CO₂ emissions flow due to international trade: multi-regional input–output approach for Spain

Abstract

As a result of globalization and free trade agreements, international trade is enormously growing and putting more pressure on the environment over the last few decades. This has drawn the attention of both environmental system analysts and economists in response to the ever-growing concerns over climate change and the urgent need for global action for its mitigation. This work aims at analysing the implication of international trade in terms of CO₂ responsibility between Spain and its important trading partners using a multi-regional input–output approach based on the data from Organisation for Economic Co-operation and Development and World Input–Output Database. The empirical results show that Spain is a net importer of CO₂ emissions, equivalent to 29% of its emissions due to domestic production. The CO₂ emissions embodied in the trade with China take the largest share and this is mainly due to the importation of energy-intensive products from China. When analysed by the end-use type, intermediate goods contribute the largest portion, which is about 67% of the total emissions associated with imported goods. Products such as motor vehicles, chemicals, a variety of machineries and equipment, textile and leather products, and construction materials are the key imports responsible for the major portion of CO₂ emissions. Being at its peak in 2005, the construction sector is the most responsible activity behind both domestic and imported emissions.

Keywords:

• CO₂ emissions
• muti-regional input–output analysis
• international trade
• embodied emissions

1. Introduction

Concerns over the economic and the social consequences of climate change continue to grow as there are scientific evidences on its potential impact (IPCC, 2007; Stern et al., 2006). The Kyoto Protocol has been an international measure adopted since 1997 in an attempt to mitigate the potential cost of climate change by committing countries that have ratified it to reduce their greenhouse gas (GHG) emissions by 5% below the 1990 levels in the first commitment period of 2008–2012. However, it is often criticized for not being effective, mainly for two basic reasons. On the one hand, the protocol binds only a subgroup of high-income countries that could result in shifting of energy- and emission-intensive goods to be produced in economically emerging but non-ratified countries such as China and India. On the other hand, the responsibility to committing countries is only from the production perspective, in a way that only emissions that occurred within the national territories are considered. Emissions embodied in imported goods are not taken into consideration. International trade is enormously growing due to free trade agreements and putting more pressure on the environment (Antweiler, Copeland, & Taylor, 2001; Copeland & Taylor, 2005; Machado, Schaeffer, & Worrell, 2001). The fact how to capture emissions embodied in international trade has been discussed for long time by both environmental and economic analysts (Lin & Sun, 2010; Peters & Hertwich, 2006a, 2006b; Weber, Peters, Guan, & Hubacek, 2008).

This paper analyses the implications of international trade between Spain and its key trading partners with respect to CO₂ emissions embodiment. The model applied is based on input–output (IO) framework (Leontief, 1941; Miller & Blair, 2009). Environmental Input–Output (EIO) analysis is a top-down approach which accounts for resource consumption and emissions release using IO tables (Leontief, 1970). EIO has been used to estimate pollution embodiment in international trade. In early 1970s, Walter (1973) applied EIO models to examine the US product profile of exports and imports and their environmental profiles. But Fieleke (1975) was the first who implemented the Leontief inverse to determine the US trade deficit in embodied energy using an EIO. Since then, a number of studies have been carried out using the EIO approach to analyse the environmental implication of international trades. Most studies have implemented the single-regional input–output (SRIO) approach, which is usually based on the very simplified assumption of the same technology for both imported and domestic products (Kondo, Moriguchi, & Shimizu, 1998; Lenzen, 1998; Machado et al., 2001; Sánchez-Chóliz & Duarte, 2004; Wyckoff & Roop, 1994). However, such assumption is far from the reality, particularly when there are high discrepancies in both technology and energy mixes between the trading partners, and therefore, it may be subjected to large errors as estimated by Lenzen, Pade, and Munksgaard (2004) and Peters and Hertwich (2006b). A detailed review of the use of SRIO models for international trade and emissions analysis can be found elsewhere (Wiedmann, 2009; Wiedmann, Lenzen, Turner, & Barrett, 2007).

On the premises of avoiding errors due to the same technology assumption in SRIO analysis, the development of multi-regional input–output (MRIO) approach emerged as an alternative in environmental analysis associated with international trade. Unlike the SRIO, a complete MRIO model differentiates the production technology and then the related energy and environmental profile of imported goods and services from domestic ones. The MRIO model fully integrates the domestic requirement matrix with imports, which is derived from international trade flows to simulate the interdependency of sectors in one region with all other sectors in trading partners. Hence, it allows seeing the entire supply chain of trades and emissions flows linked with goods and services imported to or exported from the domestic region under study. Though the application of the MRIO model dates back to mid-1900s, it was only recently that different works have emerged applying MRIO models in the analysis of emissions embodied in trade, to cite but a few are Lenzen et al. (2004), Peters and Hertwich (2006a), Wiedmann (2009), Hertwich and Peters (2009), Peters and Hertwich (2006c), Peters and Hertwich (2006b), Guan and Hubacek (2007), Nijdam, Wilting, Goedkoop, and Madsen (2005), Davis and Caldeira (2010), Chen and Chen (2011), Peters, Minx, Weber, and Edenhofer (2011), Chen, Chen, and Chen (2013) and Chen and Chen (2013).

Particular to the Spanish situation, there are few studies which analyse the environmental implication of international trade between Spain and rest of the world (RoW). Sánchez-Chóliz and Duarte (2004) used a SRIO model to study the potential impact of international trade in the level of CO₂ emissions generated by the economy for the year 1995. The study concluded that the pollution imported through the intermediate and final demand requirement of the Spanish economy is offset by the pollution exported to satisfy demands outside Spain, leaving the net trade balance of only 4237 thousand ton (1.3% of the total emissions produced in Spain). Recently, Cadarso, López, Gómez, and Tobarra (2012) analysed the impact of international trade and the shared responsibility in the Spanish economy for the period 2000–2005 using the SRIO model. They defined a set of criteria to share the responsibility of sectors for their direct emissions due to production as well as due to their input requirements. The responsibility was shared with all sectors along the global production chain depending on the value added on each step. The results reveal that the Spanish economy shows an increase of 40.8% emissions responsibility when it is based on consumption perspective than production for the year 2005.

With regard to the MRIO model application in Spain, to our knowledge, only a few studies have been conducted. Serrano and Dietzenbacher (2010) developed a two-region (Spain and RoW) MRIO model to examine the emissions responsibility of the Spanish economy due to international trade taking into account both the net trade balance and responsibility balance concepts. They considered the years 1995 and 2000 and evaluated the effect for nine different types of gases. They concluded that both concepts (net trade balance and responsibility balance) yield the same results. More recently, a fully integrated MRIO model that considers 19 regions (all 17 Spanish regions, EU and RoW) is developed by Cazcarro, Duarte, and Sánchez Chóliz (2013) to analyse the water flow and water footprints for Spain.

Applying the Organisation for Economic Co-operation and Development (OECD) IO table and bilateral trade data for 13 regions, this study aims at analysing CO₂ emissions due to international trade flow of the Spanish economy. The MRIO model allows us to understand the link between demand on production of goods and services among the regions and their environmental consequences (in terms of CO₂ emissions). Therefore, this study addresses the following questions. How would the Spanish economy be seen in terms of trade balance based on its consumption and production structure? Which partners have the most significant contribution in emissions due to trading with Spain? What are the key products responsible for the most of emissions embodied in imports? How the results from the MRIO model would be used in CO₂ emissions reduction strategies by policy-makers?

The rest of the paper is structured as follows. Section 2 presents the methodological foundations of MRIO and the data sources used. The main results and their policy implications are discussed in Section 3. Section 4 concludes.

2. Methodology and database

MRIO models are recognized as a suitable tool to analyse emission embodied in trade both from consumption and from production perspectives. They are able to trace the emissions linked with trades as a result of demand and supply interdependency of industries in domestic economy with foreign agents. In this section, we present the detailed analysis performed for the Spanish economy to understand the implication of international trade with regard to CO₂ emissions flow. The OECD-based IO tables for selected 13 regions and the bilateral trade date are used for this purpose. The MRIO model is in line with the one proposed by Peters, Hertwich, and Suh (2009) and it is briefly summarized in the following. Assuming that there are n regions, in which each region's production is classified into m sectors, the MRIO model can be formulated using the traditional IO framework as follows:(1) The linearity assumption of the IO model allows expressing the sectoral outputs for arbitrary demands. Therefore, Equation (1) can be formulated as follows:(2) where is a matrix of total output, in which each diagonal element represents the domestic total output and the off-diagonal matrix stands for the product requirement from other regions. is total inter-industry requirement matrix, which integrates both the domestic and imported inputs for each sector in each region. It can be described as, . Each diagonal matrix in is a square matrix of technology requirements on domestic production and denotes the inter-industry technological requirement from region to region . is a matrix of total final demands which comprises both demand on domestic production, import and exports. It can be represented as: where and are vectors of final demand requirements on domestic production by domestic consumers in region 1 and exports to other regions, respectively. represents the final demand of goods from region to region 1.

The economic IO model represented in Equation (2) can be extended with environmental data, for example, emissions to air, energy consumption, hazardous waste generation or other impacts so as to use it for environmental assessment. In order to determine the total emissions throughout the economy, the emission values are used with the economic IO model. An environmental extension of the model in Equation (2) can be expressed as follows:(3) where is the total environmental emissions associated with the production of output from all n regions. is a vector of direct emission intensities, in which each element represents a vector of emission intensities of sectors in region i. The emission intensity of each sector refers to the emissions released associated with each dollar of economic output, which can be determined by dividing the total sectoral emissions by the total sectoral output.

2.1. Data

The IO tables for all 13 regions and the bilateral trade data are provided by OECD. The OECD supplies a consistent and harmonized IO tables that can be used for international trade and environmental analysis. The latest OECD IO data set covers inter-industrial transactions of goods and services of 48 countries (all OECD countries but Iceland and 15 non-member countries). It provides both the domestic and import tables separately for the years 1995, 2000 and 2005. The IO tables and the bilateral trade data are in accordance with a harmonized industry structure of the International Standard Industrial Classification of all Economic Activities (ISIC) Rev. 3. The OECD IO tables are aggregated to 37 sectors. The OECD bilateral trade database provides monetary values in USD ($) of imports and exports of goods and services broken down by industrial sectors and by end-use categories (Zhu, Yamano, & Cimper, 2011). Each off-diagonal block matrix , which represents the import requirements from region to region , is derived from the bilateral trade data and total output vector of region. The bilateral data from OECD are only for intermediate goods and services; they do not include imports directly consumed by individuals. Therefore, the vector of imported final goods, elements of total final demand vector y in Equation (1), are derived based on the assumption that imported final demands are in the same proportion as imported intermediate demands from the trading partners using the following equation:¹ (4) where is the imported final demand to Spain and is the share of import of each goods, which is calculated as follows:(5) where is the total import of intermediate good c from region i to domestic region .

The data on CO₂ emissions were obtained from the World Input–Output Database (WIOD) (WIOD, 2012). WIOD is a project funded by the European Commissions, Research Directorate General as part of the 7th Framework Programme. It provides a set of harmonized supply and use tables and extensive satellite accounts for 27 EU countries and 13 non-EU major world countries for the period from 1995 to 2009. The CO₂ emissions data of WIOD include both energy-related and non-energy-related emissions and they are separately reported. In line with the OECD IO tables, the CO₂ emissions data of WIOD are classified based on ISIC Rev. 3. The CO₂ data are used to derive the direct emission intensities vector in Equation (3).

All the data sets considered in the analysis are for the year 2005. Figure 1 illustrates the import and export structure of Spain with its trade partners for the year 2005. According to the data from Instituto Nacional de Estadística (INE) (INE, 2008), around 66% of the total imported goods are from the EU and 76% of the total exports are to the EU. Germany (DE), France (FR), Great Britain (GB), Italy (IT), the Netherlands (NL), Portugal (PT) and Belgium (BE) are among the most dominant partners of Spain in both imports and exports. All together represent around 55% and 63% of the total imports and exports, respectively. The Asian market represents 17% of imports and 7% of exports of the Spanish international trade. China (CN) and Japan (JP) are important partners of Spain, which together account for 8% of the total Spanish imports. Around 9% of the total imports and 10% of exports from Spain are represented by America. The USA and Brazil (BR) are among the important partners.

Figure 1. Import and export structure of Spain with its important trading partners (own elaboration based on data from INE).

In linking IO domestic table with the bilateral trade data, the RAS procedure, an iterative method that updates IO tables, was implemented. Due to unavoidable asymmetry problems, which may result from different trade systems definition, characterization of specific goods or transaction types, different valuation of imports and exports between countries, and consideration of transit trades, re-exports and re-imports, and so on, there are some discrepancies between the values of imported goods by a given country and the corresponding exports from the other trade partner. As a result, the total intermediate output of each sector does not match with the sum of its domestic and import outputs, and for this reason, the RAS method is applied (Miller & Blair, 2009).

3. Results and discussion

This section presents the main findings and analysis of CO₂ emissions embodied in trades between Spain and the RoW for the year 2005. Generally, there are two perspectives from which CO₂ emissions embodied in international trade could be estimated: the production and consumption perspectives. Production-based emissions accounting, a method used in Kyoto protocol, considers CO₂ emissions that are emitted during the production activities of domestic economy, without regarding where the produced goods and services are consumed, that is, domestically or abroad. It includes emissions associated with the production of goods and services which are produced and consumed domestically and those which are exported to other markets. However, it does not take into account the emissions that occurred outside the national territory during the production process of goods and services which are consumed by domestic consumers. When the production base accounting framework is considered, a total of 292.5 Mton of CO₂ emissions occurred on the Spanish territory, 73.5% of which are emissions associated with the production demanded for domestic consumers and the 26.5% is due to demand by foreign market. Figure 2 shows the contribution of each sector to the total domestic CO₂ emissions linked with the production of both domestic and export final demands. When the sectoral emission intensities are analysed, electricity, gas and water supply; other non-metallic mineral; and Coke, refined petroleum products and nuclear fuel are the most important sectors that together are responsible for around 75% of the total direct emissions that occurred in Spain to provide other sectors energy and materials to produce final demand for both domestic and export consumptions. In line with Alcantara and Padilla (2006), these are the key sectors that concentrate most of the emissions caused from the production perspective of the Spanish economy. As shown in Figure 2, the emission intensity of the electricity sector is by far higher than other key sectors. This shows how the energetic outputs have great impact on the total CO₂ emissions generated in Spain. Emissions reduction measures in energy such as increasing the share of renewable energy source in the national energy production mix, increasing the efficiency of energy-generating plant by introducing more advanced technologies, switching from coal-powered turbine to natural gas or from single-cycle turbine to combined-cycle turbine and so on can contribute to a decrease in total Spanish domestic emissions.

Figure 2. CO₂ emissions produced in Spain to supply both domestic and export demands, 2005.

A different situation emerges when, instead of emission intensities, the total emissions (sum of direct and indirect) are considered. The total emissions express the emissions generated by the whole productive system in response to the production of final demand by the corresponding sector. When analysed from the domestic demand side, the emissions concentrated in the key sectors are distributed among different sectors according to the final demand on each activity and their dependence on the key sectors to produce the demanded outputs. In this regard, the construction sector of the Spanish economy is the most important, which relies on the other sectors in carrying out its activities. This implies that the sector buys pollution from other sectors through its input requirements to produce its final domestic demand for consumption. At its peak, in 2005, the production of energy and materials needed to satisfy the demand from the construction sector alone generated around 20% of the total emissions in Spain. Despite its burst after the crisis, from 2008, the sector is thought to be the significant contributor of the Spanish economy and was the most important factor linked with the economic boom in the country for almost a decade. Its high contribution to the total CO₂ emissions is resulted from its tremendous economic growth. Next to construction are electricity, gas and water supply; wholesale and retail trade, and repairs; hotels and restaurants; and food products, beverages and tobacco, which together contribute to 38% of the total indirect CO₂ emissions. When analysed from the exports demand side, transport and storage; other non-metallic mineral; and chemical and chemical products sectors generated an important amount of domestic emissions to provide goods internationally.

Unlike the production-based emissions accounting framework, the consumption-based accounting is founded on the principle that consumers are responsible for all emissions associated with the production of goods and services which they are consuming regardless of where they are exactly produced. It comprises emissions associated with all products which are produced and consumed domestically, products produced abroad, but directly consumed within the economic boundary, and finally products which are produced abroad and supplied to industries as intermediate products and finally consumed domestically. This can be estimated as the sum of domestic emissions due to domestic final demand plus the imported emissions from all other regions to the domestic region. The consumption-based emissions, which show all the emissions which are directly or indirectly produced elsewhere in order to provide goods and services consumed in Spain, are presented in Figure 3. According to the analysis, the Spanish economy is responsible for 378 Mton CO₂ emissions for the year 2005. The difference between the consumption- and production-based emissions is the net emissions trade balance, which is a surplus or deficit emission from the import and export. It reveals that Spain is a net importer of emissions with a trade balance of 85.4 Mton of CO₂ emissions, which is approximately 29% of its production emissions. This reflects the significance of international trade in the context of CO₂ emissions and its implication for policy considerations. There are several MRIO studies that assess CO₂ emissions embodied in the international trade for Spain and other regions. Our result is in line with most of the already exiting literature. For example, Peters and Hertwich (2008) carried out a quantitative comparison of CO₂ emissions embodied in international trade among 87 countries for the year 2001. The results suggest that over 5.3 Gton of CO₂ emissions are embodied in trade. Spain is listed among the net emissions importer countries with around 31 Mton differences between consumption- and production-related emissions. Davis and Caldeira (2010) also performed a global consumption-based CO₂ accounting for the year 2004. The assessment is based on data from Version 7 of the Global Trade Analysis Project, which provides extensive data on trade, economic IO, GDP, population, energy consumption and associated CO₂ emissions. According to Davis and Caldeira (2010), Spain is ranked as the seventh CO₂ importing countries with a net emissions balance of 67 Mton. More recently, Chen et al. (2013) analysed the embodiment of CO₂ emissions from fossil fuels combustion for the world economy in 2004. The difference between the imported and exported CO₂ emissions for Spain is estimated to be around 55 Mton, which makes the country a net emission importer.

Figure 3. Consumption-based emission by country where goods and services are produced, 2005.

Figure 3 also illustrates the country contribution of the total emissions due to consumption in Spain both the domestic emissions and emissions embodied in imported goods from the key trading partners. As shown, emissions associated with imported goods from China dominate to a great extent, which contributes to 17% of the total consumption-related emissions. When only the import-related emissions are considered, goods from China are responsible for around 37% the emissions. It is followed by the USA that accounts for 8% of consumption emissions or 17% of import-related emissions. Though not as high as the USA and China, considerable amounts of CO₂ are also generated in Germany, Italy, Japan, France and the UK. Aggregated emissions due to imports from the EU countries account for around 16% of total consumption-based emissions. It represents around 38% of the total embodied emissions due to imports, which is almost comparable to the emissions contribution of China alone.

A closer look at the import structure and emissions embodied in imported goods gives more insights on the key trading partners and their contribution to the global emissions derived from their exports to Spain. Table 1 summarizes by demand type the emissions that occurred outside Spain due to its import requirement. As can be seen, the largest portion of the emissions is due to demand on the intermediate goods, which are reprocessed by domestic sectors before being supplied to domestic consumers. This accounts for around 41% of the total import-related emissions. Whereas, imported goods that are directly supplied to domestic final consumer represent 33%. The Spanish economy re-exports around 26% of the total imported emissions to other countries.

Table 1. CO₂ emissions embodied in imported goods and services (Mton), 2005.

Trading partners	Intermediate demand (domestic)	Intermediate demand (re-export)	Final demand (domestic)	Total import emissions	%
BE	1539	898	1233	3670	2
BR	729	462	937	2128	1
CN	26,542	18,010	32,879	77,431	36
DE	9464	5641	4965	20,070	9
FR	7464	4426	2465	14,355	7
GB	5449	3309	3655	12,413	6
IT	8536	5119	3130	16,785	8
JP	5203	3178	4992	13,373	6
NL	1893	1117	1457	4467	2
PT	6885	3428	545	10,858	5
RW	2018	1123	2178	5319	2
USA	14,498	8921	13,406	36,825	17
Total	90,221	55,630	71,843	217,694	100

The imported goods and services that dominate the total imported emissions from the selected countries are presented in Figure 4. As can be seen, sectors such as motor vehicles, chemicals, machineries and equipment, textile and leather products, and construction materials are the key imports that drive the emissions due to their production in the respective exporting countries.

Figure 4. CO₂ emissions and import values of top 10 sectors from selected countries, 2005.

Some of these products are directly supplied as final goods to be consumed directly by the consumers, while others are imported as raw materials or intermediate products to undergo the Spanish production system before they are supplied to final consumers for both domestic and export destinations. As can be seen, a considerable amount of emissions from the importation of motor vehicles and chemicals are re-exported to other countries. Additionally, emissions from the import of building and construction sector are finally consumed by the Spanish economy.

 Figure 4 also illustrates that most goods and services imported from China have relatively low import value, but high CO₂ emissions compared with other countries such as Germany, Italy and the USA. This could be, on the one hand, due to cheap labour market in China. On the other hand, it could be due to the relatively unclean energy mix of the Chinese economy which is dominated by Coal. Until recently, coal has been covering much of the demand in Chinese energy production and its tremendously growing use has been among the most responsible sources for global CO₂ emissions.

It is obvious that in one or another way, emissions associated with the production of goods and services from a given country are dependent on the national energy mix and the production technology. Therefore, the import and emissions profile of different sectors illustrated in Figure 4 reveals the fact that Spain is importing more energy-intensive products from China than from other trading partners. A good example for this could be the emissions associated with motor vehicles imports. Representing only 7% of the total import value, 2255 M$, vehicles from China are responsible for around 10 Mt of embodied CO₂ emissions, whereas by far larger volume of the same product group imported from Germany shows quite smaller emissions. Germany is one of the leading exporters of motor vehicles to Spain, which satisfies around 28% of the Spanish total demand on motor vehicles. Despite the fact that a large number of motor vehicles are imported from Germany, a small amount of direct emissions is associated with them. This suggests that the German economy imports energy-intensive products such as different parts of vehicles for example from China, which are assembled and shipped to Spain.

4. Conclusion

The study generally analyses the CO₂ emissions generated as a result of international trade between Spain and its trading partners. The detailed analysis from the 13-region multi-directional MRIO model based on the data from OECD and WIOD enables us to examine the global carbon contribution of Spain for the year 2005. The emissions balance shows that Spain is a net importer of emissions, which is estimated to be 29% of its domestic emissions. This implies that a larger amount of emissions are embodied in Spanish imports than the emissions embodied in its exports. In line with other related studies in Spain, the findings in this paper reveal that the trade between Spain and China takes the largest share of CO₂ emissions. The still inefficient energy production mix of Chinese economy could also highlight the emissions embodied in each product that is imported from China, although the most important partners in terms of values are Germany, France, Italy and Great Britain. The embodied emission from countries with similar production structure and environmental profile is mainly driven by the total volumes imported. Regarding their import and export flow, the largest portion of embodied emissions is due to the demand by production sectors, which accounts for 67% of the total imported emissions. The Construction sector is the most responsible one behind these emissions. The year 2005 was the peak for the remarkable growth of this sector in Spain. Deliveries of the service from the sector demand large amount of input materials and, accordingly, emissions. Emissions from the import of large-value products such as motor vehicles and spare parts, machineries and equipment, textile and textile products, leather and footwear, and chemical products are also considerably high. High-value but relatively low-emission-intensive products such as computers, office machineries and other electronic devices have very little contribution.

It is worth mentioning the current position of Spain in the climate policy and its target on carbon emissions. The first Kyoto target set for Spain is 15% above the level in 1990. Past CO₂ emissions trend for Spain shows a continuous increase until 2007. But there is a decreasing trend from 2007 onwards. One is due to increasing the share of hydro, wind and nuclear power in the public electricity and heat production mix while reducing the share of thermal power production. The other emissions reduction is achieved from the transportation sector, which results from continuing economic crisis in Spain. Though there is a reduction trend, Spain still seems to be far from meeting its Kyoto target. Recently, Spain has made a number of legislation changes to renewable energy in response to reduce its GHG emissions. While there is an increasing local effort to reduce CO₂ emissions, the challenges from the international trade yet remain unanswered.

The international trade between Spain and RoW has important policy implications from a global perspective. The following policy implication could be drawn from the empirical results obtained. As it is clearly observed, emissions from countries that are economically highly emerging and dominating the global trade but not legally bounded under international emissions reduction agreements are considerably high. This has been a reason for many developed and major contributors to global GHG emissions to refuse to ratify the Kyoto Protocol; the USA and Canada could be a good example. They consider that such unbalanced treatment favours the production of goods in non-Annex I countries while it unfairly affects the competitiveness of producers who face restrictions on their emissions. Such inequity issue of international CO₂ regulations could be avoided by implementing border adjustment taxes. Border taxes or border adjustments taxes are levies imposed on imported products from countries that are not yet involved in any international CO₂ reduction protocols. Environmental tax or tradable permit price is believed by many economists and environmentalists to be the most effective and efficient way to reduce human-related CO₂ emissions. They provide incentives to polluting industries for reducing their emissions through market signals and allow internalizing the (negative) externalities caused in the environment. As the EU already established permit price on ‘cap and trade’ principle with the target of reducing GHG emissions to at least 20% below the 1990 level by the year 2020, most companies are under this restriction. The border tax on CO₂ therefore will charge companies that import goods and services from countries outside the EU the same price as current CO₂ price based on their life cycle emissions embodied in their production. The MRIO approach presented here was used as a helpful tool to analyse the global emissions flow linked with imports and exports between Spain and its important trade partners and the key sectors or product groups that are behind the embodied emissions.

The future perspective will be to construct a fully integrated MRIO model for French economy, which allows seeing the pollution flows within regions in the country and also from outside the country associated with imported good flows. It will also be used to analyse the key sectors in each region and their dependence on other regions within the country or outside. The model will be extended to address other environmental impacts such as waster footprint and resource consumption in addition to the CO₂ emissions accounting.

Acknowledgement

The authors would like to thank Dr. Sangwon Suh for his advices and supports to carry out this research.

Notes

1.  The symbol ^ accompanying a vector denotes the diagonalization of the corresponding vector.