On the Fair Accounting of Carbon Emissions in the Global System Using an Exergy Cost Formation Concept PDF 2021

Journal of Cleaner Production 280 (2021) 124438 Contents lists available at ScienceDirect Journal of Cleaner Production journal h...

Journal of Cleaner Production 280 (2021) 124438 Contents lists available at ScienceDirect Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro On the fair accounting of carbon emissions in the global system using an exergy cost formation concept Hossein Khajehpour a, b, *, Yadollah Saboohi a, George Tsatsaronis b a Energy Engineering Department, Sharif University of Technology, Tehran, Iran b Chair of Energy Engineering and Environmental Protection, Technical University of Berlin, Marchstr. 18, D-10587, Berlin, Germany a r t i c l e i n f o a b s t r a c t Article history: Carbon accounting is necessary for designing effective climate change mitigation policies. A proper and Received 17 March 2020 fair accounting method should motivate both the producers toward cleaner production methods and the Received in revised form consumers toward reducing the embodied emissions of their consumption. This research work proposes 25 August 2020 a new approach to map the production chain of carbon emissions in which every subsystem is Accepted 25 September 2020 Available online 1 October 2020 responsible for the level and efficiency of its production activities and embodied emissions for providing its economic activities or final demands. The exergy cost formation concept is used to track the emissions Handling Editor: Zhifu Mi in the production chain. The results of this accounting present the total carbon loads on economic outputs either consumed locally or exported abroad (CExA). The CExA results are then compared to the Keywords: results of conventional production-based (PBA) and consumption-based (CBA) carbon inventories. Here Climate change we show that, in addition to the levels of production and consumption, the economic structures of the Carbon accounting countries and the efficiency of the production activities are important factors differentiating the roles of Shared responsibility the countries in the global emissions. Our results show that the share of the imported emissions to the Carbon emission added total CExA varies between 14% for developing countries to 34% for the developed countries. Moreover, Exergoenvironmental analysis although the ratio of CBA to PBA for the countries is highly dependent on their economic states (0.87 for Exergy cost developing countries and 1.21 for developed countries), the ratio of CExA to PBA does not follow a unique trend among developing or developed countries. The results demonstrate that, according to the proposed sharing approach, the import-oriented developed countries, which have benefited the most from the carbon leakage effect, are mostly penalized for the embodied emissions associated with the imports to their economy, and vice versa. © 2020 Elsevier Ltd. All rights reserved. 1. Introduction countries in causing the problem. Therefore, climate equality has been the subject of numerous research works. It is widely accepted Climate change has been one of the most controversial and that in a fair allocation of the global responsibilities to the climate common environmental concerns about the quality of life of human change, factors such as the past and current shares of greenhouse beings. The problem is huge in extent and requires an integrated gases (GHG) emissions should be considered. Moreover, one main action plan involving the participation of all nations in accordance issue refers to the differences in the assigned responsibilities with “the common but differentiated responsibilities and respec- resulting from different carbon accounting approaches. tive capabilities” (UNFCCC, 1992). Sharing responsibilities has been The allocation of the responsibilities must be proportional to the one of the most debated issues of acting against climate change in shares in the global GHG emissions and the corresponding global global negotiations. This is due to the fact that there is no agree- resource consumptions. There are several approaches applicable to ment on the fair perception of the contributions of different such accounting (Eder and Narodoslawsky, 1999). Conventional allocation approaches have been developed based on the direct emissions from a country territory (Territory-Based Accounting (TBA)), direct emissions resulted from value-added production of * Corresponding author. Room 304, Department of Energy Engineering, Taymouri the countries (Production-Based Accounting (PBA)), and the direct Alley, Ghasemi Street, Habibollahi St., Azadi Avenue, Tehran, PO Box: 11365-11155, (local) and indirect (embodied in trades) emissions to meet the Iran. E-mail address: [email protected] (H. Khajehpour). final demands of a nation (Consumption-Based Accounting (CBA)) https://doi.org/10.1016/j.jclepro.2020.124438 0959-6526/© 2020 Elsevier Ltd. All rights reserved. H. Khajehpour, Y. Saboohi and G. Tsatsaronis Journal of Cleaner Production 280 (2021) 124438 Abbreviations PBA Production-Based Accounting RoW Rest of the Wolrd BRIICS (BRIIS) Major developing countries -Brazil, Russia, India, SBA Shared-Based Accounting Indonesia, South Africa-including (excluding) TBA Territory-Based Accounting China CBA Consumption-Based Accounting Nomenclature CEA Carbon Emission Added CEA* Cumulated Carbon Emission Added (t CO2e/yr) CExA Carbon Emission Added based on the exergy cost E Total embodied emission associated with a stream (t accounting CO2e/yr) FCBA Full Consumption Based Accounting E Cumulated carbon intensity (g CO2e/MJ) G7 Group of 7 (most advanced economies): Canada, ExC Cumulative exergy destruction (MJ/$) France, Germany, Italy, Japan, the United Kingdom, S Economic Sector and the United States V Monetary value of product ($) GHG Greenhouse Gases Y Total direct emissions as in PBA (t CO2e/yr) ICIO Inter-Country Input-Output tables ISIC International Standard Industrial Classification Subscripts MRIO Multiregional Input-Output tables i Index referring to an exporting country OECD Organization for Economic Co-operation and j Index referring to an importing county Development k, l Index referring to an economic sector (product) (Peters, 2008). The later, also known as the carbon footprint, is by sharing carbon loads in trades between consumers and pro- introduced as the emissions embodied in the final demands (Peters ducers (SBA). In contrast to the CBA, the SBA methods do not and Hertwich, 2008). Hence, this is equal to the local emissions exclude the embodied emissions of the exported materials and (PBA) plus the emissions embodied on the imported final products services from the producer responsibilities. This is done through minus the embodied emissions of exports (Wiebe and Yamano, modifications in the CBA by either sharing the embodied emissions 2016). These approaches have different levels of complexities in of exports among the producer and consumer based on the relative calculations, data requirements, and uncertainties. Moreover, when added values for the producers and consumers, e.g., by Lenzen et al. applied to the question of allocation of the responsibilities, they are (Lenzen et al., 2007; Marques et al., 2012; Rodrigues and Domingos, biased toward benefits of the producers or consumers. In the PBA 2008) or by assigning the share of embodied emissions resulting (as well as the TBA), the producers are the sole responsible body for from inefficient production operations to the producer, e.g., GHG emissions, while in the CBA, all the emission loads are (Kander et al., 2015). assigned to the final consumers (Davis and Caldeira, 2010). There- This paper aims at developing and implementing a new SBA fore, although the CBA can complement the functionality of PBA for method. There uncertainties in sharing the added values from raising awareness of the societies about the impacts of their actions trades among producers and consumers (Kander et al., 2015). Also, and consumptions, yet it is not clear how to combine the two in- there is possibility of failure in considering the economic incentives dicators for effective policy implications (Afionis et al., 2017). for climate policy implications (Jakob and Marschinski, 2013). In a fair allocation of responsibilities, the countries should bear Therefore, the proposed method is based on the notion that the the whole environmental burden resulting from all the supplies desirability of trade is identical for both the supply and demand and demands necessary for their economic growth (Lenzen et al., sides. As a result, both the producer and the consumer are equally 2007). In this case, every nation is responsible for the climate ex- responsible for the embodied emissions of a traded stream. A ternalities of the desired increase of its welfare. This environmental detailed discussion of the implemented methods is presented in burden is caused by either the direct emission of the pollutants the next section of this paper. from the production activities or indirectly by the import of the Assigning responsibilities can facilitate the identification of fair products or services which had caused upstream emissions abroad. contributions of the countries in financing global action plans. Ignorance of the responsibilities from the consumption of the im- However, the implications in policy implementation techniques ported products has lad to the shift of the carbon-intensive pro- have been widely discussed in the literature and are out of the ductions from developed countries (with restricted climate scope of the present work. commitments) to developing countries (Rothman, 1998). This ef- fect, also called “carbon leakage”, has continuously increased since 2. Methodology 1990 (Peters et al., 2011). Many researchers (e.g., Wiebe and Yamano (2016)) have shown the extent of carbon leakage among In any sharing carbon accounting approach, the way of assigning different categories of the countries. The leaked carbon could be as the carbon loads on the traded streams differentiates the SBA ap- high as 26% of total global emissions (Peters et al., 2011). This effect proaches from one another and from the CBA and PBA inventories. has been understood as a result of the PBA-based climate change Two main factors affect the role of individual nations in emitting policies in the past decades. The balance of the embodied carbon on greenhouse gases: the technical efficiencies of productions and the trades depends on the development stage and the corresponding structures of the economies. The technical efficiencies effect is economic structures of the countries (Davis and Caldeira, 2010). essential in accounting the responsibilities of inefficient pro- Therefore, a proper accounting method should consider both the ductions (which is already included in the PBA but is missing in the production and consumption responsibilities of the local emissions CBA). On the other hand, the economic structure is important as it and the emission loads on the imported materials and services. takes the effect of the extent of the imports and exports of an Accordingly, another group of accounting methods has been economic system into account. This is already included in the CBA developed to reduce the inefficiencies of the PBA and CBA methods but is missing in the PBA. A proper SBA approach should adequately 2 H. Khajehpour, Y. Saboohi and G. Tsatsaronis Journal of Cleaner Production 280 (2021) 124438 consider both factors simultaneously. This could be done by either a According to Equation (1), in the proposed CEA approach, the technical adjustment made to the CBA or by structural consider- carbon load of any trade stream (Eij:k ) is assigned equivalently ations added to the PBA accounting. among the consumers of the final and middle products at different Apart from the SBA methods distributing the emission burdens levels of the supply chain. In other words, in contrast to CBA, in the in proportion to the economic profitability of the trades for the CEA, the desirability of the emissions and corresponding re- producers and consumers, a group of the developed SBA methods sponsibilities are assigned for both the demand and supply sides. has been proposed based on technical adjustments made to the This is because of the fact that in the CEA, the emission loads on the CBA approach (e.g., Technology-adjusted CBA in (Kander et al., exports are not excluded from the responsibilities of the producers. 2015)). A brief introduction to the developed SBA methods and Also, the FCBA is a portion of the CEA, which is only attributed to their limitations in implementing them to climate policymaking the local consumption in the final demands, excluding the carbon are presented in Supplementary Material 1. loads on the exports. A further graphical description of how the CEA In the first proposal of SBA, Bastianoni et al. have proposed a differs from the CBA is presented in Supplementary Material 3. Carbon Emission Added (CEA) approach for sharing the re- The CEA approach was proposed for a symbolic unidirectional sponsibilities among the producers and consumers (Bastianoni supply chain in which only one product is produced and consumed et al., 2004). The method, based on the Embodied Energy Anal- among three components (Bastianoni et al., 2004). However, the ysis method (Odum, 2007), recommends the accumulation of the idea has not been further developed nor applied to any regional or emission loads in the whole supply chain of production from the global case. This seems to be because of the fact that the application raw materials to the intermediate products and the final con- of the method requires the allocation of the emission loads on in- sumptions (Bastianoni et al., 2004). The cumulative emission loads termediate streams and local consumptions, which cannot become are then normalized to fulfill the Additivity criterion of a proper easily available for each product of each country in the global sys- allocation regime. Further details about the requirements for an tem. The main contribution of this paper is to overcome this appropriate carbon accounting method are presented in Supple- shortcoming by proposing the notion that the exergy cost may be a mentary Material 2. valid proxy for apportioning the embodied emissions among The CEA method has been conceptually criticized due to the different by-products of a subsystem. seemingly unfair allocation of the emission burdens to the end- The CEA requires no more data than the global trade data pro- users, even if they do not contribute to direct emissions, e.g., final vided in the global trade yearbooks or aggregated Multiregional distributors of a product in a supply chain (Csutora and Veto } ne  Input-Output (MRIO) tables. However, the application of the mo zner, 2014). However, when considering the global trade sys- method is restricted to the development of an accounting tech- tem, such a case does not exist in a circular system of inter- nique for tracking the carbon load creation on both the interme- connected economies in which no part of the system is a sole diate and final demands as well as the trade streams. The proposed producer or consumer. Even in the case that such a subsystem method in this research work is to use the exergy cost accounting exists in the system under study, it usually plays a major role in the formulation for allocating the total CEA of the country among final promotion of the consumption, and therefore is eligible for consumptions and exported streams of different products of a assuming responsibility (Lenzen, 2008). Moreover, the CEA has country. Hereafter, the proposed method is abbreviated as CExA. been criticized because of its sensitivity to the number of system In order to calculate the emission loads on the imported and components in the supply chain (Csutora and Veto } ne  mo  zner, exported streams, the emission balance equations (as in Equation 2014). This sensitivity is indeed problematic, while the choice of (1) are formulated (Bastianoni et al., 2004). According to the level of aggregation changes the number of subsystems (Bastianoni et al., 2004), this could be done through an analogous (Berzosa et al., 2014). However, while analyzing the global econ- approach to the Embodied Energy Costs in (Odum, 2007) or other omy, the number of subsystems (countries) is fixed. Therefore, thermodynamics-based methods for tracking the accumulation of there would be no such sensitivity in the definition of the system the exergy destruction (Szargut and Morris, 1987) thermos- and its components. ecological cost (Szargut and Stanek, 2010), or exergoenvir- Based on the method proposed in (Bastianoni et al., 2004), the onmental costs within an energy system (Khajehpour et al., 2017). CEA of the i’th component of a (linear) system is equivalent to the As the fossil-fuel related emissions account for almost 84% of the normalized value of the summation of the direct emissions from global CO2 emissions (IPCC, 2014), only the energy-related CO2 the component (PBAi ) and the cumulative emission loads of the emissions have been considered in this study and the reference imported streams of type k from other system components j study for the CBA accounting presented in (Wiebe and Yamano, PP ( Eji:k ) (Equation (1)). 2016). Therefore, the approach of the extended exergoenvir- jsi k onmental method has been applied to the calculation of the emission loads in this research work. This idea suggests the use of XX cumulative exergy destructions, rather than economic added-value From the production point of view : CEA*i ¼ PBAi þ Eji:k production chain, as proxies for cumulative carbon emissions. The jsi k proportionality of the exergy destructions and embodied carbons XX on traded materials and services are discussed in the Supplementary From the consumption point of view : CEA*i ¼ FCBAi þ Eij:k Material 4. jsi k According to the exergoenvironmental formulation (Meyer et al., 2009), Equation (2) presents the main carbon balance equa- X.X  tion for each subsystem i (which are countries in the global system) CEAi ¼ CEA*i  PBAi CEA*i (1) having trades of goods and services from k different economic i i sectors with other j subsystems. Also, from the consumption point of view, the CEA is equivalent to the total carbon loads on the activities benefiting an economic XX XX Yi þ Eji:k ¼ Eij:k (2) entity, either from the consumption in the final demand or pro- jsi k j k duction sectors (FCBAi ) or by added-values from the export of the PP products and services k to destinations j ( Eij:k ). In Equation (2), Yi is the PBA of the subsystem i and Eij,k stands 3 H. Khajehpour, Y. Saboohi and G. Tsatsaronis Journal of Cleaner Production 280 (2021) 124438 for the total embodied emission of product k exported from i to j. 2016). The OECD developed the ICIO_2016 database. Accordingly, Eij,k is calculated according to Equation (3): the CExA emission inventories are calculated for 62 countries plus a 63rd subsystem representing the rest of the world. The latter ag- Eij:k ¼ eij:k  Vij:k (3) gregates the trade data for 177 less developed countries of the world, for which the individual MRIO tables are not available. Also, The variable eij,k, is the specific cumulated carbon intensities based on the third revision of the International Standard Industrial (embodied emission per unit value of the product) and Vij,k is the Classification of All Economic Activities (ISIC Rev.3) the economic amount (monetary or physical value) of product k produced in i and activities are categorized into 34 sectors. ICIO_2016 database allows consumed in j. The specific cumulated carbon intensities eij,k are tracking carbon inventories before and after the Kyoto protocol. identical for products of the same type k from the same producer i This is important as the Kyoto protocol and its mandates are un- for all of the j destinations. derstood as the main reasons for the outsource of emissive in- In case of the aggregation of some subsystems (e.g., the eco- dustries to the developing world, which has resulted in the carbon nomic sectors of a country), according to the extended exer- leakage effect (Aichele and Felbermayr, 2015). The comparison of goenvironmental accounting (Khajehpour et al., 2017), the total CBA, PBA, and CExA is especially helpful in tracking the temporal emission loads are to be distributed among different types of out- changes before and after the Kyoto protocol in developed and puts (k and l) in proportion to their cumulative exergetic costs developing countries. Considering the data availability from the (Equation (4)): ICIO_2016, the CExA of the countries is calculated from 1995.. through 2011. Eij:k ExCij:k ¼ Eij:l ExCij:l (4) 3.2. Comparison of CExA results with PBA and CBA In Equation (4), the ExCij,k is the cumulative exergy destruction of product k, which is traded from country i to country j. Combining Fig. 1 presents the CExA per capita of the countries for the year Equations (3) and (4) gives: 2011.. . ..  The global average value of the CExA per-capita is 3.4 t CO2 eij:k eij:l ¼ ExCij:k Vij,k ExCij:l Vij,l (5) (equivalent to those of CBA and PBA) while the average values of the developed and the developing countries are 10.8 and 2.9 t CO2, The values of the unit exergy costs of products produced in i respectively. Table 1 compares the carbon emission per capita in the ðExCij:k =Vij:k Þ are different for every economic activity k. However, developed and developing countries for different allocation due to lack of country-specific sectoral exergy cost accounting re- approaches. sults, the ratio of unit exergy cost of product k to unit exergy cost of According to Table 1, the averaged CBA and PBA per-capita are product l ðExCij:k =Vij:k Þ=ðExCij:l =Vij:l Þ are assumed to be the same for 11.1 and 9.8 t CO2 for the developed countries and 2.8 and 3.1 for the all subsystems i. For instance, the ratio of unit exergy cost of textile developing countries (calculated from the results of Wiebe and production (k) to unit exergy cost of cement production (l) are Yamano (2016)). The last column of Table 1 shows higher shares assumed to be identical in all of the modeled countries (i). These of the imported carbon loads for the developed economies than the values are calculated from national exergetic input-output analysis developing countries. This is a result of the carbon leakage phe- research by Rocco (2016), and the results are used in this modeling. nomenon. Also, Table 1 shows that, in an overall view, the CExA Assuming the validity of this proxy, the detailed import data of shares the responsibilities among producers (in the developing any sector of a country could be aggregated into an overall import countries) and the consumers (in the developed countries). This is profile of the country, no matter how it is distributed among the according to the expectation from an SBA. Further detailed numeric economic activities of the country. Although this aggregation may values of the calculated shared emission inventories of the CExA are introduce some uncertainties in the exergy cost accounting of the presented in tabular and scatter-plot forms in the Supplementary aggregated economic sectors, it reduces the amount of the required Material 6. Also, Fig. 2 shows the changes in the carbon inventories data by 97% and hence improves the applicability of the accounting according to the CExA as a percentage change from the PBA and method for the countries for which the detailed data of the MRIO CBA. are not available. This uncertainty in the use of the exergy cost In contrast to the CBA, the expectations referring to the SBA proxy is a manageable source of uncertainty and, therefore, could emission inventories are not systematically supporting the con- be reduced in further detailed accountings. ventional categorization of the countries into developed (OECD) The novelty of this research work is that a new approach for and developing (non-OECD) countries (Kander et al., 2015). This is accounting the emission loads on the materials and services is due to the fact that the countries within each one of these groups proposed, which is based on the proportionality of the cumulative differ from each other according to two important aspects of exergy costs and the embodied emissions. As a result, the CExA climate responsibility: their economic structure and the carbon responsibility allocation method may be applied for accounting the intensities of their economic activities. The economic structure is shared emission inventories in the global system. Considering the important because it refers to the relative extent of the demand for fairness of the CExA approach in accounting for the upstream and international trade in the entire economy. Also, the carbon in- downstream responsibilities in an equal manner, the sharing tensities of products incorporate the technical efficiencies of pro- method may better reflect the policy implications of the national ducers in carbon accounting of the traded streams and, therefore, inventories. should be accounted for in a fair climate inventory regime. 3. Results and discussion 3.3. Effect of the economic structures on CExA 3.1. Data source for CBA and CExA accounting According to the product portfolio of the economic sectors, the average carbon (direct and indirect) intensity of the countries is For the sake of comparability of the results with a reference CBA calculated and depicted in Fig. 3 below. inventory, the trade data according to the 2016 edition of the Inter- According to Fig. 3, on average, the direct emissions account for country Input-Output (ICIO_2016) is used (Wiebe and Yamano, 80.4% of the countries’ emissions while the rest are the emission 4 H. Khajehpour, Y. Saboohi and G. Tsatsaronis Journal of Cleaner Production 280 (2021) 124438 Fig. 1. World map of the distribution of the CExA per-capita of the countries in 2011. Table 1 Comparison of the CO2 emission per capita in 2011 according to different accounting approaches. CO2 emission per capita per year (tCO2/ca.yr) PBA CBA CExA Share of indirect emissions in total carbon load (%) Developed countries 9.8 11.1 10.8 38.8% Developing countries 3.1 2.8 2.9 28.9% Global average 3.4 3.4 3.4 19.6%a a This is for the global average among all 198 countries. The average share of the imported carbon to the total carbon load among the 62 studied countries is 34.4%. Fig. 2. Changes from the PBA (left) and the CBA (right diagram) to the CExA sharing approach of this study. Fig. 3. Total carbon intensities of the economies and the shares of the direct and indirect carbons for the modeled economies. loads on the imported products. This share is as high as 92% for as low as 50% for import-oriented countries (e.g., Singapore, major developing countries (e.g., China, India, and Russia) while it Austria, Belgium). 5 H. Khajehpour, Y. Saboohi and G. Tsatsaronis Journal of Cleaner Production 280 (2021) 124438 In order to better analyze the effect of sharing, the countries of the major developing countries, Brazil (because of its high share of the world are categorized based on the economic structure and bio-fuel production), India (as the least carbon-efficient economy emission intensity criteria. The developed economies (OECD) are (see Fig. 3)), and Saudi Arabia (because of its fossil-fuel-based categorized into three groups: USA (less importing), “other G700 economy) show different levels and compositions of the carbon countries (Canada, France, Germany, Italy, Japan, United Kingdom), intensities. Oil-reliant economies have very low shares of indirect and “the rest of the OECD” countries. The “other G700 group of carbon loads for their consumptions. Therefore, they experience countries are cleaner producers while importing more than the the same CExA-based carbon inventory relative to PBA and CBA. USA and less than the other OECD countries. In contrast, “the rest of This relatively low share of the imported carbon is because of the the OECD” group of countries are more net importers among the fact that the crude oil production industry, as the main motivator of developed economies. On the other hand, the less carbon-efficient the economy, is itself a very energy (and carbon) intensive industry countries in the developing world are categorized into China (major which leads to high PBA, while the main imports to these countries producer and exporter), “BRIIS” countries (Brazil, Russia, India, are associated with less carbon-intensive products imported to Indonesia, South Africa) and a group of the rest of the world cover the final demands and not for intermediate processing countries. The BRIIS group of countries are major developing productions. countries with higher carbon intensities than China and fewer Fig. 5 shows the values of the CExA and (the shares of indirect imports. emission loads) in comparison to the PBA and CBA accounting re- Also, to interpret the effect of the economic factors causing sults (calculated from the results presented in (Wiebe and Yamano, different carbon inventories, some facts and figures referring to the 2016)) for the mentioned 6 categories. economic and exergetic indicators of the modeled countries in this Furthermore, along with the changes in the production levels, research work are presented in Supplementary Material 5. Fig. 4 economic structures, and the production efficiencies, the accounted depicts the relative climate-economic characteristics of the CExA of different countries experience gradual changes in time. mentioned groups of countries. The economic structure of the Fig. 6 shows the temporal changes of the accounted inventories as countries (in terms of the ratio of imported (indirect) carbon to the well as the PBA and CBA of the above group of countries (from the direct emissions) and their production efficiency (in terms of their reference study (Wiebe and Yamano, 2016)). carbon intensities of national gross domestic products in kgCO2/ $2010) are analyzed for the 6 categories to help understand the 3.4. Temporal changes of the CExA behavior of CExA (Fig. 4). Based on the data presented in Figs. 3 and 4, the carbon intensity The relative behavior of the CBA and PBA inventories of the of the developed (developing) countries is normally lower (higher) countries has been intensively studied in the literature (Davis and than the global average of 0.22 kgCO2/$2010. More specific to the Caldeira, 2010; Davis et al., 2011; Wiebe and Yamano, 2016) and economic structure of the country groups in this study, the average is not further discussed here. The stable proportionality of the SBA carbon intensity of the other G7 has the smallest value in the chart to the PBA and CBA through time shows that there have been no and, after other OECD, the second-largest share of indirect to direct major structural shifts in the relative production and consumption emissions. This clearly shows the effect of carbon leakage for behavior of the countries. However, according to the changes in the outsourcing polluting industries from these countries. More production levels and the efficiency improvements, there have import-dependent OECD economies (e.g., Luxembourg (LUX), been gradual increases (decreases) in the shares of the carbon in- Sweden (SWE), and Switzerland (CHE)), as well as the less pro- ventories of developing (developed) countries (Fig. 6). ductive developing economies (e.g., Costa Rica (CRI), Cambodia Taking further “underlying stories” into account reveals the (KHM), Singapore (SGP)), have the highest shares of the imported differences between the subcategories in the developed and emissions. Also, a further comparison of the economic structures of developing countries (Kander et al., 2015). For instance, the relative the countries clarifies how the countries are different according to imported carbon to the direct emissions of the USA is 16.7%, the the most productive sector of their economies. For instance, among smallest number among the OECD countries, while the average values for the OECD and non-OECD countries are 37.4% and 15.4%, respectively (Fig. 4). Therefore, although the CBA of the developed countries is more than their PBA (due to the carbon leakage effect), different shares of the imports in the total consumption of these countries cause different CExA behaviors. Thus, the CExA inventory of the United States, as the most self-sufficient economy among all developed countries, has always been less than its PBA. This is because of its high PBA in place and the lower share of the demand for import of the embodied carbons, relative to the size of its pro- ductions and direct emissions of the USA (Figs. 4 and 5). In other words, relative to its high production-based emissions, while considering the total carbon trades for both the intermediate in- dustrial productions and final consumptions, the USA is adequately charged for their high PBA and, therefore, have a lower re- sponsibility for the global trade of the embodied carbons. However, the increasing demand for the embodied carbon of imports for the other G7 countries increases the CExA of these countries almost to their CBA. The lower productivity of the high- consuming other OECD countries causes the highest share of the demand for indirect carbons and therefore places their CExA higher than their CBA. Fig. 4. Comparison of the economic structure and the production efficiencies of the On the other hand, in the developing world, the different ratios country groups. of dependence on the imported carbons cause different CExA- 6 H. Khajehpour, Y. Saboohi and G. Tsatsaronis Journal of Cleaner Production 280 (2021) 124438 Fig. 5. Comparison of the CExA carbon accounting with the PBA and CBA approaches in 2011. based carbon inventories. Similar to the developed countries, the developed and applied to the global system. The CExA approach higher the share of the imported carbon in total consumptions in satisfies the fairness criteria and recommends a holistic view for industries and final consumers, the higher the CExA. Therefore, for accounting the global responsibilities related to the GHG emissions. China, the CExA is almost as low as the CBA, while it increases more It incorporates the functionalities of both the PBA and CBA princi- than the CBA for the BRIIS group of countries. ples. Therefore, it can account for levels and qualities of productions Finally, a clear relationship between the behavior of carbon in- as well as the levels and choice of the sources of the imports for the ventories has been experienced. For instance, a dramatic increase in total consumptions in both the final demands and intermediate the consumption of goods and services for both the final demands industries. Consequently, while considering the CExA carbon in- and the intermediate productions in the developed countries, have ventories, the countries are motivated toward reducing their direct resulted in a dramatic decrease in the consumption of the low and emissions (by less and greener productions) and the indirect medium-income developing countries (“Rest of the World”) in the emission loads on the imports (by lower total consumptions and year 2000 (Fig. 6). Detailed investigations of the changes in the choice of greener sources for imports), at the same time. global trades of different products and among different regions, With the aid of exergy cost accounting for the allocation of the clarify the underlying factors causing such changes. Therefore, carbon emissions among the multi-products of each component of further analyses of individual countries are available from the re- the global system, the CExA approach is applied to track the carbon sults presented in Supplementary Material 6. loads on the traded streams. As a result, the CExA approach rec- ommends a fair SBA allocation of the carbon responsibilities among the countries. 4. Conclusions In summary, according to the accounting results presented here, there are two features in the CExA: Tackling climate change needs integrated will and action in the First, sharing is not equivalent to averaging: the SBA inventories global community. In an ideal case, the differentiated re- do not necessarily lie between the PBA and CBA as they are not sponsibilities of countries should consider different aspects of averaging the conventional inventories; instead, they are incorpo- climate equity. This paper proposes a carbon accounting method rating further economic and technical details in the accounting. for a proper allocation of carbon loads among the countries. This Second, in contrast to the CBA and PBA inventories, there are no can help in an accurate accounting of the equity aspects of emission systematic benefits expected for the developed or developing responsibilities. groups of countries. This is because both the technical efficiencies A fair allocation of the contributions of individual countries to and the economic structures of countries matter in a proper SBA the global carbon emissions necessitates effective consideration of accounting. Therefore, considering the differences in these two the causes of emissions in both the demand and supply. Therefore, characteristics, there are different relative SBA inventories among from the production point of view, the producers shall be respon- the countries in both groups. sible for both the extent of their production activities and the Nevertheless, the method proposed here accounts for the technical efficiency of their productions. Whereas, simultaneously, induced consumption in the producing country by direct consid- from the consumption point of view, economies should take the eration of the PBA as an inherent part of the SBA. Therefore, the responsibility of the extent and the source of their consumptions in CExA approach of this research work is valid for full consideration both the final demands of their country (e.g., residential and of the differences in the region-specific technical efficiencies and transportation) and the intermediate consumptions at the local emission factors as they are inherently incorporated in the PBA part economic activities. This comprehensiveness guarantees the of the proposed SBA inventory. maximum effectiveness of the climate policies by motivating both As in any accounting method, there would be some losers and the producers and consumers toward lower emissions. winners under the non-PBA accounting rules. For instance, our In this research work, by combining the rational of the CEA and results show that, according to the proposed sharing approach, the the extended exergoenvironmental method, a new SBA approach is 7 H. Khajehpour, Y. Saboohi and G. Tsatsaronis Journal of Cleaner Production 280 (2021) 124438 Fig. 6. Temporal changes in the carbon inventories in different accounting approaches. The legends are similar for all graphs. group of the developed countries which have benefitted the most recent years. Also, the developed CExA method should be applied to from the carbon leakage effect by outsourcing the polluting in- more comprehensive and updated ICIO tables in the future. dustries to the developed nations, are found to be relatively most penalized for the embodied emissions on the imports to their economy. CRediT author statement The holistic consideration of different responsibility-generating factors in both the supply and demand sides, which is present in the Hossein Khajehpour: Conceptualization, Data curation, Formal proposed accounting method in this research work, is an advantage analysis, Investigation, Methodology, Software, Validation, Visual- over the other approaches. The rationality of the method can ization, Writing - original draft, Writing - review & editing. overcome resistance against its applicability in sharing the contri- Yadollah Saboohi: Funding acquisition, Project administration, butions in integrated global climate action. Conceptualization, Methodology, Resources, Supervision. George Indeed, all accounting methods have limitations for the design Tsatsaronis: Funding acquisition, Project administration, Concep- and implementation of climate policies. The SBA methods are tualization, Methodology, Resources, Supervision, Validation, Re- developed to improve the effectiveness of conventional accounting view and editing. methods. However, they cannot include many climate equity con- siderations and the “indirect and dynamic” effects of global climate policies (Kander et al., 2015). This necessitates, for example, the Declaration of competing interest incorporation of SBA accountings into the climate dynamic models in future studies. Data on inter-country input-output tables is The authors declare that they have no known competing improving by covering a higher number of countries and for more financial interests or personal relationships that could have appeared to influence the work reported in this paper. 8 H. Khajehpour, Y. Saboohi and G. Tsatsaronis Journal of Cleaner Production 280 (2021) 124438 Acknowledgment Jakob, M., Marschinski, R., 2013. Interpreting trade-related CO2 emission transfers. Nat. Clim. Change 3 (1), 19e23. https://doi.org/10.1038/nclimate1630. Kander, A., et al., 2015. National greenhouse-gas accounting for effective climate The authors are very thankful to the Alexander von Humboldt policy on international trade. Nat. Clim. Change 5 (5), 431e435. https://doi.org/ Foundation for supporting this research work. 10.1038/nclimate2555. Khajehpour, H., Saboohi, Y., Tsatsaronis, G., 2017. Environmental responsibility ac- counting in complex energy systems. J. Clean. Prod. 166 (Suppl. C), 998e1009. Appendix A. Supplementary data https://doi.org/10.1016/j.jclepro.2017.08.013. Lenzen, M., 2008. Consumer and producer environmental responsibility: a reply. Supplementary data to this article can be found online at Ecol. Econ. 66 (2), 547e550. https://doi.org/10.1016/j.ecolecon.2008.03.012. Lenzen, M., et al., 2007. Shared producer and consumer responsibility d theory and https://doi.org/10.1016/j.jclepro.2020.124438. practice. Ecol. Econ. 61 (1), 27e42. https://doi.org/10.1016/j.ecolecon.2006.05.018. Marques, A., et al., 2012. Income-based environmental responsibility. Ecol. Econ. 84, References 57e65. https://doi.org/10.1016/j.ecolecon.2012.09.010. Meyer, L., et al., 2009. Exergoenvironmental analysis for evaluation of the envi- ronmental impact of energy conversion systems. Energy 34 (1), 75e89. https:// Afionis, Stravros, Sakai, Marco, Scott, Kate, Barret, John, Gouldson, Andy, 2017. doi.org/10.1016/j.energy.2008.07.018. Consumption-based carbon accounting: does it have a future? WIREs Climate Odum, H.T., 2007. Environment, Power, and Society for the Twenty-First Century Change 8 (1), 438. https://doi.org/10.1002/wcc.438. the Hierarchy of Energy. Columbia University Press, null. Aichele, Rahel, Felbermayr, Gabriel, 2015. Kyoto and Carbon Leakage: An Empirical Peters, G.P., 2008. From production-based to consumption-based national emission Analysis of the Carbon Content of Bilateral Trade. The Review of Economics and inventories. Ecol. Econ. 65 (1), 13e23. https://doi.org/10.1016/j.ecolecon. Statistics 97 (1), 104e115. https://doi.org/10.1162/REST_a_00438. 2007.10.014. Bastianoni, S., Pulselli, F.M., Tiezzi, E., 2004. The problem of assigning responsibility Peters, G.P., Hertwich, E.G., 2008. Post-Kyoto greenhouse gas inventories: produc- for greenhouse gas emissions. Ecol. Econ. 49 (3), 253e257. https://doi.org/ tion versus consumption. Climatic Change 86 (1), 51e66. https://doi.org/ 10.1016/j.ecolecon.2004.01.018.  Barandica, J.M., Ferna 10.1007/s10584-007-9280-1. Berzosa, A., ndez-S anchez, G., 2014. A new proposal for Peters, G.P., et al., 2011. Growth in emission transfers via international trade from greenhouse gas emissions responsibility allocation: best available technologies 1990 to 2008. Proc. Natl. Acad. Sci. U.S.A. 108 (21), 8903e8908. https://doi.org/ approach. Integrated Environ. Assess. Manag. 10 (1), 95e101. https://doi.org/ 10.1073/pnas.1006388108. 10.1002/ieam.1489. Rocco, M.V., 2016. Chapter 6: case studies: applications of the exergy based Input- Csutora, M., Veto } ne mo  zner, Z., 2014. Proposing a beneficiary-based shared re- Output analysis. In: Primary Exergy Cost of Goods and Services: an Input e sponsibility approach for calculating national carbon accounts during the post- Output Approach. Springer International Publishing, Cham, pp. 101e137, 978-3- Kyoto era. Clim. Pol. 14 (5), 599e616. https://doi.org/10.1080/14693062. 319-43656-2. 2014.905442. Rodrigues, J., Domingos, T., 2008. Consumer and producer environmental re- Davis, S.J., Caldeira, K., 2010. Consumption-based accounting of CO2 emissions. Proc. sponsibility: comparing two approaches. Ecol. Econ. 66 (2), 533e546. https:// Natl. Acad. Sci. U.S.A. 107 (12), 5687e5692. https://doi.org/10.1073/pnas. doi.org/10.1016/j.ecolecon.2007.12.010. 0906974107. Rothman, D.S., 1998. Environmental Kuznets curvesdreal progress or passing the Davis, S.J., Peters, G.P., Caldeira, K., 2011. The supply chain of CO2 emissions. Proc. buck?: a case for consumption-based approaches. Ecol. Econ. 25 (2), 177e194. Natl. Acad. Sci. Unit. States Am. 108 (45), 18554e18559. https://doi.org/10.1073/ https://doi.org/10.1016/S0921-8009(97)00179-1. pnas.1107409108. Szargut, Jan, Morris, David R., 1987. Cumulative exergy consumption and cumula- Eder, P., Narodoslawsky, M., 1999. What environmental pressures are a region’s tive degree of perfection of chemical processes. International Journal of Energy industries responsible for? A method of analysis with descriptive indices and Research 11 (2), 254e261. https://doi.org/10.1002/er.4440110207. inputeoutput models1A previous version of this paper was presented at the Szargut, J., Stanek, W., 2010. Thermo-climatic cost of the domestic consumption Fourth Biennial Meeting of the International Society for Ecological Economics, products. Energy 35 (2), 1196e1199. https://doi.org/10.1016/j.energy.2009. Boston, 4e7 August 1996.1. Ecol. Econ. 29 (3), 359e374. https://doi.org/10.1016/ 04.025. S0921-8009(98)00092-5. UNFCCC, 1992. UNFCCC treaty. In: FCCC/INFORMAL/84- GE.05-62220 (E) 200705. IPCC, 2014. Summary for policymakers. In: Edenhofer, O., Pichs-Madruga, R., 1992: United Nations Framework Convention on Climate Change (UNFCCC) Sokona, Y., Farahani, E., Kadner, S., Seyboth, K., Adler, A., Baum, I., Brunner, S., Earth Summit in Rio de Janeiro. Eickemeier, P., Kriemann, B., Savolainen, J., Schlo € mer, S., von Stechow, C., Wiebe, K., Yamano, N., 2016. Estimating CO2 Emissions Embodied in Final Demand Zwickel, T., Minx, J.C. (Eds.), Climate Change 2014, Mitigation of Climate Change. and Trade Using the OECD ICIO 2015. OECD Publishing, Paris. https://doi.org/ Contribution of Working Group III to the Fifth Assessment Report of the 10.1787/5jlrcm216xkl-en. Intergovernmental Panel on Climate Change. Cambridge, United Kingdom and New York, NY, USA. 9