Environmental Research Letters LETTER • OPEN ACCESS A multi-level climate club with national and sub-national members: theory and application to US states To cite this article: Nick Martin and Jeroen C J M van den Bergh 2019 Environ. Res. Lett. 14 124049 View the article online for updates and enhancements. This content was downloaded from IP address 92.57.32.56 on 23/06/2020 at 16:06 Environ. Res. Lett. 14 (2019) 124049 https://doi.org/10.1088/1748-9326/ab5045 LETTER A multi-level climate club with national and sub-national members: OPEN ACCESS theory and application to US states RECEIVED 3 June 2019 Nick Martin1 and Jeroen C J M van den Bergh1,2,3 REVISED 1 5 October 2019 Institute of Environmental Science and Technology, Universitat Autònoma de Barcelona, Bellaterra, Spain 2 Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain ACCEPTED FOR PUBLICATION 3 School of Business and Economics & Institute for Environmental Studies, VU University Amsterdam, Amsterdam, The Netherlands 22 October 2019 PUBLISHED E-mail: nicholas.martin@uab.cat 11 December 2019 Keywords: climate policy, climate club, UNFCCC, Paris Agreement, carbon import tariff Supplementary material for this article is available online Original content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Abstract Any further distribution of The minilateral approach of a climate club of countries has been suggested as an intermediate phase in this work must maintain attribution to the a transition towards a global agreement that enforces national climate policies through harmoniza- author(s) and the title of the work, journal citation tion. To garner critical mass, we propose an extended club configuration including sub-national states and DOI. or provinces, resulting in a multi-level club. This would allow considerable contributions from important emitters like the US to be brought on board, relevant given its intended withdrawal from the Paris Agreement. We elaborate this idea and clarify potential roles of participants at distinct levels. The concept is operationalized by developing a method for identifying suitable entities at each level that uses a set of likelihood-of-involvement indicators capturing existing carbon dependence, public opinion, government policy and climate coalition membership. Application at the national level identifies a subset of seven of the highest emitting countries representing 21% of global emissions. This rises to 51% assuming that China, the dominant global emitter, could be enticed into the group. Given that US involvement remains unlikely for now, we illustrate selection at the sub-national level for US states. Here, an initial group of 21 states appear as potential members, jointly accounting for 36% of national emissions. An additional group, representing a further 34% of emissions, are potentially receptive to enticement via trade dependencies on four key countries identified within the group of national members. Accordingly, some 70% of US emissions, representing 11% of global totals, may be subject to climate club involvement via a combination of these pathways. While the implementation of such a club requires various political and legal hurdles to be overcome, the ongoing threat of climate change and inadequacies of the Paris Agreement suggest that novel solutions of this kind deserve serious attention from scientists and politicians. 1. Introduction far—the defunct Kyoto Protocol and the current Paris Agreement—have both relied on voluntary actions, The United Nations Framework Convention on meaning that levels of commitment and compliance Climate Change (UNFCCC) has now been widely among member countries have not been uniform in accepted as the default facilitator of global negotiations nature. Accordingly, the incentive for self-interested on climate issues (Karlsson-Vinkhuyzen and McGee parties to free-ride continues to be present. 2013). However, the multilateral approach it employs Acknowledging the limitations of this approach, has been slow to address the ongoing threat of global many are now suggesting the importance of alternative warming, primarily owing to the logistical limitations solutions. One of the more promising of these is the of large groups and the voluntary nature of involve- minilateral approach of a climate coalition or club of ment (Nordhaus 2015). Indeed, the two most ambi- ambitious countries that implements a uniform pol- tious initiatives undertaken by the UNFCCC thus icy, most likely some form of carbon pricing. This © 2019 The Author(s). Published by IOP Publishing Ltd Environ. Res. Lett. 14 (2019) 124049 so-called ‘climate club’ would offer exclusive trade 2. Climate clubs beyond countries benefits or club goods to members and seek to attract further membership by imposing penalties on imports Interest in minilateral solutions to climate change has from non-members to limit unfair competition from increased in recent years and a great deal of discussion unregulated production sources. By creating moral regarding the characteristics and potential benefits of and economic pressure on other countries, the club climate club arrangements can be found in the would likely expand over time. Once critical mass is literature (Leycegui and Ramírez 2015, Nordhaus achieved, it could grow to cover the bulk of global 2015, Victor 2015, Falkner 2016, Gampfer 2016, Hovi emissions and potentially influence post-Paris et al 2016, Keohane et al 2017, Nordhaus 2017, UNFCCC negotiations to focus on global policy coor- Victor 2017, Sprinz et al 2018, Hagen and Eisenack dination rather than mere targets (Eckersley 2012, 2019, Paroussos et al 2019). Nevertheless, discussions Nordhaus 2015). on the subject to date have tended to assume that Attempting to affect environmental change in membership would be limited to national actors. Yet, smaller groups outside of the UNFCCC framework in countries with considerable spatial variation in has a number of key benefits. Firstly, it is thought that terms of economic activities and climate policies, smaller groups enable negotiations to be undertaken membership could also be offered to sub-national more efficiently, which may result in faster and more governments who wish to deviate from national streamlined responses (Biermann et al 2009, Weischer actions by implementing their own, more ambitious et al 2012). This is especially critical when mitigating a climate policies. Indeed, a growing number of such problem such as climate change, where time is a deci- government bodies are mobilizing to address climate sive factor. Secondly, smaller groups of like-minded change and implement stricter emissions targets participants are able to tackle ‘narrow-but-deep’ goals within coalition groups, often as a direct reaction to and achieve much more focused and ambitious out- weak policy or inaction at the national level. comes than larger disparate groups (Aldy et al 2003). As the effectiveness of a climate club is largely Additionally, the relatively simple and egalitarian determined by the share of emissions and economic nature of climate clubs that results from homo- influence of its members, we contend that a multi- geneous market policies presents less barriers and level club could target motivated potential members encourages entry to a broader spectrum of stake- with high economic pull and emissions reduction holders (Biermann et al 2009, Weitzman 2014). capacities regardless of the strata of government they Lastly, countries which have consistently blocked occupy. Furthermore, the participation of significant ambitious proposals at UNFCCC Conference of the sub-national entities may ultimately also be capable of Parties meetings—particularly oil exporters such as imposing critical pressure on national governments to Russia and Australia—will continue to do so, mean- consider club membership. ing that the current mechanisms of the UNFCCC are Nevertheless, as club members, sub-national gov- unlikely to lead to global policy harmonization (King ernments are unlikely to be capable of imposing trade and van den Bergh 2019). It is particularly telling in barriers on imports from non-members. For example, this respect that the Nationally Determined Con- it is doubtful that individual states within the United tributions of the ten largest fossil fuel producers States (US) could legally impose trade restrictions on under the Paris Agreement lack any policy limiting other countries as members of a climate club. Theore- fossil fuel supply, with the exception of India (Piggot tically, Article 1, section 10 of the US Constitution pre- et al 2017). vents states from imposing their own trade policies on Notwithstanding such observations, another com- other states or foreign countries unless permitted to do pelling aspect of climate club membership lies in the so by Congress. Such actions have proven difficult in economic benefits it provides. Empirical modeling the past and would seem very unlikely to be approved that tested four target carbon prices and 11 tariff rates by the current legislative branch of the federal govern- for non-members for a range of club configurations ment, particularly for purposes that directly challenge found that net gains in welfare were achieved for the the policies of the current administration. club under all regimes, even using lower carbon prices However, sub-national members such as US states and tariff rates (Nordhaus 2015). Higher benefits were would still be able to implement the uniform climate reaped as the prices and tariffs increased. And, while policies of the club internally, be exempted from not all members benefitted under all combinations carbon border tariffs on trade with national club tested, individual net gains were achieved 88% of the members, benefit from club goods such as R&D coop- time. Tellingly, many of the simple regimes tested eration and, in the case of high-profile sub-national resulted in all seven major emitters achieving simulta- members, provide legitimacy to the club as a whole. neous benefits. As such, climate clubs are not only cap- This introduces a club design consisting of two types able of providing a feasible reduction in emissions but of members: (1) countries that can enjoy club goods are thought to offer more immediate financial benefits and both employ and avoid penalties; and (2) sub- to members under the majority of configurations. national governments that can enjoy club goods and 2 Environ. Res. Lett. 14 (2019) 124049 Table 1. Criteria used for assessing likelihood-of-involvement of 15 countries with the highest carbon emission rates. Criteria Description Relevance Source Carbon independence ‘Cleanliness’ or lack of dependence on Reflects ability to conform to club World Bank (2017a, 2017b) carbon-intensive industries in emission standards or integrate generating income. Defined as GDP other carbon-related mechanisms earned per tonne of CO2 emitted into local economy. A lower value of this indicator increases the likelihood of considering club membership Public opinion Composite of data from two public Reflects voter concern about climate Pugliese and Ray (2009) and regarding climate opinion polls relating to climate change and support of mitigation Stokes et al (2015) change change issues strategies. Indicates likelihood of prioritizing climate change and considering club membership Government policy Quantified assessment of Nationally Reflects current government policy Climate Action Tracker position Determined Contribution (NDC) on climate change. Indicates (2017), Robiou du Pont pledge under the Paris Agreement. likelihood of considering club and Meinshausen (2018) membership Climate coalition Involvement of governments in Reflects current policy on climate Individual coalition membership current climate-focused coalitions change. Indicates likelihood of membership records considering club membership be enticed into clubs to avoid penalties imposed by a location’s ambitions regarding climate policy and existing members. hence its potential willingness to be involved in an At the receiving end, the application of trade international climate club. penalties within a multi-level climate club arrange- As a first step, the maximum score for each criter- ment may indeed be complicated by the fact that sub- ion was used to normalize individual scores on a scale national members are positioned within the trade bor- from 0 to 1. Where more than one source exists for a ders of a non-member country and are thus liable to given criterion, normalized scores were initially calcu- pay the penalties that apply to the country as a whole. lated for the individual sources and then averaged to Administratively, it would seem complex for a climate generate an overall score for the criterion. The final club to only apply penalties to certain regions of a non- likelihood-of-involvement scores were then derived member country. A simpler solution could be to apply by aggregating the normalized scores for all criteria. penalties to all sales from non-member countries and This was achieved by calculating a geometric mean as then return those penalties to sub-national members this method is believed to deliver a more meaningful as part of a ‘refund’ scheme. outcome for indicators of this kind than using an arithmetic mean (Ebert and Welsch 2004). A higher final aggregated score is interpreted as indicating that a 3. Assessing potential club members country or state is more motivated to undertake cli- mate action and, therefore, more likely to consider In order to predict the ‘likelihood-of-involvement’ for membership in a climate club. A full description of the governments at multiple levels, we developed a data, method and sensitivity analysis used is provided method that utilizes four complementary criteria: (1) in the supplementary materials available online at carbon independence; (2) public opinion regarding stacks.iop.org/ERL/14/124049/mmedia. climate change; (3) government policy position; and (4) climate coalition membership. The first criterion quantifies the lack of dependence on carbon-intensive 4. Potential national members industries by comparing the gross domestic product (GDP) or gross state product (GSP) earned per tonne Likelihood-of-involvement scores were first calculated of CO2 emitted. The second criterion captures the level for the 15 countries4 with the highest carbon emission of localized voter support for action on climate change rates utilizing the data sources listed in table 1. The and, thus, the likelihood of governments implement- results, as shown in figure 1, suggest that two distinct ing more stringent climate policies. The third criterion groups of countries exist. A group of seven ‘likely’ club considers the current political leanings or policy members—including the EU, Japan, South Korea, positions of government bodies with respect to climate Canada, Brazil, Mexico and Australia–were deemed to issues. The fourth and final criterion assesses the be the most receptive to club membership. These extent of local involvement in the growing number of countries represent approximately 20.5% of global climate-related coalition groups. While these criteria may correlate, they are largely complementary in 4 The term ‘countries’ is used hereafter for all potential national scope and are believed to provide a good indication of members, including the European Union (EU). 3 Environ. Res. Lett. 14 (2019) 124049 Figure 1. Normalized indicators and final ‘net likelihood’ scores for top 15 carbon emitting countries in order of current CO2 emission levels (World Bank 2017a). Large red rings denote seven ‘likely’ countries and small-sized rings denote eight ‘not likely’ countries. A threshold line delineating the two categories is also shown. carbon emissions. It is also noted that this group economies’. Moreover, the EU currently has active includes all Group of Seven (G7) members except trade agreements with three other members of the the US. ‘likely’ group—Japan, South Korea and Mexico—and A relative gap of some 25.5% then exists between only one member of the ‘not likely’ group (South Japan (0.606) and the US (0.451) as the seventh and Africa). This suggests that a substantial level of influ- eighth highest scoring countries, respectively. This ence and existing affiliations exist within the identified represents a significant increase compared to the ‘likely’ countries, with the EU as the member with the intervals observed within the ‘likely’ group, where highest apparent level of potential influence. changes were generally around 2%–3% and all were The capacity of the EU to guide initiatives of this less than 9%. Accordingly, this acts as the delineation kind is not only demonstrated by empirical data. It point between the ‘likely’ group and a second group of also has a long history of leadership within the sphere eight ‘not likely’ club members. This group consists of of climate change mitigation activities (Parker and China, the US, India, Russia, Iran, Saudi Arabia, South Karlsson 2010, Schmidt and Fleig 2018) and has tradi- Africa and Indonesia. tionally shown greater commitment to UNFCCC pro- Because these results were derived by taking the grams than either China or the US (Keohane and geometric means of four normalized indicators, equal Victor 2011, Sommer and Hain 2017). Following a weighting was implicitly applied. As a sensitivity test, period of reinvention in the UNFCCC following the six alternative combinations using two weightings at 2009 Copenhagen Summit, the EU adopted the addi- 10% and another two at 40% were examined (see sup- tional role of both leader and mediator by attempting plementary materials). Although slight variations to reconcile the growing concern of the developing occurred in their order, the countries within each world with the apparent reticence of the US and China group remained constant for all combinations sug- (Bäckstrand and Elgström 2013). As such, it played a gesting a consistency and robustness in the delineation key role in the coordination of what was to become the between the two groups. Paris Agreement (Oberthür and Groen 2017). As the third-highest emitter with the second-high- est GDP, the EU is the most desirable potential club member within the group of more ‘likely’ countries in 5. Challenges at the national level terms of potential emissions reduction and economic influence. Four EU countries—France, Germany, Notwithstanding the elevated likelihood of EU mem- Italy and the United Kingdom—alongside Canada and bership, the second ‘not likely’ group includes both the Japan represent six of the seven members of the G7, a US and China—the two highest emitters with the group considered by the International Monetary Fund highest and third-highest GDPs. Two such high- to represent the world’s seven largest ‘advanced profile countries remaining outside such a club is likely 4 Environ. Res. Lett. 14 (2019) 124049 to greatly limit its chances of success. Indeed, previous Climate Alliance—were formed as a direct reaction to modeling simulations found that climate clubs are far Trump’s election and expected withdrawal (Crooks more likely to be successful with US involvement 2017, Worland 2017). (Hovi et al 2019, Sprinz et al 2018). Despite this apparent gridlock at the national level, It is nevertheless worth noting that previous mod- the proliferation and consolidation of determined eling simulations also suggest that it would be rela- sub-national governments suggests that US involve- tively easy for others to entice China into a club (Hovi ment in international climate clubs could take the et al 2019) and that, alongside India, China has the form of membership by motivated states as part of a most to gain from club membership (Paroussos et al multi-level configuration. Existing coalitions may 2019). This is predominantly the result of the high even be capable of building upon their present rela- volume of emissions it produces, the high cost of tionships to negotiate with—or even aid in initiating energy transitions and its vulnerability to the ongoing —burgeoning clubs as a collective unit. Furthermore, impacts of global emissions. As the world’s highest as a notably decentralized country, where individual emitter of carbon—producing some 30.4% of global states are able to enact relatively high levels of political totals—China imposes a huge cost on its own future, decision-making at the local level, the US is an ideal and should be less sensitive to free-rider temptation candidate for operationalizing the multi-level club than smaller players. What is more, it already appears notion. to be implementing more sustainable domestic energy As many states control significant economies with policies (Green and Stern 2017) and to be making high emissions in their own right—some exceeding greater efforts to raise its profile on the international those of many of the top 20 countries in both of these climate stage. Furthermore, China and the EU have dimensions—such an arrangement could offer a path- recently announced a joint commitment to intensify way for more enthusiastic governments to sidestep the implementation of the Paris Agreement and current administration and achieve significant reduc- strengthen their ties on climate change and clean tions in overall US emissions. This could be achieved energy issues (European Commission 2018). In light via involvement of single states, or integration of of this evidence, it is assumed that, in reality, China is regional coalitions of states, in pre-existing interna- more likely to consider club membership than our cal- tional clubs. Operational climate clubs may also prove culations suggest. This is especially true if the EU were capable of using the market-based mechanisms that to initiate such a club. As a result, 50.9% of global car- drive them to entice otherwise reluctant US states bon emissions would be encompassed within this towards a climate club arrangement, particularly group of eight countries. where strong trade dependencies with member coun- In any case, what remains clear is that, considering tries exist. the Trump administration’s ongoing intention to withdraw from the Paris Agreement in 2020, US mem- bership in a climate club regime would appear unlikely 6. Potential US state members for now. That leaves it as the ‘odd one out’ of the three key countries highlighted as preferred members of a In order to estimate the likelihood of individual US potential climate club situation. Being the second- states to consider membership in an international highest emitter with the highest GDP, the non- climate club the previous method for assessing poten- inclusion of the US would represent a major challenge tial members was applied to US states utilizing the data for the feasibility of implementing a truly effective glo- sources listed in table 2. Recognizing the complexity of bal climate club. political mechanisms in the US, it is acknowledged Although Trump’s decision to step away from the that one cannot fully capture the myriad factors that Paris Agreement came as no surprise, the news was determine the likelihood of state involvement in a greeted with widespread condemnation in the US and climate club. Nevertheless, the chosen data is thought abroad (Diringer 2017). Perhaps the most visible reac- to provide a reasonable indication of the general public tion to the announcement was the rapid mobilization and governmental tendencies of each state. of coalitions of companies and sub-national govern- The assessment ultimately assigns each state as ment bodies specifically denouncing the decision and being ‘very likely’, ‘moderately likely’ or ‘not likely’ to vowing to uphold the spirit of the Paris Agreement and consider involvement in a climate club situation. prioritize climate change mitigation efforts. Two of While classifying groups is always somewhat arbitrary, the most significant—the Under2 Coalition, a global three groups were defined based on the size of the rela- alliance spearheaded by California governor Jerry tive gaps between scores. Brown, and the Mayors National Climate Action Ultimately, 12 of the 50 states were selected as Agenda (‘Climate Mayors’)—were organized in being ‘very likely’ to consider club membership. The advance of the 2015 Paris conference but subsequently relative gap between the lowest ranked state in this first revitalized by Trump’s statement (Alvarez 2017, group, New Jersey (12th, 0.838), and the next state, Henderson 2017, State of California 2017). The most New Hampshire (13th, 0.803), corresponds to a notable others—We Are Still In and the United States change of 4.2%, the highest of the first 12 intervals 5 Environ. Res. Lett. 14 (2019) 124049 Table 2. Criteria used for assessing likelihood-of-involvement of US states. Criteria Description Relevance Source Carbon independence ‘Cleanliness’ or lack of dependence Reflects ability to conform to club Bureau of Economic Analysis on carbon-intensive industries in emission standards or integrate (2017) and US Energy Informa- generating income. Defined as other carbon-related mechan- tion Administration (2018) GSP earned per tonne of CO2 isms into local economy. A emitted lower value increases the like- lihood of considering club membership Public opinion Public opinion poll data relating to Reflects voter concern about cli- Howe et al (2015) regarding climate climate change issues mate change and support of change mitigation strategies. Indicates likelihood of prioritizing cli- mate change and considering club membership Government policy Composite indication of political Reflects current government com- Federal Election Commission position position (Republican or position considering the (2016), United States Senate otherwise) based on presidential increasingly partisan division on (2019), United States House of voting and composition of climate issues in US politics. Representatives (2019), legislative branches at federal States with lower Republican Ballotpedia (2019a, 2019b) and state level Party influence are broadly assumed to be more likely of consider club membership Climate coalition Involvement of governments in Reflects motivation to undertake Individual coalition membership membership current climate-focused action on climate change. Indi- records coalitions cates likelihood of considering club membership (most others are less than 2%). Observation of the data ‘moderately likely’ and ‘not likely’ groups based on the for these states also reveals a notable consistency in size and location of the observed gaps and trends high scoring across most or all categories when com- observed in the data for the individual states within pared to the states that follow. The resulting group of this range. ‘very likely’ states collectively cover a total of 17.8% of So, while the nine states within the final ‘moder- the total emissions and 36.6% of GDP. ately likely’ group display less noteworthy results New York and California are the key candidates than the former (‘very likely’) group, they do never- for climate club membership in this group; their like- theless offer some suggestion of a willingness to con- lihood scores occupy the second and fourth rankings, sider membership in initiatives such as climate clubs while they control the third-largest and largest state when compared to the 29 remaining states. This economies, respectively, jointly accounting for 22.1% group of ‘moderately likely’ states collectively of total GDP. In fact, as countries, California and New account for 18.2% of emissions and 19.8% of GDP. York would represent the fifth and eighth largest Considering their emissions and GSP, the two most economies in the world. What is more, California is significant states in this list are Illinois and Florida, the second-highest emitting state while New York sits among states the fourth and sixth highest emitters at number nine. with the fifth and fourth largest economies, Almost all gaps between scores for the remaining respectively. 38 states are less than 2%. The only notable exceptions The final results for each state, in order of CO2 are the 5.6% interval between Virginia (14th, 0.786) emissions, are shown in figure 2. All ‘very likely’ and and Maine (15th, 0.742), the 4.6% interval between ‘moderately likely’ states are further summarized in Florida (21st, 0.699) and New Mexico (22nd, 0.667) table 3 while a complete listing of the indicator data is and the 5.2% interval between Arizona (27th, 0.641) provided in the supplementary materials. and Ohio (28th, 0.608). As the gap between Virginia As a sensitivity test, six alternative combinations and Maine occurs only two positions after the previous using two weightings at 10% and another two at 40% group this interval is not deemed suitable. Meanwhile, were again examined (see supplementary materials). the individual characteristics of states as far down the While slight variations occurred in the order, the states list as Arizona indicate diminishing tendencies contained within each group remained largely con- towards climate action and these could not realistically stant with a small number of exceptions. The top 12 be considered ‘likely’ states. Accordingly, the interval ‘very likely’ group members remained within the between Florida and New Mexico is thought to pro- group for all combinations tested with the exception of vide the most suitable delineation between the Delaware and New Jersey which transferred out of the 6 Environ. Res. Lett. 14 (2019) 124049 Figure 2. Normalized indicators and final ‘net likelihood’ scores for US states in order of current CO2 emission levels (US Energy Information Administration 2018). Large red rings denote 12 ‘very likely’ states, medium-sized rings denote nine ‘moderately likely’ states and small rings denote 29 ‘not likely’ states. Threshold lines delineating the three categories are also shown. The names of all states identified as ‘very likely’ and ‘moderately likely’ are highlighted in gray. Table 3. Indicator scores for 12 ‘very likely’ states and nine ‘moderately likely’ states, ranked in final order of likelihood. Likelihood-of-involvement State % of total US CO2 emissions % of total US GDP Score Ranking Rating Mass 1.2 2.7 0.961 1 Very likely N York 3.1 8.1 0.953 2 Connecticut 0.6 1.4 0.924 3 California 6.6 14.0 0.919 4 Maryland 1.1 2.1 0.882 5 R Island 0.2 0.3 0.876 6 Vermont 0.1 0.2 0.862 7 Washington 1.4 2.5 0.859 8 Oregon 0.7 1.2 0.858 9 Delaware 0.2 0.4 0.850 10 Hawaii 0.3 0.5 0.847 11 N Jersey 2.1 3.2 0.838 12 N Hampshire 0.3 0.4 0.803 13 Moderately likely Virginia 1.9 2.7 0.786 14 Maine 0.3 0.3 0.742 15 Minnesota 1.8 1.8 0.735 16 Illinois 4.3 4.3 0.725 17 Nevada 0.7 0.8 0.721 18 Colorado 1.7 1.8 0.711 19 Michigan 3.0 2.6 0.704 20 Florida 4.2 5.0 0.699 21 top group during four scenarios. Both states possess three states have exceptionally low membership in high public opinion and government scores and mod- coalitions despite scoring well in public opinion and erate carbon dependence but are sensitive to variations government categories. As a result, they fell out of the in coalition membership weighting as a result of their group during six, three and one of the sensitivity sce- low scores in this category. narios, respectively. However, as the groupings were Similar observations pertain to Michigan, Nevada generally shown to be resilient to these changes, and and Illinois within the ‘moderately likely’ group. All the observed substitutions all related to the coalition 7 Environ. Res. Lett. 14 (2019) 124049 Figure 3. US state exports sold to four key countries and combined sum of all four as percentage of gross state product (GSP). Threshold line representing the 75th percentile of combined scores is also shown. States previously identified as ‘very likely’ and ‘moderately likely’ climate club members are highlighted in gray. criterion, we can conclude that the results are con- Canada, Mexico and China. Since the previous analy- sistent and robust overall. sis of potential members at the national level found the To summarize, a total of 21 states were found to involvement of three of these to be likely, and provided display tendencies that mark them as more likely can- additional reasons for China to be included, the four didates for considering club membership, represent- key US trading partners could potentially exert pres- ing a total of 36.0% of national emissions and 56.3% of sure on certain US states to join the climate club. With GDP. The remaining 29 states were deemed to be ‘not this in mind, figure 3 reports data on state exports to likely’ in the analysis. Significantly, ten of the highest each of these countries and their sum (‘Four Com- 15 emitters are in this group, including Texas as the bined’) as a percentage of the GSP for that state (see highest overall emitter with the second-highest GSP. supplementary materials for full listing). Previously As we demonstrate in the following section, a func- identified ‘likely’ states are highlighted in gray. tional climate club could entice some of these states Considering all 50 states, the third quartile (75th into membership via a trade-related mechanism. percentile) of these scores is some 5.7% of GSP. The results indicate that exports from several states— 7. State-country trade as a push force for many of which are among the highest emitters—are additional state membership well above this level. What’s more, a total of 10 pre- viously ‘not likely’ states—accounting for 33.9% of The ability of operational multi-level climate clubs to national emissions and 21.4% of GDP—are among entice new members can be further understood by those with trade dependencies on the EU, Canada, examining existing international trade links with US Mexico and China that are well above the national states. The previous analysis attempted to identify states average. This includes five of the seven highest emit- with higher likelihoods of considering climate club ting ‘not likely’ states, all of whom are within the coun- membership solely based on their individual character- try’s top 12 highest emitting states overall. istics. However, once a club is in operation, member The most notable state in this list is Texas—the top nations may be capable of enticing even the less ‘likely’ carbon emitter with the second-highest GSP—which US states via the power of existing trade dependencies. sells 61.4% of its total exports to these four countries States with particularly high dependencies on club and 37.0% to Mexico alone. These exports account for member countries may consider membership simply in some 10.6% of the state’s total GSP, making it the order to maintain beneficial trade relationships and avoid fourth highest of all 50 states in this category. Indeed, the imposition of trade penalties on non-members. Texas was highlighted as being particularly vulnerable Examining international exports from individual to revenue losses when Donald Trump threatened to US states (United States Census Bureau 2018) identi- withdraw from the North American Free Trade Agree- fied the top four US trading partners as the EU, ment in 2018 for this very reason (Roberts 2018). 8 Environ. Res. Lett. 14 (2019) 124049 Other notable states in the list include Louisiana and architectures that could evolve should climate clubs Kentucky—the seventh and twelfth highest carbon become a reality and offers an example of trade-related emitting states—which have the third and second- mechanisms that could influence further club growth. highest proportion of exports to GSP, respectively. Nevertheless, we fully recognize the legal and political On the strength of such results, it seems plausible hurdles that climate clubs could face, including those that the EU, Canada, Mexico and China could be capable employing a multi-level structure. Such hurdles have of using the power of these trade dependencies to entice already received much attention in the literature, and still additional high-profile US states into considering mem- merit further research, particularly in the context of the bership in club arrangements. This is especially likely if ongoing discussion regarding environmental policy and all four are able to apply club mechanisms in unison. protection measures permitted by the World Trade Orga- nization (Cottier et al 2009). Here we merely intended to offer a new angle on the composition of climate clubs and 8. Discussion and conclusions the selection of potential members at multiple levels. While it is easy to judge proposals for genuinely The climate club is emerging as a promising alternative for solving climate change as politically difficult, con- mitigating global climate change. Recognizing the urgency sidering the limitations of the Paris Agreement and the of addressing climate change and the shortcomings of the urgency of implementing effective climate action all-inclusive UNFCCC approach, a climate club prioritizes sooner rather than later, the time is ripe for debating deeper reductions in carbon emissions by smaller groups daring solutions. More ambitious policies within a of more motivated governments. However, the chances of small coalition of powerful and motivated countries such a club succeeding would appear to be greatly and sub-national governments already promise sig- improved by US involvement, an unlikely scenario for the nificant emissions reduction in the short run. More time being. Acknowledging the potential of sub-national importantly, through moral and economic pressure government involvement in minilateral actions of this exerted on non-members, such a climate club could kind, it was demonstrated that climate club arrangements expand over time and thus significantly contribute to a that include US states may be able to maximize US transition towards a global policy level playing field. participation in lieu of involvement at the national level. This would allow for harmonized and gradually more Based on various motivation-level indicators, 12 stringent climate policies worldwide. of the 50 US states were found to be ‘very likely’ to con- sider club membership while a further nine states were deemed to be ‘moderately likely’ to do so. In total, Acknowledgments these states account for 36.0% of national emissions and 56.3% of GDP. Even so, 29 states were found to be This study has received funding through an ERC ‘not likely’ to join a climate club based on these indica- Advanced Grant from the European Research Council tors. Tellingly, ten of the 15 highest emitting states are (ERC) under the European Union’s Horizon 2020 in this group, including Texas as the highest overall research and innovation programme (grant agreement emitter with the second-highest GSP. no 741087). Further investigation revealed that even less moti- vated US states could potentially still be enticed into an Data availability statement operational climate club via strong export dependencies, particularly with the four key US trading partners—the The data that support the findings of this study are EU, Canada, Mexico and China. In fact, five of the seven available from the corresponding author upon reason- highest emitting ‘not likely’ states, and 10 ‘not likely’ able request. The data are not publicly available for states in total, were found to be among the highest quar- legal and/or ethical reasons. ter of states when considering their individual export dependencies in relation to total GSP. Collectively, these states represent some 33.9% of emissions and 21.4% of ORCID iDs GDP, suggesting that a club including the four key US trading partners may be capable of augmenting club Nick Martin https://orcid.org/0000-0001-9023-9696 membership significantly via trade influences. This is Jeroen C J M van den Bergh https://orcid.org/ consistent with one important role of a climate club 0000-0003-3415-3083 recognized in the literature, namely to put pressure on exports of non-members so as to encourage them to References become club members. In all, the analysis suggests that states representing a total of 69.9% of national emis- Aldy J E, Barrett S and Stavins R N 2003 Thirteen plus one: a sions, 10.8% of global emissions and 77.7% of total US comparison of global climate policy architectures Clim. Policy 3 373–97 GDP may be amenable to club membership. Alvarez L 2017 Mayors, sidestepping Trump, vow to fill void on The method devised for assessing likely members climate change New York Times (https://nytimes.com/ within the study provides a broad first analysis of the 2017/06/26/us/mayors-trump-climate-change.html) 9 Environ. Res. Lett. 14 (2019) 124049 Bäckstrand K and Elgström O 2013 The EU’s role in climate change Centre for Trade and Sustainable Development (ICTSD) and negotiations: from leader to ‘leadiator’ J. Eur. Public Policy 20 World Economic Forum) (www.e15initiative.org/) 1369–86 Nordhaus W 2015 Climate clubs: overcoming free-riding in Ballotpedia 2019a Partisan Composition of State Houses (https:// international climate policy Am. Econ. Rev. 105 1339–70 ballotpedia.org/Partisan_composition_of_state_houses) Nordhaus W 2017 Climate clubs and carbon pricing Global Carbon Ballotpedia 2019b Partisan Composition of State Senates (https:// Pricing: The Path to Climate Cooperation ed P Cramton et al ballotpedia.org/Partisan_composition_of_state_senates) (Cambridge, MA: MIT Press) pp 109–23 Biermann F, Pattberg P, van Asselt H and Zelli F 2009 The Oberthür S and Groen L 2017 The European Union and the Paris fragmentation of global governance architectures: a agreement: leader, mediator, or bystander? Wiley Interdiscip. framework for analysis Glob. Environ. Polit. 9 14–40 Rev. Clim. Change 8 1–8 Bureau of Economic Analysis 2017 Regional Economic Accounts Parker C F and Karlsson C 2010 Climate change and the European (https://bea.gov/regional/downloadzip.cfm) Union’s leadership moment: an inconvenient truth? Climate Action Tracker 2017 Improvement in Warming Outlook as J. Common Mark. Stud. 48 923–43 India and China Move Ahead, But Paris Agreement Gap Still Paroussos L, Mandel A, Fragkiadakis K, Fragkos P, Hinkel J and Looms Large (https://climateactiontracker.org/documents/ Vrontisi Z 2019 Climate clubs and the macro-economic 61/CAT_2017-11-15_ImprovementInWarmingOutlook_ benefits of international cooperation on climate policy Nat. BriefingPaper.pdf) Clim. Change 9 542–6 Cottier T, Nartova O and Bigdeli S Z 2009 International Trade Piggot G, Erickson P, Lazarus M and van Asselt H 2017 Addressing Regulation and the Mitigation of Climate Change: World Trade Fossil Fuel production Under the UNFCCC: Paris and Beyond Forum ed T Cottier et al (Cambridge: Cambridge University (Working paper 2017-09) (Seattle: Stockholm Environment Press) Institute) (https://sei.org/mediamanager/documents/ Crooks E 2017 US states form alliance to meet Paris climate Publications/SEI-2017-WP-addressing-fossil-fuel- commitments Financ. Times (https://ft.com/content/ production.pdf) 27c5bad2-4895-11e7-919a-1e14ce4af89b) Pugliese A and Ray J 2009 A heated debate: global attitudes toward Diringer E 2017 Let Trump claim a better deal on climate Nature climate change Harvard Int. Rev. 31 64–8 (www.jstor.org/ 546 329 stable/42763325) Ebert U and Welsch H 2004 Meaningful environmental indices: a Roberts K 2018 Texas Stands to Lose Most in NAFTA Turmoil, social choice approach J. Environ. Econ. Manage. 47 270–83 Chart of 50 States Shows Forbes (https://forbes.com/sites/ Eckersley R 2012 Moving forward in the climate negotiations: kenroberts/2018/09/29/texas-stands-to-lose-most-in- multilateralism or minilateralism? Glob. Environ. Polit. 12 nafta-turmoil-chart-of-50-states-shows/) 24–42 Robiou du Pont Y and Meinshausen M 2018 Warming assessment European Commission 2018 EU and China Step Up Cooperation of the bottom-up Paris agreement emissions pledges Nat. on Climate Change and Clean Energy (https://ec.europa.eu/ Commun. 9 4810 clima/news/eu-and-china-step-cooperation-climate- Schmidt N M and Fleig A 2018 Global patterns of national climate change-and-clean-energy_en) policies: analyzing 171 country portfolios on climate policy Falkner R 2016 A minilateral solution for global climate change? On integration Environ. Sci. Policy 84 177–85 bargaining efficiency, club benefits, and international Sommer B and Hain A 2017 Europe as a green leader? A brief legitimacy Perspect. Polit. 14 87–101 evaluation of both the European Union’s Climate and energy Federal Election Commission 2016 Federal Elections 2016: Election policy and common agricultural policy Cult. Pract. Eur. 2 Results for the US President, the U.S. Senate and the US 33–45 House of Representatives (Washington, DC) (https:// Sprinz D F, Sælen H, Underdal A and Hovi J 2018 The transition.fec.gov/pubrec/fe2016/federalelections2016.pdf) effectiveness of climate clubs under Donald Trump Clim. Gampfer R 2016 Minilateralism or the UNFCCC?The political Policy 18 828–38 feasibility of climate clubs Glob. Environ. Polit. 16 62–88 State of California 2017 Governor Brown Welcomes New Members Green F and Stern N 2017 China’s changing economy: implications to Under 2 Coalition (https://gov.ca.gov/2017/09/17/ for its carbon dioxide emissions Clim. Policy 17 423–42 news19956/) Hagen A and Eisenack K 2019 Climate clubs versus single coalitions: Stokes B, Wike R and Stewart R 2015 Global Concern About Climate the ambition of international environmental agreements Change, Broad Support for Limiting Emissions (Washington Clim. Chang. Econ. 10 1–19 DC: Pew Research Center) (http://pewglobal.org/files/ Henderson P 2017 California, Other States Vow to Lead as Trump 2015/11/Pew-Research-Center-Climate-Change-Report- Exits Paris Climate Pact Reuters (https://reuters.com/ FINAL-November-5-2015.pdf) article/us-usa-climatechange-states-idUSKBN18S6PI) United States Census Bureau 2018 State and Metropolitan Area Hovi J, Sprinz D F, Sælen H and Underdal A 2016 Climate change trade data (https://census.gov/foreign-trade/statistics/ mitigation: a role for climate clubs? Palgrave Commun. 2 state/index.html) 16020 United States House of Representatives 2019 Directory of Hovi J, Sprinz D F, Sælen H and Underdal A 2019 The club Representatives (https://house.gov/representatives) approach: a gateway to effective climate co-operation? Br. J. United States Senate 2019 Senators (https://senate.gov/senators/ Polit. Sci. 49 1071–96 index.htm) Howe P D, Mildenberger M, Marlon J R and Leiserowitz A 2015 US Energy Information Administration 2018 Energy-Related Geographic variation in opinions on climate change at state Carbon Dioxide Emissions by State, 2000–2015 (Washington, and local scales in the USA Nat. Clim. Change 5 596–603 DC: US Department of Energy) Karlsson-Vinkhuyzen S I and McGee J 2013 Legitimacy in an era of Victor D G 2015 The Case for Climate Clubs, E15Initiative (Geneva: fragmentation: the case of global climate governance Glob. International Centre for Trade and Sustainable Development Environ. Polit. 13 56–78 (ICTSD) and World Economic Forum) (www.e15initiative. Keohane N, Petsonk A and Hanafi A 2017 Toward a club of carbon org/) markets Clim. Change 144 81–95 Victor D G 2017 Three-dimensional Climate Clubs: Implications for Keohane R O and Victor D G 2011 The regime complex for climate Climate Cooperation and the G20 (Geneva: International change Perspect. Polit. 9 7–23 Centre for Trade and Sustainable Development (ICTSD) and King L C and van den Bergh J C J M 2019 Normalisation of Paris World Economic Forum) (https://ictsd.org/sites/default/ agreement NDCs to enhance transparency and ambition files/research/three-dimensional_climate_clubs_victor_ Environ. Res. Lett. 14 084008 climate_and_energy_final_0.pdf) Leycegui B and Ramírez I 2015 Addressing Climate Change: A WTO Weischer L, Morgan J and Patel M 2012 Climate clubs: can small Exception to Incorporate Climate Clubs (Geneva: International groups of countries make a big difference in addressing 10 Environ. Res. Lett. 14 (2019) 124049 climate change? Rev. Eur. Community Int. Environ. Law 21 World Bank 2017a CO2 Emissions (kt) from Carbon Dioxide 177–92 Information Analysis Center, Environmental Sciences Weitzman M L 2014 Can negotiating a uniform carbon price help to Division, Oak Ridge National Laboratory, Tennessee, United internalize the global warming externality? J. Assoc. Environ. States (https://data.worldbank.org/indicator/EN.ATM. Resour. Econ. 1 29–49 CO2E.KT) Worland P 2017 A New US Environmental Alliance is Trying to World Bank 2017b GDP (Current US$) from World Bank National Take Trump’s Place on the World Stage Time (http://time. Accounts Data (https://data.worldbank.org/indicator/NY. com/4806118/michael-bloomberg-states-climate-change/) GDP.MKTP.CD) 11