherwise. Here the WTO Trade Facilitation Agreement and the plurilateral Investment Facilitation for Development Agreement provide some examples for such an architecture. 

----------

Robert Howse is Lloyd C. Nelson Professor of International Law, NYU School of Law.

Petrus van Bork is an independent consultant.

Arunima Agarwal, LLM 2026, NYU School of Law, provided assistance.

-----

Synergies is an online platform featuring expert commentary and opinions curated by TESS. We foster dialogue and incubate ideas on how to shape a global trading system that effectively addresses global environmental crises and advances sustainable development. Synergies draws on perspectives from leading experts and practitioners across policy communities from around the world. We cultivate solutions-oriented policy analysis for a sustainable future.

The Executive Editor is Fabrice Lehmann.

Disclaimer

Any views and opinions expressed in Synergies are those of the author(s) and do not necessarily reflect those of TESS or any of its partner organizations or funders.

License

All of the content on Synergies is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) license. This means you are welcome to adapt, copy, and share it on your platforms with attribution to the source and author(s), but not for commercial purposes. You must also share it under the same CC BY-NC-SA 4.0 license.

If you would like to reuse any material published here or if you have any other question related to Synergies, send an email to Fabrice Lehmann.