Effects of asymmetric policies to achieve emissions reduction on energy trade: A North American perspective

Resources, Environment and Sustainability 18 (2024) 100179 Contents lists available at ScienceDirect Resources, Environment and Sustainability journal homepage: www.elsevier.com/locate/resenv Research article Effects of asymmetric policies to achieve emissions reduction on energy trade: A North American perspective Sara Giarola a,b, Iván García Kerdan c,∗, Peter Johnston d,1, Nick Macaluso d,1, Baltazar Solano Rodriguez e,f, Ilkka Keppo e,g, Adam Hawkes a, David Daniels h aChemical Engineering Department, Imperial College London, London, UK bDepartment of Management Engineering, Polytechnic of Milan, Milan, 20141, Italy cTecnologico de Monterrey, School of Engineering and Sciences, Mexico dEnvironment and Climate Change Canada, Gatineau, Quebec, Canada eUCL Energy Institute, University College London, London, UK fTransition Modelling Lab, London, UK gDepartment of Mechanical Engineering, Aalto University, Espoo, Finland hThe Swedish National Road and Transport Research Institute (VTI), Gothenburg, Sweden ARTICLE INFO Keywords: Asymmetric carbon policy Global fossil fuel trade Energy systems model Carbon leakage North America ABSTRACT The implementation of asymmetric emission reduction policies can not only increase the cost of reducing emissions but also reduce the effectiveness of climate policies themselves, leading to policy inefficiencies such as carbon leakage. This paper investigates the impact of asymmetric emission reduction policies on the costeffectiveness and efficiency of climate strategies in North America. Using a model inter-comparison approach, which combines two bottom-up global models and one top-down global model, this study assesses the effects of such policies on fuel substitution, global fossil fuel trade, and emissions in North America and globally. It is the first work where a multi-model approach is used for exploring how different energy systems react to asymmetric carbon policies. This provides critical insights into regional policy design within a global emissions framework. Quantitatively, the study reveals that asymmetric carbon pricing can lead to more than 60% global emissions reduction in certain models, but can also drive trade distortions, where U.S. exemptions result in emissions rising by more than 10% compared to reference scenarios. Qualitatively, significant fuel substitution patterns across Canada, Mexico, and the U.S. demonstrate increased coal consumption when carbon prices are unevenly applied. While no global emission increase was observed, asymmetric policies result in inefficiencies between local policy costs and emissions reduction outcomes, such as rising fossil fuel trade in non-abating regions. The findings suggest that harmonising carbon policies across regions would reduce inefficiencies and minimise carbon leakage. 1. Introduction In the context of a fast-pacing societal change, the scientific community has been calling for years for a transformation of the energy systems facilitated by international collaboration to achieve a net reduction of the anthropogenic emissions and mitigate the effects of global climate change (IPCC,2007;COP21,2015). This call for cooperation has not yet been translated into internationally agreed policies to best enable an effective transition to a lower greenhouse gas (GHG) emitting energy system. There are 75 carbon pricing instruments in operation worldwide, comprising both emissions trading systems (ETS) and carbon taxes. Although these instruments are implemented across ∗Corresponding author. E-mail address: [email protected] (I. García Kerdan). 1Views expressed in this paper are those of the authors and do not reflect those of Environment and Climate Canada or the Government of Canada. 53 national and 40 sub-national jurisdictions, covering about 24% of global emissions (World Bank,2024a,b), they remain insufficient to meet the Paris Agreement goals. The World Bank reports (World Bank,2024a) that carbon pricing revenues reached a record $104 billion in 2023, but only a small fraction of global emissions is priced at the levels recommended to limit global temperature rise to well below 2◦C. In presence of not-unified policies, businesses may transfer production to other countries which have less stringent constraints on GHG emissions. This situation, named carbon leakage, makes policies overall inefficient. For instance, the biggest carbon trading scheme in the world, the EU ETS (European Emissions Trading Scheme), can https://doi.org/10.1016/j.resenv.2024.100179 Received 8 August 2024; Received in revised form 15 October 2024; Accepted 6 November 2024 Available online 15 November 2024 2666-9161/©2024 The Author(s). Published by Elsevier B.V. on behalf of Lishui Institute of Ecology and Environment, Nanjing University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 produce carbon leakage through three channels: (i) a direct cost deriving from reducing emissions inversely proportional to the ETS cap; (ii) another cost derived from buying the initial amount of permits in an auction; and (iii) an indirect cost caused by the electricity price increases due to the CO2emission mitigation costs. When the third EU ETS phase was announced, with the more stringent EU-wide cap on emissions, 170 sectors were identified as potentially exposed to carbon leakage (European Commission,2018). As the fourth phase of the EU ETS (2021–2030) (European Commission,2024) has introduced even stricter caps on emissions, energy audits, climate neutrality plans and a Carbon Border Adjustment Mechanism, supposed to phase out free allocations, will be in place to mitigate carbon leakage risks. In such a complex international context, the issue of asymmetric carbon pricing poses significant challenges to the energy transition in North America (NA) and in the whole world. Despite national and subnational differences, not only NA is a region with strong interlinks between the three countries (Canada, Mexico, and the United States (U.S.)) from both the economic and the energy resource perspective but also a major energy exporter in the global market. With differing carbon policies across Canada, Mexico, and the U.S., the risk of carbon leakage, policy misalignment, and trade distortions increases, which could undermine the effectiveness of carbon reduction strategies. This misalignment not only complicates energy system planning at a national level but also impacts regional energy trade and fuel substitution patterns, key drivers in achieving broader climate goals. As the urgency to reduce greenhouse gas emissions intensifies at a global scale, this research provides crucial insights into how varied carbon pricing mechanisms across interconnected economies can affect both local, regional, and global emissions targets. By addressing these asymmetries through energy systems models, the study contributes to designing more integrated and effective climate policies for the region. This paper explores the energy systems transformation, with a focus on energy systems and trade flow changes specifically of fossils (crude oil, natural gas, and refined liquids), within NA and between NA and the rest of the world by analysing the EMF-34 scenarios of fragmented and global action for climate change mitigation. The methodology is based on the use of energy systems modelling to explore future pathways and deviations from a reference scenario in an intercomparison model study where the scenarios impose either symmetric or asymmetric geographical coverage of the carbon price. The focus is on domestic fuel mix and international trade, with the aim to assess patterns which could lead to higher/lower propensity of each country to increase/reduce their energy linkages at an intra-regional or international level in the context of more/less collaborative climate mitigation which would result in presence of symmetric or asymmetric carbon pricing. The rest of the paper is organised as follows, after an overview of the approaches present in the literature to assess asymmetric carbon policies, the methodology will be presented, followed by the case study definition,results and discussions and conclusions. 2. Literature survey 2.1. Contributions on asymmetric policies There is a broad literature on the effects of asymmetric carbon pricing on trade, with significant variation in estimates of the magnitude of carbon leakage (Larch et al.,2018). Most contributions are based on regression analyses of bilateral flows (empirical input–output models) and estimate the deviations in embodied carbon of traded goods from global trade and production datasets. Regarding the geographical applications (see ) most studies concentrated on CO2emissions embodied in North–South (i.e. developed - developing countries) trades, with China and Brazil among the relevant developing countries as trading partners (Schaeffer and de Sá,1996). 2.2. Contributions on asymmetric policies: a focus on North America Regarding the applications related to asymmetric carbon policies, the CO2flows in North–North trade, i.e. among developed countries, have not been analysed much. Among the limited number of North– North contributions, Wang and Zhou (2019) studied the embodied CO2 emissions between Germany and U.S.. Sarker et al. (2023) investigated the asymmetric effects of climate policy uncertainty (CPU), geopolitical risk (GPR), and crude oil prices (WTI) on clean energy prices (CEP) in the U.S.. Using a non-linear autoregressive distributed lags (NARDL) model from January 2001 to December 2021, they found that CPU, GPR, and WTI influence CEP returns and volatility differently over time, with significant implications for investment decisions. Although empirical contributions differ in energy sector mapping, geographical focus, and the detail of the decomposition analysis performed on bilateral trade. the implications on energy dynamics which depend on inter-related aspects, such as socio-economic growth, emission standards and performance, as well as techno-economic features of energy technologies can only be studied with energy systems models (Weyant,2012). Only a small number of works, led by the Energy Modeling Forum (EMF), have addressed the implications of energy trade flows using an energy systems modelling approach. Energy systems models, depicting scenarios of possible energy futures under pre-defined assumptions, can be used alongside advanced modelling techniques such as Artificial Intelligence (AI), to enhance the accuracy of energy projections (McLaughlin and Choi,2023). Within the context of EMF34, the group addressed the fragmented analysis of energy markets in Canada, Mexico, and the United States, emphasising the need for integrated approaches. Huntington et al. (2020) analysed results from 17 models and underscored the importance of improved data quality and consistency to enhance policy decision-making on energy trade. Siddiqui et al. (2020) examined the impacts of altering cross-border electricity transmission and natural gas transport costs on energy trade policies for North America and found that, in absence of cross-border mechanisms promoting renewables, the expansion of the electricity transmission capacity between Canada and the U.S. could increase natural gas-driven electricity production. Lastly, Brown et al. (2021) focusing on the effects of natural gas market disruptions on energy security, evaluated responses to hypothetical price shocks across the three countries and showed that flexible policy designs can help navigate volatile markets and enhance regional stability. Beyond the work focused on the intra-regional trade carried out by EMF, in terms of global trade, the most recent reference scenario of the Energy Information Administration (EIA) reports that the United States continues to be a net exporter of natural gas, largely due to the growth in liquefied natural gas (LNG) exports (EIA,2024). The EIA projects this trend will persist with LNG exports leading the growth. However, for petroleum liquids, the U.S. is anticipated to remain a net exporter through 2040, driven by ongoing increases in crude oil production and rising global demand for U.S. petroleum products. The International Energy Agency (IEA,2024) also notes the significant impact of U.S. shale oil and gas on global markets, predicting the U.S. will maintain its net exporter status for both fuels into the 2030s, although the share of exports is expected to decline as a result of climate policies and national energy security programmes (see Table 1). 3. Research gap and contributions This paper is the first to assess the effects of asymmetric carbon policies as a driving force behind carbon leakage, specifically by measuring deviations in fossil fuel demand – namely, the trade flows of natural gas, oil, and refined products – from a North American perspective. This focus is relevant because energy systems in exporting regions like North America are vulnerable to the uneven enforcement of global CO2 emissions policies as well as to intra-region policy misalignment. 2 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Table 1 Key findings from literature review on asymmetric carbon policies. Study Year Region Study’s focus Key findings Guo et al. (2010) 2010 China, US Sino-US trade -Suggested that the introduction of low-carbon technologies in China could significantly reduce embodied carbon in traded goods, indicating the potential for collaboration in achieving global emission reduction targets. Clò (2010) 2010 Europe Impact of EU ETS on trade flows -The cessation of the grandfathering mechanism in the EU ETS raised concerns over carbon leakage, prompting sectors to relocate to regions with less stringent emissions regulations, indicating significant potential risks to European industries. Böhringer et al. (2012b) 2012 Global Economic efficiency of tariffs -Showed that import tariffs on embodied carbon could complement emissions pricing, improving economic efficiency by accounting for indirect emissions, thus enhancing the overall effectiveness of climate policies. Weyant (2012) 2012 Global Border carbon adjustments -Highlighted the effectiveness of border carbon adjustments in mitigating carbon leakage, particularly for energy-intensive industries. Emphasised that unilateral climate policies may lead to competitive disadvantages without appropriate measures. De Cian et al. (2013) 2013 Europe European climate policy interactions -Emphasised the interplay between European climate policy and international regimes, demonstrating how global actions influence regional trade flows and emissions, necessitating coordinated policies. Dong et al. (2015) 2015 China Border carbon adjustments -Analysed the implications of border carbon adjustments on Chinese trade, concluding that such measures could preserve competitiveness but might have limited impact on domestic emissions reductions. Deng and Xu (2017) 2017 Global Multiregional input– output analysis -Identified that the embodied carbon in global trade varies significantly among top economies, with different contributions from imports, exports, and domestic consumption, underscoring the need for nuanced policies addressing trade dynamics. Larch et al. (2018) 2018 Global Estimation of carbon leakage -Found significant variability in carbon leakage estimates across different models, highlighting the importance of model choice and assumptions in predicting trade impacts under asymmetric carbon policies. He and Hertwich (2019) 2019 49 countries Flow of embodied carbon -Demonstrated that direct and indirect emissions depend heavily on import/export shares of intermediates, revealing how trade relationships influence national emission intensities and calling for tailored policy responses. Huntington et al. (2020) 2020 North America Response of energy production to policies -Analysed how carbon taxes and renewable policies influence oil and natural gas production, stressing the need for improved data consistency across borders to inform effective policy-making in North America. Sarker et al. (2023) 2023 USA Climate policy uncertainty -Investigated the asymmetric effects of climate policy uncertainty on clean energy prices, showing that volatility in energy markets is influenced by geopolitical risks and crude oil prices, which affect investment decisions. 3.1. Intra-region and global trade of North America The three countries in the NA region, have a 30 year-long agreement ruling their commercial relations, the North American Free Trade Agreement (NAFTA), which has been updated into the United States Mexico Canada Agreement (USMCA) (called CUSMA in Canada and T-MEC in Mexico). As a result of their geographical proximity and longlasting trade negotiations, Canada, Mexico, and the U.S. have become privileged trade partners of primary resources and refined products. As of 2023, on the Canada-U.S. side, over 80% of Canada’s crude oil supply and 45% of the natural gas supply went to U.S. From 2010 to 2021, natural gas exports to the US decreased by 18% due to increased natural gas production in the northeast U.S. (Canadian Association of Petroleum Producers,2024). On the U.S. side, Canada is the dominant source of crude oil imports, being 5 times larger than the second most dominant source: Mexico. Also, around 97% of U.S. natural gas imports came from Canada (Government of Canada,2020). Conversely, the bilateral Mexico-U.S. energy trade has shown growing volumes of petroleum products (mainly motor gasoline and distillate fuel oil) imports from the U.S. to Mexico and declining volumes of Mexican crude oil sold to the U.S., although Mexico remains one of largest crude oil exporters to the U.S. behind Canada. In 2022, the total value of energy trade between the United States and Mexico reached $81.9 billion (in real prices), setting a new record. U.S. crude oil imports from Mexico averaged 536,000 barrels per day, a 9% increase from 2021, with the value rising by 47% to $20.7 billion due to higher global prices. Mexico was the largest export market for U.S. petroleum products, accounting for nearly 20% of exports, with volumes increasing 33% to 1.7 million barrels per day and value rising to $44.2 billion. U.S. natural gas exports to Mexico were 5.7 billion cubic feet per day, a 4% decrease from 2021, with a 1.5% decline in trade value (EIA,2023). Besides the strong intra-regional integration based on the trade of fossil fuels, NA has important trade of energy goods with the rest of the world. Globally, the U.S., a net exporter of refined products, is the largest refined products exporter (EIA,2020), whereas Canada is the fourth largest crude oil exporter (Government of Canada,2020). 3 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 3.2. Emission intensity regulation in North America North American countries have intensified efforts to mitigate emission intensity, for example: •Canada: Initiatives involve collaboration among government bodies, oil companies, and NGOs to evaluate methods for reducing oil and gas production (CAPP,2023). •United States: There is a strong commitment to renewable energy expansion, with states like California implementing policies to lessen reliance on fossil fuels (CARB,2022). •Mexico: The Energy Transition Law sets ambitious clean energy and emissions reduction targets (SEMARNAT,2021). Despite these efforts, a unified regional agreement to significantly curb emissions is elusive and missing in the revised NAFTA agreement. 3.3. Research gap This paper not only contributes to the understanding of asymmetric carbon policies and their implications for energy trade but also offers valuable insights for policymakers aiming to design effective climate strategies in a region characterised by complex energy interdependencies. 1. Novel Focus: It uniquely examines the implications of asymmetric carbon pricing on fossil fuel demand in North America, filling a critical gap in the literature regarding regional dynamics and carbon leakage. 2. Methodological Innovation: By employing a model comparison approach that harmonises assumptions across heterogeneous models, this research contrasts traditional statistical decomposition methods. The analysis utilises energy systems models to depict future scenarios for North America’s energy system under various ‘‘what-if’’ assumptions. 3. Scenario Analysis: This study contrasts the effects of a globally adopted carbon price with those stemming from individual state policies (Mexico and Canada) or coalitions (North America or other countries excluding the U.S.). This approach highlights the significant interlinkages in energy trade within North America and with the rest of the world, demonstrating how unequal implementations of carbon pricing could impact trade dynamics and emissions outcomes. 4. Policy Implications: The findings underscore the necessity for coordinated carbon pricing mechanisms in North America to mitigate carbon leakage and enhance the effectiveness of climate policies. 4. Methodology A model inter-comparison approach is used to address the implications of carbon policies on the trade of fossils fuels, primarily, crude, natural gas, and petroleum products of NA. Carbon pricing is used as a proxy for carbon policies and is applied with an asymmetric geographical distribution within the NA region compared to the rest of the world. Three global models are used: •𝐸 𝐶−𝑀 𝑆 𝑀 𝑅: is a recursive dynamic CGE model, applying a solution approach based on the equilibrium between demand and supply of all the commodities in the market (general equilibrium). The model uses a top-down approach to the modelling of energy futures, where key technologies in each energy sector are modelled as well as the energy sector interaction with the whole economy (Zhu et al.,2018). The model solves investment, consumption, and production decisions related to the energy systems only by the economic situation in the current period and investment in the prior period. •𝑀 𝑈 𝑆 𝐸(the ModUlar energy system SImulation Environment): is an agent-based simulation model of the energy system, as it explicitly characterises the decision-making process of multiple agents, representing firms and consumers in the energy system, with knowledge imperfection and biases (Giarola et al., 2022). The model balances the demand and supply of the energy commodities (partial equilibrium) using a limited foresight approach (García Kerdan et al.,2019), which means that it solves a configurable number of years for which there is knowledge of demand and price projections. The model is rich in energy technologies, each of which is modelled in costs, energy consumption, and environmental emissions (bottom-up approach). Modelled decisions relate to investments including technology retirement, new capacities, as well as operations such as production level, fuel consumption, emissions, and constraints on maximum uptake and production limits. •𝑇 𝐼 𝐴𝑀−𝑈 𝐶 𝐿(The TIMES Integrated Assessment Model): is an integrated assessment model which uses an optimisation framework to solve the energy system, adopting a social planner perspective to maximise the economic surplus for both consumers and suppliers of energy. The model offers a rich suite of technologies, which includes currently available technologies and future options. Each technology is characterised with a bottom-up approach, using a technical, economic, and environmental characterisation as well as the inclusion of constraints on maximum capacity uptake and operation (Loulou and Labriet,2008). The model works using an inter-temporal optimisation approach which ensures knowledge of commodity demand, prices, technology availability, and costs over the full simulated period. The chosen models are highly complementary. A computable general equilibrium model, such as 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅, offers an analysis of the energy market as well as of all the commodity markets, providing an overview on how the energy system would interact with the whole economy and with the rest of the world with trade. Energy systems models focus on energy commodities and on a representation of the energy systems highly rich in energy technologies. A simulation model, such as 𝑀 𝑈 𝑆 𝐸, projects a world where investors’ behaviours may make the system departing from an optimal solution. An optimisation model, such as 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿provides a platform for estimating how energy system operations could evolve maximising the net surplus, whereas meeting a specific target, such as renewable penetration and limits on emissions. The three models use a multi-sectoral approach to the modelling of the energy systems, which identifies key sectors: the extraction of primary energy resources (which includes fossils, biomass, and nuclear supply curves); the conversion of this primary energy into useful forms (such as electricity and fuels), and the use of these fuels in a range of energy service applications, such as transportation, building heating and cooling, and the powering of industrial manufacturing plants. In Appendix A, an overview of the model characteristics is presented. Harmonisation approach. All three models represent the energy flows among the three largest countries of NA: Canada, the U.S., and Mexico. For the rest of the world, the three models are characterised by a different regional breakdown. In Table 2, a mapping of the regions across the models is presented. The models were harmonised in two ways: •Service demand projections: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿and 𝑀 𝑈 𝑆 𝐸use demand projections based on SSP2 (Riahi et al.,2017), which is a ‘‘Middle of the Road’’ scenario, aiming to reproduce a continuation of historical trends. 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅uses multiple external projections to generate the reference case; from a comparison with the key components of the SSP2 pathway, the external projections used are within the suggested ranges for CO2emissions, GDP and primary energy values in 2100. 4 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Table 2 Region-to-region matching across models. EC-MSMR MUSE TIAM-UCL Canada Canada Canada U.S. U.S. U.S. Mexico Mexico Mexico Europe 9 regions: Denmark, Norway, Iceland, Sweden, Finland, EU18 (remaining Western Europe) EU7 (Eastern Europe), Emerging Europe, Switzerland and Turkey 2 regions: Western Europe and Eastern Europe Japan Japan Japan Australia Oceania (Australia and New Zealand) Oceania (Australia and New Zealand) New Zealand Russia 2 regions: Russia, Caspian region Russia and Central Asia China China China India India India South Korea South Korea South Korea Brazil Brazil South Africa South Africa OPEC 2 regions: Middle East and Israel Middle East Central and South America Central and South America Central and South America Chile Rest of the World 4 regions: Other Developing Asia, ASEAN, Africa (except South Africa), and Africa 2 regions: South and Central Asia and Africa •2015 energy balance across the models: This involves the calibration of base year stock for the energy technologies. The calibration of the global trade flows was partial due to the different regional disaggregation of the models which did not allow a full trade mapping; the focus was on the bilateral flows within NA and between NA and the rest of the world. The harmonisation does not cover the techno-economics of the energy sectors, as assumed within the EMF-34 study framework. In addition, the three models (CGE, agent-based, and optimisation) apply different approaches to policy modelling implying that this harmonisation level was not achievable in the study. For example, bottom-up models, such as 𝑀 𝑈 𝑆 𝐸and 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿need to explicitly model technology-specific policies. Specifically, the retirement of unabated coal in Canada and clean energy power generation targets in Mexico, are explicitly modelled. Differently, 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅does not explicitly impose any policy for the increase or decrease in the use of specific technologies, but the baseline, being calibrated to energy, emission and economic information from the 2019 Annual Energy Outlook (AEO) for U.S., 2019 Reference Case from E3MC for Canada, and the 2017 International Energy Outlook (IEO) for all remaining regions, would incorporate all policies at the time of release in regards to retirement regulations. Models apply different approaches regarding revenues from carbon tax: while bottom-up models such as 𝑀 𝑈 𝑆 𝐸and 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿did not include ways to re-integrate carbon tax into the economy, in 𝐸 𝐶− 𝑀 𝑆 𝑀 𝑅model, all of the carbon tax revenues are recycled directly back to the regions representative agent who will use the revenues to make capital and labour allocation decisions maximising their own utility through increasing consumption. Differences in costs and policies among the three models are in the reference scenario. The scenario structure focuses on assessing the differential implications of additional efforts beyond the reference scenario (e.g., the asymmetric imposition of a carbon price) in the form of deviations from the reference scenario. The lack of harmonisation on policies is an advantage of this work as it recognises the inherent uncertainty of policy futures and avoids modelling biases introduced by asserting a common future policy environment. 5. Case study The case study is organised using a set of five scenarios: one reference and four sensitivities. This approach allows one to assess deviations from a reference scenario, including parametric and structural differences across the models in mitigation policies, technologies, and their techno-economic characterisation. From the reference, four additional scenarios are built considering the effects on emission reduction coming from either the imposition of a global carbon price profile or the imposition of the same carbon price profile to selected groups of countries, NA and its subsets, in order to assess the relative deviation on fossil trade and emission within NA and between NA and the world. The scenarios are: •A reference scenario (Ref.) is built considering the modeller’s settings for techno-economic parameters (technology types, costs, efficiencies, and growth rates) as well as policies. In terms of mitigation policies, the models differ to the extent that stated policies (IEA,2019) are implemented and these deviations are embedded in the reference scenario. Due to the global nature of the models used, and the global structure of the models, subnational policies are not implemented in this model comparison exercise. An overview of current sub-national policies is presented in Appendix Bas a reference for the readers. •A scenario (Ctax.WORLD) in which a uniform carbon price is applied in NA and in the rest of the world, shows the impact of a global energy transition on NA trade. This scenario highlights the relative impacts of additional policies limiting the use of fossils on the energy systems in each country of the NA region compared to the rest of the world. •A scenario (Ctax.NA) where a uniform policy on carbon price is applied in NA but not in the rest of the world, demonstrates the consequences on NA exports of energy commodities. A local carbon price might produce an increase in exports due to a reduced local demand. However, the higher generation costs for fuels induced by local carbon policies could stimulate the non-NA countries to increase their national capacities and be less reliant on imports from NA. 5 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Table 3 Carbon price trajectory, 2015 USD per t of CO2. Year 2020 2030 2040 2050 Carbon price 21 64.77 105.51 105.51 •A scenario where a uniform carbon price is applied worldwide except in the U.S. (Ctax.WORLDxUS), highlights the consequences of the U.S. failing to enact decarbonisation policies. While fossil consumption could be maintained in the absence of carbon policies, an asymmetric carbon price could increase the costs of imported fuels. •A scenario with a uniform carbon price applied in Mexico and Canada but not in the U.S. or the rest of the world (Ctax.CANMEX), shows the consequences on the trade of a highly fossil exporting country (like Canada) and a highly importing country (like Mexico). A first impact could be seen in the trade relation between Canada and Mexico with the U.S., as the carbon policies might reduce local fuel consumption and therefore reduce energy trade with the U.S. A second impact can be seen in comparison with the rest of the world: as Canada has a large export volume of crude oil, local production might still increase sustained by exports. The scenarios apply the same carbon price trajectory reported in Table 3and only differ for the geographical distribution of the carbon price, according to the scenario definition. The carbon price trajectory is aligned with the assumptions of the EMF-34 study. Accordingly, a carbon price of 21, 28, and 35 USD (2015) would apply for 2020, 2021, and 2022; from 2022 onwards a 5% yearly increase would apply until 2040; after 2040, it is assumed that the taxation regime would remain constant. This study does not analyse deep global decarbonisation; rather, it focuses on deviations in emission reductions in each country of the NA region and globally, when mediated by a symmetric or an asymmetric carbon price. The chosen carbon price trajectory, milder than those which can be seen in deep mitigation studies (Rogelj et al.,2015), is in line with the assumptions of the general EMF-34 study. 6. Results In the analysis, the energy systems are studied from 2015 through to 2050 with a 5-year step. Results are presented from 2020, which is the first modelled year after the calibration year. Results are organised as follows: A discussion of the variations observed between the reference scenario and the sensitivities is presented first in terms of local energy consumption and trade flows of refined products, crude oil, and natural gas, then in terms of CO2emissions. Further charts denoted with C are reported in the Appendix D. 6.1. Domestic consumption of fossils NA is expected to meet a growing energy service demand; this results in a growing final energy consumption of fossils as well as of electricity generation for the NA region. In Canada, oil products in the energy systems are increasingly consumed in 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅, driven by an increasing demand across the end use sectors. 𝑀 𝑈 𝑆 𝐸has a peak in consumption of oil products in 2040 for transportation. 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿shows a more relevant fuel substitution in the Ref. scenario, driven by cost competitiveness of natural gas technologies. In presence of a carbon price, the consumption of oil products diminishes in all the models. In 2050, when the carbon tax reaches its maximum, the most relevant deviations occur with differences depending on an asymmetric imposition of carbon price among the countries. •In 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅, the reductions in domestic fossil consumption are the greatest when the carbon tax applies to Canada and Mexico only (Ctax.CAN-MEX), as this configuration would lead to a large reduction of oil products imports (Fig. 4) •In 𝑀 𝑈 𝑆 𝐸oil products consumption remains relatively unaffected by the levels of the assumed carbon price •In 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿similarly to 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅, displays the largest reductions of oil products when an asymmetric imposition of carbon price in NA applies (Ctax.CAN-MEX). This comes also with higher imports, which are used domestically for a small fraction and are then re-exported to the rest of the world (Fig. D.33) The Ref. energy mix shows a growing dependence on natural gas in Canada (Fig. 1). However, while this is the dominant fossil fuel in 𝑀 𝑈 𝑆 𝐸and 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅,𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿shows a growing demand of coal. It can be observed that: •Natural gas-fired electricity generation does see a lot of expansion for Canada in 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅as coal-fired electricity generation is phased out in 2030 except for certain provinces who have negotiated equivalency agreements. •In 𝑀 𝑈 𝑆 𝐸the phase-out of unabated coal power plants triggers more natural gas plants. Coal is then substituted with natural gas generation as renewable power generation costs does not decrease as fast as in 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅(Fig. D.20) and renewables (Fig. D.22). •In 𝑇 𝐼 𝐴𝑀 natural gas gains an increasing share of the electricity generation, but it competes with the availability of cheaper coal generation, not forced to phase out by policies. As a carbon tax is applied, different dynamics can be seen in the three models •In 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅, natural gas tends to decrease with little effects of asymmetric carbon price •𝑀 𝑈 𝑆 𝐸shows a decrease in the consumption of natural gas. The decrease in this fossil demand is more rapid when a global carbon tax applies (Ctax.WORLD and Ctax.WORLDxUS) and is accompanied by a higher uptake of renewables in the power sector. •In 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿the future consumption of local natural gas consumption is more influenced by the U.S. policies. When a carbon tax applies in the whole world but not in the U.S., (Ctax.WORLDxUS), this would reduce natural gas consumption. The domestic fuel substitution promoted by a carbon tax, could lead to the increase in crude oil imports, but more sensitivity analyses would be needed for a definitive answer (Fig. D.38). Conversely, the implementation of homogeneous carbon prices in NA reduces the availability of cheaper crude oil and favours the consumption of natural gas. When a carbon price applies globally (Ctax.WORLD) and within NA (Ctax.NA and Ctax.CAN-MEX), a higher electrification rate occurs in the energy system •Electricity generation moves from natural gas to hydropower and renewable sources in 𝐸 𝐶−𝐸 𝐶 𝑀 𝑆 𝑅(Fig. D.22). Oil products consumption in the energy systems, which reflects the consumption of oil in the power sector, decreases the most when a carbon tax applies to either NA only or to the cluster Canada and Mexico (Ctax.NA and Ctax.CAN-MEX), aligned with a minimum import of crude oil. •In 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿shows an energy system less electrified when a carbon price applies. Coal consumption diminishes quite substantially only if a carbon tax applies globally (Ctax.WORLD), but remains relatively constant or increases as the carbon price does not apply homogeneously. This suggests that coal trade with U.S. and with the rest of the world would be empowered as a carbon tax applies exclusively to NA or to the cluster CAN-MEX. 6 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Fig. 1. Natural gas consumption in Canada. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Mexico continues to steadily use oil products in the Ref. scenario (Fig. D.16), mainly driven by the transportation sector demand, where fuel efficiency justifies the slight downward trend, and reduction of oil products in the power generation tends to decrease with time in all the models, as shown in (Fig. D.23). In the presence of a carbon price, the model response varies depending on its geographical allocation. •In 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅the consumption of oil products decreases with time, both in the demand sector and power generation, but no relevant consequences can be attributed to asymmetric policies implementation as the import of oil products remains quite unchanged (Fig. 5); •In 𝑀 𝑈 𝑆 𝐸, results show an upward trend of oil products consumption, due to increased demand from transportation, which, as the service grows, would need a much higher carbon price compared to the modelled values to show a relevant fuel substitution; •In 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿, results display a downward trend as the carbon price applies. It is interesting to note that oil products consumption is remarkably dependent on the inclusion of the U.S. in the cluster of countries adopting a carbon price, as the Ctax.NA leads to the minimum oil products consumption. The Ctax.WORLDxUS and Ctax.CAN-MEX scenarios lead to the highest levels of oil products consumption, as a result of increased imports from the U.S. A coal-to-gas switch can be seen in all the models. The coal-based generation tends to reduce over time (Fig. D.24), favouring natural gas (see Fig. 2), and hydropower in all models (Fig. D.26), although not leading to a complete phase-out. Renewables other than hydropower are less cost-effective in 𝑀 𝑈 𝑆 𝐸compared to natural gas, justifying a higher uptake of natural gas in the power sector compared to other models (Fig. D.25). In presence of carbon price: •In 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅, results show the lowest coal consumption. An asymmetric carbon price does not lead to major deviations, with the exception of an increase in coal power generation when the U.S. does not apply a carbon price; •In 𝑀 𝑈 𝑆 𝐸, results show a reduction in the use of domestic coal; the rate of substitution is higher if a carbon tax only applies to Canada and Mexico (Ctax.CAN-MEX); •In 𝑇 𝐼 𝐴𝑀−𝑈 𝐶 𝐿coal consumption increases if a carbon price only applies to Canada and Mexico (Ctax.CAN-MEX), driven by higher coal power generation as a cheaper alternative to renewables. The results also reflect the modelling approach used by 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿 which being a global cost optimisation model could lead to global solution which generates local discontinuities of the kind. In the U.S., oil products consumption decreases over time due to improvements of fuel efficiency. In presence of a carbon price applied domestically, the oil products consumption decreases in all the models (Fig. D.17). Asymmetric carbon price implementation produces small fuel substitutions in which scenarios Ctax.NA and Ctax.WORLD align closely with each other as the scenario Ctax.WORLDxUS would do with the scenario Ctax.CAN-MEX. The consumption of natural gas (Fig. 3) is driven by the share of natural gas in the power sector. Depending on the rate of the coal-to-gas substitution, different behaviours will depend on the model, as outlined below: •In 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅, where coal declines in the Ref., favouring natural gas, as the implementation of power plant emission standards is implicit in the calibration, a domestic carbon price would rather favour a switch from natural gas to renewables in the power sector (Figs. D.30 and D.32). •In 𝑀 𝑈 𝑆 𝐸where no constraints are set on coal, this fossil fuel tends to increase compared to other sources (Fig. D.14) in the Ref. scenario. In presence of a carbon price locally, natural gas increases (and oil, too) compared to the reference (Ctax.WORLD, Ctax.NA), as this promotes a coal-to gas and a coal-to-oil substitution (Fig. D.28). •In 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿presents a higher share of coal in the Ref. scenario, because of the policy implementation. Coal does show a more rapid phase-out when a carbon price applies everywhere (Ctax.WORLD). Coal is then substituted with oil and natural gas; renewables gain more share after 2040 when subjected to a sufficiently high costs reduction compared to carbon price value. 6.2. Fossil trade Fossil trade is shown for imports of oil products in the main body of the work for the reference and the sensitivity cases are presented. Additional diagrams are reported in Appendix Dwhere Tables D.5 and D.6 provide an estimate of the relative variation of the total imports and exports from NA in each scenario compared to the Ref. scenario. Natural gas. In 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅, the world demand of natural gas reduces worldwide when a carbon price is applied (Ctax.WORLD), resulting in an overall contraction of supply and exports. Exports decreases the least when only Canada and Mexico implement the carbon price (Ctax.CANMEX), as the demand for natural gas falls in these regions, the supply of natural gas is exported (Figs. D.37,D.43, and D.47). On the contrary, natural gas imports exhibit the greatest increase when only the U.S. 7 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Fig. 2. Natural gas consumption in Mexico. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. 3. Natural gas consumption in the U.S.. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. 4. Imports of oil products in Canada from 2020 through to 2050: reference and sensitivity scenarios. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. refrains from imposing a carbon price (Ctax.WORLDxUS) as the local gas consumption increases. Imports decrease the most when only NA imposes a carbon price as it relies less on natural gas for domestic consumption (Fig. D.46). 𝑀 𝑈 𝑆 𝐸increases natural gas exports as Canada and Mexico exports more than the Ref. scenario to the U.S. when a carbon price is applied (Ctax.WORLD and Ctax.NA). This triggers a switch, especially in the power sector from coal to natural gas. Conversely, natural gas imports 8 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Fig. 5. Imports of oil products in Mexico from 2020 through to 2050: reference and sensitivity scenarios. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. 6. Imports of oil products in U.S. from 2020 through to 2050: reference and sensitivity scenarios. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. decrease both in Canada and in Mexico, as a carbon price applied would translate mainly to a higher share of renewables and natural gas in the power sector. U.S. displays a rather opposite behaviour as both oil and natural gas are used more than renewables for substituting coal, and imports are enhanced (Figs. 6and D.46). A large share of the U.S. imports is expected to be supplied by Canada. In 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿, natural gas exports reduce in Canada, U.S., and Mexico when a carbon price is applied globally (Ctax.WORLD). In the case of an asymmetric carbon price, a reduction in the natural gas demand in the countries outside NA drives a reduction in the exports (Ctax.WORLDxUS). U.S. also increases its exports if a price applies exclusively to NA or to the cluster Canada-Mexico, driven by cheaper natural gas availability (Ctax.NA and Ctax.CAN-MEX). Imports tend to increase if all three countries enforce a carbon price, primarily due to a coal-to-gas substitution in the power sector (Ctax.WORLDxUS); for the U.S. the trend is also observed in the Ctax.CAN-MEX scenario. Crude oil. In 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅, results show that crude oil exports from NA to the rest of the world decrease under the four carbon pricing scenarios, with larger decreases when regions outside of NA put a price on carbon emissions (Ctax.WORLD). The result is driven by the share of Canadian crude oil exports (Fig. D.39. NA crude oil imports increase in the scenarios except for when only Canada-Mexico and the U.S. impose a carbon price. This result is driven by an increase in imports by Canada from regions outside of NA (mostly OPEC) as domestic crude oil production falls more than demand. In 𝑀 𝑈 𝑆 𝐸, the increase in crude oil imports in the Ctax.CANMEX and the Ctax.WORLDxUS scenarios, occurs as U.S. increases the demand for oil especially in power generation (Fig. D.28). In Canada, the increase in imports as increasing demand for oil in industrial cogeneration, in the building sector also in addition to support local supply. There is weak crude oil import demand in Mexico, to sustain the demand of local refinery capacity. Crude oil imports are stable in the U.S. sustaining the demand of local refineries. If a carbon price applies, there is no deviation from the Ref. scenario in Canada. The U.S. tends to increase the import of cheaper crude oil when a carbon price applies in Canada and Mexico only (Ctax.CAN-MEX). Mexico tends to import cheaper crude oil (especially when a carbon price applies in Mexico and Canada only, Ctax.CAN-MEX) as the level of carbon price does not change the demand of oil products in transportation, but rather promotes imports to compensate local demand gaps. In terms of crude oil exports, NA still maintains the base year level of intratrade and develops further commercial trade with Europe for example. In presence of a carbon price, there is minimal variation from the Ref. scenario in the three countries, due to the nearly stable demand of oil products of refineries outside the NA. In a case of asymmetric carbon price (Ctax.WorldxUS), cheaper crude oil for NA are imported, to sustain the demand of local refineries as the level of carbon price is not sufficiently high to create a shift in the transportation sector. In 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿, crude oil imports increase in Canada and Mexico when a carbon price applies, especially as the power sector switches 9 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Nova scotia. Nova Scotia implemented a cap-and-trade programme as of January 1, 2019. Companies will be given allowances to emit carbon dioxide-equivalents to a certain level, and will be able to buy and sell credits to pollute above that level. Each year, the total number of allowances in the province will decrease to help push Nova Scotia towards its overall emissions reduction goals. Nunavut. Nunavut is supportive of the federal government applying its own carbon pricing framework in its territory. Large emitters and most fossil fuels will be subject to carbon prices as of July 2019, with exemptions for aviation fuels and diesel-powered electricity generation. Ontario. Ontario cancelled its cap-and-trade system and overall climate change Program in 2018. Ontario will be covered by a provincial OBPS for industrial facilities emitting more than 50,000 tonnes of carbon dioxide-equivalent per year will be priced starting January 2019. Companies emitting between 10,000 and 50,000 tonnes will be able to opt into the system as well. The Federal Carbon Levy will be applied to consumers. Prince edward island. Large industrial emitters in Prince Edward Island will be subject to the Federal OBPS for industry emitting more than 50,000 tonnes of CO2e per year, with facilities emitting between 10,000 and 50,000 tonnes of CO2𝑒𝑞 per year allowed to opt in. The province has imposed their own carbon levy on consumer fuels, exempted furnace oil and propane. Quebec. Quebec’s cap-and-trade system came into effect in 2013. The cap-and-trade system, called the Western Climate Imitative (WCI), is linked with California and with a linkage to Nova Scotia happening shortly. The province aims to have its total emissions 20% below where they were in 1990 by 2020, and 37.5% below the 1990 level by 2030. Industrial, electricity and fossil fuel companies which emit more than 25,000 tonnes of carbon dioxide-equivalent chemicals must take part in the cap-and-trade market. Companies see slight annual reductions in their cap-and-trade allowances, which pushes the province as a whole towards its emissions reduction targets. Revenues from the system are put towards various measures to help the province reduce carbon emissions and adapt to climate change. Sasatchewan. Saskatchewan unveiled a system which partially meets the federal standard. Under Saskatchewan’s plan, large industrial facilities which emit at least 25,000 tonnes of carbon dioxide-equivalent gases per year will have to reduce their emissions by 10% by 2030. This will cover approximately 1.1 per cent of the province’s total greenhouse gas emissions. The federal government announced that it will impose the Federal OBPS for electricity generation and pipelines as they are exempt under the Saskatchewan system. The Canadian Federal government also imposed the Federal Carbon Levy on consumers. Yukon. The Yukon government has asked the federal government to apply the same plan in its territory as is being imposed on provinces that have not developed their own plans. There are some slight differences in how the plan is applied in Yukon. It will not take effect until July 2019, and will not include a carbon-related price on commercial fishing and agriculture. There will also be fuel charge relief for aviation fuel, Diesel electricity generation in remote communities and partial relief for greenhouse operators. B.3. Overview of Mexico legislation on ETS Mexico has enforced a target for a 30% reduction in GHG emissions below business as usual by 2020, and a 50% reduction below 2000 levels by 2050 through the General Law on Climate Change Law (GLCC), issued in 2012. After signing the Paris Agreement in April 2016, the country has committed to a non-conditional target of 22% GHG emission reduction, compared to a reference case without climate policies, and to increase this to 36% subject to financial support from the international community. These targets were included in the reform to the General Law of Climate Change on July 13th, 2018. The GLCC and the further amendments in 2017 enable the Ministry of Environment to establish a mandatory ETS to start after an initial 36-month Pilot Programme. The pilot programme has begun on October 1st, 2019 with the announcement of the Preliminary Guidelines by the Mexican Ministry of the Environment and Natural Resources (SEMARNAT). The scheme imposed a National Emissions Registry in 2014, which is comprised of an emissions registry for both direct and indirect GHG emissions for facilities with annual emissions above 25,000 t of CO2(IETA,2019). In this pilot programme, CO2will be the only greenhouse gas included and it will apply only to facilities which annual emissions are equal or larger than 100,000 t of CO2from the industrial and energy (oil and gas, electricity generation) sectors, which represent 90% of the emissions reported to the National Emissions Registry. Although there will not be economic sanctions during this pilot programme if participants do not comply, it could influence the behaviour of stakeholders within the industrial and energy sectors, getting them to fully integrate emissions reduction in their decision making processes. Following this pilot programme, the ETS will be officially implemented in 2022. In 2013, a carbon tax of 3.5 USD/ t of CO2was introduced on fossil fuel production, excluding natural gas. In addition, in the same year, a voluntary carbon exchange (MEXICO2) was also established to trade carbon credits as a potential means to comply with the tax. Some companies in Mexico, including two of the largest national airlines, Aeromexico and Volaris, have gained valuable experience through the voluntary carbon market. A system of certified emission reductions was also introduced for all the sectors primarily promoted through the United Nations Clean Development Mechanisms (ICAP,2019). B.4. Overview of U.S. legislation on ETS Carbon mitigation initiatives at a state level include the Regional Greenhouse Initiative and the California cap-and-trade system (The New York Times,2019). The Regional Greenhouse Gas Initiative is a cap-and-trade system based on auctions supplying carbon emission permits to power plants in 9 Northeastern states. Although the scheme has been characterised by a fairly low carbon price (5 USD/t of CO2) to date and a low coverage (up to 18% of the emissions), the states have used the money raised by the auctions to invest in efficiency and clean energy programmes. More states are considering to join the Regional Greenhouse Gas Initiative, such as Virginia and New Jersey. California has a cap-and-trade programme for power plants, manufacturers, and refineries. Carbon prices under the system are higher then in the Northeastern states (up to 15 USD/t of CO2) and the share of covered emissions is up to 85% of the total. Appendix C. Literature on asymmetric policies This literature review focuses on the geographical coverage of the approaches presented. C.1. European union perspectives A large part of the literature focuses on the EU ETS with concerns on carbon leakage especially after the grandfathering mechanism ceased (European Commission,2018;Clò,2010). Conversely, evaluations of energy efficiency programmes in the EU demonstrate how effective policies can enhance sectoral energy savings and mitigate carbon leakage risks through collaboration (Johansson et al.,2022). Regression analyses on EU imports between 2004 and 2011 indicated modest variations in embodied carbon due to emission permit values being lower than transportation costs (Naegele and Zaklan, 16 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 2019). The EMF led two initiatives, EMF-28 and EMF-29, to project the implications of European climate policies on trade. De Cian et al. (2013) examined interactions between European climate policy and the international climate regime using various models. The EMF-29 study highlighted a higher carbon leakage risk for energy-intensive industries, suggesting that border carbon adjustments could reduce leakage (Böhringer et al.,2012a). Paroussos et al. (2015) used the GEME3 model to quantify carbon leakage due to fragmented global climate policies, while Arroyo-Currás et al. (2015) evaluated the effectiveness of unilateral and joint mitigation efforts using the REMIND model, identifying causes of carbon leakage in price changes and inter-fuel substitution. These studies highlight the concerns over asymmetric embodied carbon of traded goods and the potential for carbon leakage due to stringent regulations. This context underscores the importance of effective energy policies, which is further explored in Xue et al. (2022) focusing on economic policy uncertainty (EPU) and clean energy consumption in France from 1987 to 2019. Utilising the STIRPAT framework and the Augmented ARDL method, the study reveals that while clean energy consumption does not significantly reduce emissions in the long run, EPU and economic growth contribute to increased CO2emissions. In contrast, urbanisation positively affects environmental quality. C.2. Global south perspectives Literature has also assessed the role of the Global South, particularly China’s trade as a major exporter. Guo et al. (2010) used an empirical input–output model from the Chinese Input–Output Association to analyse Sino-US trade, concluding that introducing less carbon-intensive technologies in China’s manufacturing would aid global emission targets. Several studies on bilateral trade between China and Australia2 and China and Japan3decomposed time-series of traded goods to identify changes in embodied carbon emissions at the sector level. Using a CGE model, Dong et al. (2015) examined border carbon adjustments and export taxes on Chinese trade, suggesting these measures would be weak instruments for reducing domestic CO2emissions while preserving competitiveness in unilaterally regulated countries. In many Global South nations, such as India and China, coal-fired power plants still play a central role in energy generation, and are considered fundamental for economic development in the medium term. Nevertheless, substantial amounts of both air and groundwater pollution is produced. For instance, recent research has proposed using fly ash as a landfill geoliner to help mitigate groundwater contamination, offering a potential solution to two major environmental challenges at once (Chowdhury et al.,2022). This reflects the broader asymmetries in environmental policies, where developing countries often face the dual burden of managing both industrial waste and environmental degradation with less regulatory support compared to the Global North. Recently, Li et al. (2022) analysed the impact of economic policy uncertainties (EPU) on China’s carbon emissions trading (CET) market using a nonlinear ARDL model from 2013 to 2021. They found that trade policy uncertainty (TPU) and monetary policy uncertainty (MPU) positively affect CET prices, while exchange rate policy uncertainty (ECPU) has a negative influence, highlighting the need for stable policy environments for effective emission reductions. Additionally, Ma et al. (2023) assessed China’s carbon neutrality pathways using the Global Change Analysis Model (GCAM), revealing trade-offs between the pace of carbon neutrality and reliance on carbon dioxide removal (CDR) technologies. The most aggressive scenario (‘S00’) reduces mitigation 2Jayanthakumaran and Liu (2016) linked input–output tables of Australia (Australian National Accounts,2013) and China (National Bureau of Statistics of China,2012). 3Long et al. (2018) based their analyses on world input–output tables (NWO,2016). costs by 1.04% of GDP by 2100 and decreases reliance on biomass energy with carbon capture and storage (BECCS) by 36%. Furthermore, Lin et al. (2024) found that although green finance boosts the green total factor productivity of China’s energy-intensive industry, regional disparities necessitate tailored policy approaches to maximise green development benefits, while Shang et al. (2024) emphasised the importance of comprehensive policy frameworks that address regional disparities in the promotion of electric vehicles, advocating for coordinated efforts to ensure equitable access to resources and incentives. Finally, Wu et al. (2023) underscored the urgent need for policies that not only optimise energy structures but also promote sustainable industrial practices, reinforcing the interconnectedness of effective policy frameworks across various sectors in driving China towards its environmental goals. These insights into China’s energy transition highlight the vital role of renewable energy in limiting global warming to 1.5 ◦C while promoting socio-economic development in the Global South. In this sense, Vanegas Cantarero (2020) stressed the need for advancing technology, societal engagement, and supportive policies for successful energy transitions. By advocating for commercially available technologies and citizen participation in energy planning, they propose a roadmap for China and other developing countries to achieve sustainable development. Similarly, Mungai et al. (2022) noted that unlocking climate finance can enhance renewable energy and energy efficiency initiatives, addressing investment barriers and fostering sustainable growth. The aforementioned outputs, which are mainly derived from energy systems modelling exercises, emphasise the importance of energy equity. For instance, Chowdhury and Naz (2023) studied the impacts of implementing large-scale energy projects in the Indian Sundarbans, where energy poverty, carbon emissions, and health risks, were assessed. Findings suggest improvements for 1220 families receiving novel and efficient technology, which besides improvements in energy justice, also includes reduction on PM2.5 exposure while CO2 emissions decreased by about 803 kg per person annually. C.3. Other regional perspectives Other studies addressed the role of asymmetric policies for macroregions. Deng and Xu (2017) proposed a multi-regional input–output model based on the World Input–Output Tables (NWO,2016) to calculate carbon emissions in global trade. They applied a structural decomposition analysis on China, Japan, India, and the United States, revealing contributions to embodied carbon from imports, exports, and self-consumption. Similarly, He and Hertwich (2019) used EXIOBASE 3.3 (EXIOBASE Consortium,2018) to assess the flow of embodied carbon through 49 countries, focusing on direct and indirect emissions in China, the U.S., and Europe, and demonstrating that emission intensities depend on the share of imports and exports. Multi-sector, multi-regional general equilibrium models based on the GTAP Data Base (Naegele and Zaklan,2019) facilitate qualitative and quantitative assessments of climate policy-induced economy-wide changes (Rutherford and Paltsev,2000). Caron (2012) developed a micro-consistent dataset expanding GTAP’s industrial coverage from 16 to 51 sectors, concluding that industrial trade responses to carbon pricing were higher in models with aggregated sectors, with leakage rates smaller than in partial equilibrium studies. Among energy systems modelling contributions, Böhringer et al. (2012b) assessed import tariffs on embodied carbon as a supplement to unilateral emissions pricing using a multiregion, multi-sector CGE model, arguing that efficient policies should consider indirect electricity emissions. Appendix D. Additional data See Figs. D.12,D.13,D.15,D.18,D.19,D.21,D.27,D.29,D.31,D.36, D.40–D.42,D.44 and D.45 and Table D.4. 17 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Fig. D.12. Coal consumption in Canada. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.13. Coal consumption in Mexico. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.14. Coal consumption in U.S.. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. 18 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Fig. D.15. Oil product consumption in Canada. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.16. Oil product consumption in Mexico. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.17. Oil product consumption in U.S.. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. 19 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Fig. D.18. Electricity generation by oil in Canada. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.19. Electricity generation by coal in Canada. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.20. Electricity generation by natural gas in Canada. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. 20 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Fig. D.21. Electricity generation by hydropower in Canada. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.22. Electricity generation by renewables in Canada. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.23. Electricity generation by oil in Mexico. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. 21 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Fig. D.24. Electricity generation by coal in Mexico. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.25. Electricity generation by natural gas in Mexico. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.26. Electricity generation by hydropower in Mexico. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. 22 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Fig. D.27. Electricity generation by renewables in Mexico. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.28. Electricity generation by oil in U.S.. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.29. Electricity generation by coal in U.S.. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. 23 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Fig. D.30. Electricity generation by natural gas in U.S.. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.31. Electricity generation by hydropower in U.S.. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.32. Electricity generation by renewables in U.S.. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. 24 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Fig. D.33. Exports of oil products in Canada from 2020 through to 2050: reference and sensitivity scenarios. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.34. Exports of oil products in Mexico from 2020 through to 2050: reference and sensitivity scenarios. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. Fig. D.35. Imports of oil products in U.S. from 2020 through to 2050: reference and sensitivity scenarios. Left: 𝐸 𝐶−𝑀 𝑆 𝑀 𝑅; Centre: 𝑀 𝑈 𝑆 𝐸; Right: 𝑇 𝐼 𝐴𝑀 −𝑈 𝐶 𝐿. 25 S. Giarola, I. García Kerdan, P. Johnston et al. Resources, Environment and Sustainability 18 (2024) 100179 Pye, S., Butnar, I., Cronin, J., Welsby, D., Price, J., Dessens, O., Rodríguez, B.S., Winning, M., Anandarajah, G., Scamman, D., Keppo, I., 2020. The TIAM-UCL model (version 4.1.1) documentation. URL https://www.ucl.ac.uk/energy-models/ sites/energy-models/files/tiam-ucl-manual.pdf. Riahi, K., van Vuuren, D.P., Kriegler, E., Edmonds, J., O’Neill, B.C., Fujimori, S., Bauer, N., Calvin, K., Dellink, R., Fricko, O., Lutz, W., Popp, A., Cuaresma, J.C., KC, S., Leimbach, M., Jiang, L., Kram, T., Rao, S., Emmerling, J., Ebi, K., Hasegawa, T., Havlik, P., Humpenöder, F., Silva, L.A.D., Smith, S., Stehfest, E., Bosetti, V., Eom, J., Gernaat, D., Masui, T., Rogelj, J., Strefler, J., Drouet, L., Krey, V., Luderer, G., Harmsen, M., Takahashi, K., Baumstark, L., Doelman, J.C., Kainuma, M., Klimont, Z., Marangoni, G., Lotze-Campen, H., Obersteiner, M., Tabeau, A., Tavoni, M., 2017. The shared socioeconomic pathways and their energy, land use, and greenhouse gas emissions implications: An overview. Global Environ. Change 42, 153–168. Rogelj, J., Luderer, G., Pietzcker, R.C., Kriegler, E., Schaeffer, M., Krey, V., Riahi, K., 2015. Energy system transformations for limiting end-of-century warming to below 1.5 ◦C. Nature Clim. Change 5 (6), 519–527. http://dx.doi.org/10.1038/ nclimate2572. Rutherford, T., Paltsev, S., 2000. GTAP-energy in GAMS: The dataset and static model. Working paper 00–2. GTAP Working Papers. University of Colorado, Boulder. Sachs, J., Hidayat, S., Giarola, S., Hawkes, A., 2018. The role of CCS and biomassbased processes in the refinery sector for different carbon scenarios. In: Friedl, A., Klemeš, J.J., Radl, S., Varbanov, P.S., Wallek, T. (Eds.), 28th European Symposium on Computer Aided Process Engineering. In: Computer Aided Chemical Engineering, vol. 43, Elsevier, pp. 1365–1370. http://dx.doi.org/10.1016/B978-0-444-642356.50239-4. Sachs, J., Meng, Y., Giarola, S., Hawkes, A., 2019. An agent-based model for energy investment decisions in the residential sector. Energy 172, 752–768. http://dx.doi. org/10.1016/j.energy.2019.01.161. Sarker, P.K., Bouri, E., Marco, C.K.L., 2023. Asymmetric effects of climate policy uncertainty, geopolitical risk, and crude oil prices on clean energy prices. Environ. Sci. Pollut. Res. 30 (6), 15797–15807. http://dx.doi.org/10.1007/s11356-02223020-w. Schaeffer, R., de Sá, A., 1996. The embodiment of carbon associated with Brazilian imports and exports. Energy Convers. Manage. 37 (6), 955–960. http://dx.doi.org/ 10.1016/0196-8904(95)00283-9, Proceedings of the International Energy Agency Greenhouse Gases: Mitigation Options Conference". SEMARNAT, 2021. Ley de transición energética. URL https://www.gob.mx/semarnat. Shang, H., Sun, Y., Huang, D., Meng, F., 2024. Life cycle assessment of atmospheric environmental impact on the large-scale promotion of electric vehicles in China. Resour. Environ. Sustain. 15, 100148. http://dx.doi.org/10. 1016/j.resenv.2024.100148, URL https://www.sciencedirect.com/science/article/ pii/S266691612400001X. Siddiqui, S., Vaillancourt, K., Bahn, O., Victor, N., Nichols, C., Avraam, C., Brown, M., 2020. Integrated North American energy markets under different futures of cross-border energy infrastructure. Energy Policy 144, 111658. http: //dx.doi.org/10.1016/j.enpol.2020.111658, URL https://www.sciencedirect.com/ science/article/pii/S0301421520303888. The New York Times, 2019. These Countries Have Prices on Carbon. Are They Working?. Technical Report, The New York Times, URL https://www.nytimes.com. Vanegas Cantarero, M.M., 2020. Of renewable energy, energy democracy, and sustainable development: A roadmap to accelerate the energy transition in developing countries. Energy Res. Soc. Sci. 70, 101716. http://dx.doi.org/10. 1016/j.erss.2020.101716, URL https://www.sciencedirect.com/science/article/pii/ S2214629620302917. Wang, Q., Zhou, Y., 2019. Uncovering embodied CO2 flows via north-north trade – A case study of US-Germany trade. Sci. Total Environ. 691, 943–959. Weyant, J., 2012. Introduction to the EMF 29 special issue on the role of border carbon adjustment in unilateral climate policy. Energy Econ. 34, S95–S96. http: //dx.doi.org/10.1016/J.ENECO.2012.10.002. World Bank, 2024a. State and trends of carbon pricing 2024. Retrieved from https://www.worldbank.org/en/news/feature/2024/05/29/state-and-trendsof-carbon-pricing-2024. World Bank, 2024b. World bank carbon pricing dashboard. Retrieved from https: //carbonpricingdashboard.worldbank.org/. Wu, H., Wang, Q., Xu, Y., Ye, Y., Zeng, X., 2023. Coal life-cycle analysis embedded with land–energy nexus of a coal-based city in China. Resour. Environ. Sustain. 12, 100109. http://dx.doi.org/10.1016/j.resenv.2023.100109, URL https://www. sciencedirect.com/science/article/pii/S2666916123000026. Xue, C., Shahbaz, M., Ahmed, Z., Ahmad, M., Sinha, A., 2022. Clean energy consumption, economic growth, and environmental sustainability: What is the role of economic policy uncertainty? Renew. Energy 184, 899–907. http://dx.doi. org/10.1016/j.renene.2021.12.006, URL https://www.sciencedirect.com/science/ article/pii/S0960148121017250. Zhu, Y., Ghosh, M., Luo, D., Macaluso, N., Rattray, J., 2018. Revenue recycling and cost effective GHG abatement: An exploratory analysis using a global multi-sector multi-region CGE model. Clim. Chang. Econ. 9 (01), 1840009. 32