ESIL Reflection: The G7’s Pledge to End Fossil Fuel Subsidies by 2025: Mere Rhetoric or a Sign of Post-Paris Momentum?

PDF Version                                                   Vol 5, Issue 8
Editorial board: Anne van Aaken (editor-in-chief), Jutta Brunnée, Başak Çali, Jan Klabbers  

Henok Birhanu Asmelash
Bocconi University and Max Planck Institute Luxembourg for Procedural Law

The G7´s Pledge to End Fossil Fuel Subsidies by 2025:
Mere Rhetoric or a Sign of Post-Paris Momentum?



At their Ise-Shima Summiton 26‒27 May 2016, the G7 leaders agreed to end fossil fuel subsidies (FFSs) and called on other countries to do the same.[1] This agreement is the latest in a series of non-binding intergovernmental agreements to phase out FFSs since the 2009 G20 Summit. The Ise-Shima Agreement, however, represents the first attempt to set a specific deadline for phasing out FFSs at the intergovernmental level. In a statement released shortly after its announcement, Sierra Club Executive Director Michael Brune hailed the agreement as “an important step forward for our climate and the millions of people around the world suffering from the effects of pollution from dirty fuels”.[2]  Shelagh Whitely of the Overseas Development Institute (ODI) called it “a new chapter in the climate history books”.[3] This Reflection considers whether this warm reception is justified and how the substantive content of the Ise-Shima Agreement differs from past agreements to phase out FFSs.

The last few years have witnessed growing recognition that FFSs are not only economically inefficient but also environmentally harmful. This has led to intergovernmental agreements to phase out FFSs. Despite these agreements, however, FFSs remain prevalent worldwide. One of the major shortcomings of previous intergovernmental agreements has been their failure to set a target date for the phasing out FFSs. The Ise-Shima Agreement has generated high levels of interest because of the expectation that the process of phasing out FFSs will be expedited as a result of the Agreement’s deadline. This Reflection argues that the Agreement is, however, unlikely to live up to this expectation. Each G7 country is allowed to define for itself what constitutes a FFS. This means that G7 countries can define FFSs as narrowly as they wish and thereby exclude the various forms of public support they provide for fossil fuel production and/or consumption. A closer look at the Agreement also reveals that the deadline for phasing out FFSs is ambiguous at best and counterproductive at worst.

Before proceeding to examine the substance of the Ise-Shima Agreement in depth, this Reflection will first briefly review the environmental case for eliminating FFSs and previous intergovernmental agreements to phase out FFSs.


1.    The Case for Eliminating FFSs

The Ise-Shima Agreement came less than six months after the Paris Agreement, in which 196 countries undertook to limit the increase in the global average temperature to well below 2 °C and to pursue efforts to limit the temperature increase to 1.5 °C above pre-industrial levels.[4] This ambitious target is to be achieved through nationally determined contributions to reducing greenhouse gas emissions.[5] The energy sector, accounting for more than two-thirds of global greenhouse gas emissions, is critical for any country’s emission reduction strategy.[6] Fossil fuels, which consist primarily of coal, oil and natural gas, currently comprise over 80% of the global energy supply. To meet the 2 °C climate target, the vast majority of proven fossil fuel reserves need to remain in the ground.[7] This can only be achieved through a massive improvement in energy efficiency and a rapid transition of the global energy system from one relying heavily on carbon-intensive fossil fuels to one depending mainly on renewable and low-carbon energy sources. The much-sought transition towards renewable energy is already underway but not at a sufficient pace to meet the 2°C target. The history of previous energy transitions –from wood to coal and from coal to oil and natural gas ‒ suggests that energy transition is a lengthy process that will take several decades.

Government policies can play a crucial role in influencing both the direction and the speed of energy transition. Accordingly, governments worldwide have introduced a wide range of energy transition measures, including renewable energy subsidies.[8] However, in the absence of carbon pricing, the measures that have been implemented so far are unlikely to bring about the energy transition on their own. Putting a price on carbon (in the form of carbon taxes or cap and trading schemes) is widely regarded as the most efficient way to expedite the transition towards a sustainable energy future. Whilst there have been growing efforts in this direction, only few jurisdictions have so far introduced carbon taxes or emission trading systems.[9] Moreover, as pointed out by Tim Grosser, “it is completely incoherent for the world to be tentatively coordinating actions to put a price on carbon, while simultaneously massively subsidizing the consumption of carbon”.[10]

Governments worldwide subsidize fossil fuel production and consumption with $500 to $5300 billion every year, depending on how subsidies are defined and measured.[11] These subsidies are both economically inefficient and environmentally harmful. Besides unnecessarily draining government budgets, they encourage wasteful energy consumption, crowd out public spending on other priorities, such as health and education, and discourage investment in energy efficiency and renewable energy technologies. They are often justified under the guise of providing poor households with access to modern energy services, but studies have shown that FFSs rather benefit high-income households.[12] The latest IMF study on FFSs estimated that eliminating FFSs could reduce global CO2 emissions by more than 20%, and raise global economic welfare by $1.8 trillion.[13] It is for these reasons that FFS reform is considered as the “low-hanging fruit” of climate change mitigation policies.

FFS reform was discussed as one of the policies and measures that could be employed to tackle climate change during the Paris Agreement negotiations. Two provisions of the draft negotiating text made reference to FFSs: the first urging “parties to reduce international support for high-carbon investments, including international fossil fuel subsidies”, and the second listing “the phasing down of high-carbon investments and fossil fuel subsidies” as a climate finance option.[14] Neither of these provisions, however, made it into the final text of the Paris Agreement. What remained in the final text (adopted on 12 December 2015) is Article 2(c), which commits parties to:  “[make] finance flows consistent with a pathway towards low greenhouse gas emissions and climate resilient development”. This is far from a clear requirement to end FFSs, but there is no doubt that FFSs go against the objective of Article 2(c) insofar as they divert financial flows away from renewables, and thereby delay the transition towards low carbon development.[15] From this perspective, eliminating FFSs is an important step towards the implementation of the Paris Agreement. The central question is whether the Ise-Shima Agreement iscapable of inducingsuch elimination.


2.    Intergovernmental Agreements To Phase Out FFSs

FFSs have been the subject of discussion in various intergovernmental forums ranging from the IMF and the WTO to the UNFCCC, confirming the crosscutting nature of the issue and underscoring the fragmentation of global environmental and energy governance. From an environmental perspective, the issue of FFS reform first appeared on the international agenda during the Kyoto Protocol negotiations in the 1990s. The Consolidated Negotiating Text (CNT) by the Chairman included an explicit reference to FFSs. The corresponding provision to Article 2.1(a)(v) of the Kyoto Protocol in the CNT listed the “progressive phasing out of market imperfections and fiscal incentives that run counter to the objective of the Convention, including, inter alia, subsidies on all fossil fuels” as one of the priority policies and measures to address climate change.[16] The reference to FFSs was, however, replaced by “subsidies in all greenhouse gas emitting sectors” in the final text of the Protocol.[17] Their implicit inclusion in the Kyoto Protocol’s list of emission reduction policies and measures hardly attracted any attention, perhaps because of the broad discretion given to Annex I countries in terms of their choice of policy measure to meet their emission reduction commitments under the Protocol.  

Serious international attention was given to FFSs only in the aftermath of the 2009 G20 agreement to "phase out and rationalize" inefficient FFSs over the medium term.[18] This started a wave of non-binding intergovernmental agreements to phase out FFSs.[19]

Nevertheless, FFSs remain prevalent around the world. In fact, global fossil fuel consumption subsidies increased from $409 billion in 2010 to $493 billion in 2014, according to the IEA.[20] The intergovernmental agreements’ failure to induce the phasing out of FFSs can be attributed to a number of issues, including: the lack of a precise definition of FFSs; the lack of mechanisms to improve transparency; the lack of specific timelines for phasing out FFSs; the lack of monitoring and surveillance mechanisms; and the lack of enforceable commitments.[21] As discussed below, the potential role of the Ise-Shima Agreement in eliminating FFSs depends on whether and to what extent it will address these issues.


3.    The Effective Content of the Ise-Shima Agreement

The Ise-Shima Agreement is the first intergovernmental response to the widespread calls for eliminating FFSs in the post-Paris era. It is partly for this reason that the Agreement has been viewed as a positive sign of post-Paris momentum in some quarters.[22] However, we need to examine the substance of the agreement to disentangle the rhetoric from the reality.

The logical place to start assessing the substance of the Ise-Shima Agreement is the definition of FFSs. The definition determines its scope and coverage. The broader the definition is, the wider the scope of application of the Agreement. The key part of the Ise-Shima Agreement that deals with FFSs appears in the last paragraph of the Climate Change subsection of the section on Climate Change, Energy and Environment. This paragraph provides no definition whatsoever of “fossil fuel subsidies”. This would not be a concern if there was a single internationally agreed definition of “subsidies”. However, despite its frequent use, the term “subsidy” has no generally accepted definition. Existing definitions range from as narrow as a direct budgetary payment by a government to a producer or consumer to as broad as any government interventions that affect prices or costs.[23] Choosing one definition over another has far-reaching implications. For example, based on its relatively narrow definition, the IEA estimated the value of FFSs worldwide to be $548 billion in 2013, whereas for the same year the IMF estimated FFSs to be $1.9 trillion, using a broader definition that factors in negative externalities associated with fossil fuel production and consumption (e.g., climate change, congestion and air pollution).[24] The significant difference between “subsidy” definitions means that the scope of application of the Ise-Shima Agreement will vary depending on the particular definition adopted by each G7 country.

Since combating climate change is the main rationale behind the agreement to end FFSs, one may argue that (in the absence of a precise definition) the term “subsidies” will (or should) be given a broad meaning to capture the different forms of government support measures for the production and consumption of fossil fuels. However, two considerations suggest that this will not be the case. First, by adding the term “inefficient”, G7 leaders have already indicated their intention to limit the scope of their pledge only to “inefficient FFSs”. The problem is that it is already hard enough to define FFSs, let alone to identify inefficient ones (especially when no criteria or guidelines are prescribed). Moreover, it is worth noting that this is not the first time that the term “inefficient FFSs” has appeared in intergovernmental agreements to phase out FFSs: it was used in both the 2009 G20 and APEC agreements but with no definition. It is beyond the scope of this Reflection to discuss what an inefficient or efficient FFS is (whether such questions can be answered ex ante is questionable in any case). However, it should be noted that, unless one argues that “efficient FFSs” are “environmentally beneficial”, distinguishing FFSs based on “efficiency” makes little sense from an environmental perspective. Moreover, the ongoing process of implementing the 2009 G20 agreement, which as mentioned left the term “inefficient FFSs” undefined, has already shown that (given the choice) countries tend to define FFSs as narrowly as possible. For instance, the UK claims to have no FFSs, defining them as “government action that lowers the pre-tax price to consumers to below international market levels”.[25] This definition is “extremely narrow”.[26] For instance, it excludes the various forms of support that the UK Government provides to fossil fuel production, which are estimated at $9 billion per year on average.[27] The inference from the foregoing is that, by leaving the definition of “inefficient FFSs” open, the Ise-Shima Agreement has provided a way for G7 countries to define FFSs as narrowly as they wish and thereby claim not to have any.

A second and perhaps even more intriguing issue concerns the deadline to end FFSs. First, the paragraph of the Ise-Shima Agreement addressing FFSs is vaguely formulated. The relevant part of the paragraph reads: “We remain committed to the elimination of inefficient fossil fuel subsidies and encourage all countries to do so by 2025”. As pedantic as it may seem (especially in light of the “G7 pledge to end FFSs by 2025” media headlines), it is important to note that the literal reading of this paragraph would cast doubt on whether the 2025 deadline is set for the G7 countries or simply constitutes part of the call for non-G7 countries to eliminate inefficient FFSs. It is of course hypocritical for G7 countries to urge non-G7 countries to end FFSs by 2025 if they are will not act similarly. An optimistic reading would even suggest that G7 countries are simply urging non-G7 countries to end FFSs by 2025, while they intend to eliminate theirs even earlier. Such a reading would actually reflect the difference between G7 and non-G7 countries in terms of the size and types of FFSs as well as the technical and institutional capacities to bring about the end of FFSs. Unless we apply a broad subsidy definition that captures “uninternalized externalities” from fossil fuel combustion and consumption, the size of FFSs in G7 countries will be much smaller than that of non-G7 countries.

Most G7 FFSs also come in the form of production subsidies, which are relatively easier to scrap than consumption subsidies (due to public resistance). G7 countries also have sufficient institutional and technical capacity to carry out an expedited FFS reform. The combination of these factors therefore suggests that G7 countries are better placed to end FFSs ahead of their non-G7 counterparts. When viewed from this perspective, setting 2025 as a deadline for ending G7 FFSs might even be counterproductive. If it takes almost ten years for G7 countries to end FFSs (which some of them claim not to have anyway), how long will it take for non-G7 countries, particularly those with a massive amount of fossil fuel consumption subsidies? This question is of vital importance given that some developing countries have legitimate justifications (e.g. poverty alleviation, income redistribution) for subsidizing the consumption and/or production of fossil fuels. Some of these countries may also lack the necessary technical and institutional capacity to undertake subsidy reforms.



Despite the widespread media attention it has garnered, the Ise-Shima Agreement suffers from the same issues as previous intergovernmental agreements to phase out FFSs, including the lack of a precise definition of FFSs. The most notable feature of the Agreement is that for the first time it sets a deadline to end FFSs. However, the deadline itself raises more questions than it answers. Aside from its vague drafting, the Ise-Shima Agreement does not justify the selection of the year 2025 as opposed to, say, the year 2020. It also lacks detail regarding how this deadline will be met and compliance will be monitored. Setting a deadline is doubtless of paramount importance in the global fight against FFSs. An expiry date gives both urgency and importance to FFS reforms. However, it goes without saying that what is to be phased out needs to be identified first. It is also important that the deadline is binding and stringent enough to provide certainty to FFS reforms. The deadline set out in the Ise-Shima Agreement falls far short of such standards. There is no doubt, however, that the agreement will move the discussion on FFSs one step forward. Only over the last few years has the question surrounding FFSs shifted from whether they should be removed to how best to remove them. The Ise-Shima Agreement marks the beginning of another phase in the FFSs debate – when should FFSs be eliminated and at what rate? Will one size fit all?  

[1] See G7 Ise-Shima Leaders’ Declaration: G7 Ise-Shima Summit, 26-27 May 2016 (Ise-Shima Agreement). In evaluating the potential impact of international agreements, legal analysis often focuses on their binding nature. However, since such an agreement is not yet forthcoming, debating that point here would be futile. Like its predecessors, the Ise-Shima Agreement is not binding.  

[2] See Cindy Carr, ‘G7 Leaders Pledge To End Fossil Fuel Subsidies By 2025’ (Sierra Club, 2016) < accessed 30 May 2016.

[3] See Shelagh Whitley, ‘G7 Must Stop Propping up Doomed Fossil Fuel Industries’ (Climate Home, 2016) < accessed 30 May 2016.

[4] See Paris Agreement to the United Nations Framework Convention on Climate Change (adopted 12 December 2015) FCCC/CP/2015/L.9 (Paris Agreement), art.2.1(a).                                           

[5] See ibid art.3&4.

[6] See IEA, Energy and Climate Change: Special Report of World Energy Outlook (International Energy Agency 2015) at 20.

[7] See IEA, World Energy Outlook 2012 (International Energy Agency 2012) at 25.

[8] At least 126 countries around the world had implemented renewable energy support policies of one form or another as of early 2015. See Renewable Energy Policy Network for the 21 Century, Renewables 2015: Global Status Report (REN21 2015) at 91.

[9] As of early 2015, carbon pricing has been implemented in 40 national and 20 subnational jurisdictions around the world. See World Bank and Ecofys, State and Trends of Carbon Pricing 2015 (The World Bank 2015).

[10] ‘Statement Made at Cancun Climate Negotiations in December 2010’ <> accessed 30 May 2016.

[11] David Coady and others, ‘How Large Are Global Energy Subsidies?’ (IMF 2015) IMF Working Paper WP/15/105; IEA, World Energy Outlook 2015 (International Energy Agency 2015).

[12] Javier Arze del Granado, David Coady and Robert Gillingham, ‘The Unequal Benefits of Fuel Subsidies: A Review of Evidence for Developing Countries’ (International Monetary Fund 2010) IMF Working Paper.

[13] See Coady and others (n 13) at 7.

[14] Ad Hoc Working Group on the Durban Platform for Enhanced Action Second session, part eight Geneva, 8–13 February 2015, Negotiating Text, FCCC/ADP/2015/1, 25 February 2015 (Paris Agreement Negotiating Text) paras.81 and 128(1)(d).

[15] It is perhaps in recognition of this that at least 13 countries included FFS reform in their INDCs to the Paris Agreement Anika Terton and others, ‘Fiscal Instruments in INDCs: How Countries Are Looking to Fiscal Policies to Support INDC Implementation’ (International Institute for Sustainable Development 2015) Discussion Paper.

[16] See UNFCCC, Completion of a Protocol or Another Legal Instrument: Consolidated Negotiating Text by the Chairman, 13 October 1997, FCCC/AGBM/1997/7.

[17] See Kyoto Protocol to the United Nations Framework Convention on Climate Change (adopted 11 December 1997, entered into force 16 February 2005) 2303 UNTS 148 (Kyoto Protocol) art.2.1(a)(v).

[18] See G-20 Leaders’ Statement: The Pittsburgh Summit, September 24-25, 2009 (Pittsburgh Agreement).

[19] See Asia Pacific Economic Community, 2009 Leaders’ Declaration, Singapore 14 Nov 2009; United Nations Resolution adopted by the General Assembly on 27 July 2012 – the Future We Want, (11 September 2012) A/RES/66/288, 66th Session (UNGA Resolution 66/288); Leaders’ Declaration G7 Summit 7-8 June 2015.

[20] See IEA, World Energy Outlook 2015 (n 13).

[21] See Henok Birhanu Asmelash, ‘Falling Oil Prices and Sustainable Energy Transition: Towards a Multilateral Agreement on Fossil Fuel Subsidies’ (UNU-WIDER 2016) WIDER Working Paper No 2016/13.

[22] Since the Paris Agreement is yet to be ratified, sustaining the momentum that led to its successful conclusion is vital for both its ratification and implementation. 

[23] For a detailed discussion on FFS definitions, see Kerryn Lang, ‘Defining Fossil-Fuel Subsidies for the G-20: Which Approach Is Best?’ (Global Subsidies Initiative 2010) Policy Brief.

[24] See IEA, World Energy Outlook 2014 (International Energy Agency 2014); Benedict J Clements and others, Energy Subsidy Reform: Lessons and Implications (International Monetary Fund 2014).

[25] Department of Energy & Climate Change, ‘Re: Freedom of Information Request, 17 August 2015, Ref: FOI 2015/15308’ < accessed 1 June 2016.

[26] See Kyle Mandel, ‘What Does the G7 Pledge to Phase Out Fossil Fuel Subsidies Mean for the UK?’ < accessed 1 June 2016.

[27] See Elizabeth Bast and others, ‘Empty Promises G20 Subsidies to Oil, Gas and Coal Production’ (Overseas Development Institute & Oil Change International 2015) at 80. Also entirely excluded from the UK definition are post-tax subsidies (the difference between efficient and actual taxation).