A multi-regional input-output analysis of ozone precursor emissions embodied in Spanish international trade

A multi-regional Input-Output analysis of ozone precursor emissions embodied in Spanish international trade Román, Rocíoa*; Cansino, José M.a; Rueda-Cantuche, José M.b a Universidad de Sevilla (Spain) and Universidad Autónoma de Chile (Chile). University of Seville. Facultad de CC. Económicas y Empresariales Avda Ramon y Cajal 1 41018 Sevilla bEuropean Commission, DG Joint Research Centre Edificio EXPO, Inca Garcilaso, 3,41092Seville E-mail: [email protected] *Corresponding author: Rocio Román University of Seville. Facultad de CC. Económicas y Empresariales Ramon y Cajal 1 41018 Sevilla Spain +34 954551657 [email protected] 2 A multi-regional Input-Output analysis of ozone precursor emissions embodied in Spanish internationl trade 1. Introduction Many countries and international organizations are concerned about the effects that critical levels of ozone have on human health and the environment, and therefore the links between greenhouse gases and air pollution.1 Both issues share common sources of emissions — primarily from fuel combustion in industry and households, transport and agriculture — but also through cross-issue pollutant effects. Rypdal et al. (2005) advised on the need to include tropospheric ozone and aerosol precursors into climate agreements. At present, ozone is one of the most problematic pollutants in European Union (EU) in terms of harm to human health. High levels of ozone cause respiratory and cardiovascular health problems and lead to premature mortality. According to the European Environment Agency EEA (2014), around 14-17% of the urban population in the EU is exposed to ozone concentrations above EU reference levels (120 µg/m3 per eight-hour period) during 2010-2012. However, these figures are worse when the World Health Organization (WHO) reference levels are considered, as they are stricter than EU target values. In that case, more than 95% of the urban population in Europe is exposed to ozone concentrations above WHO reference levels (100 µg/m3 per eight-hour period) (WHO, 2006). In order to provide the European Commission (EC) with evidence-based advice on the health aspects of air pollution, the report published by the WHO/Europe (2013) in relation to the effects of air pollution on health concludes that, in the case of ozone, the harmful effects for health can occur at air pollution concentrations lower those established by the WHO 2005 Guidelines (WHO, 2006). For this reason, the WHO report recommends a revision of the European Union Air Pollution policy that updates the limits of the main pollutants that are harmful to health. 1 The strong dependence of ozone levels on atmospheric conditions suggests that the projected changes in climate leading to warmer temperatures could also result in increased ground-level ozone concentrations in many regions. 3 In EU, the Air Quality Directive (2008/50/EC), which replaced nearly all the previous EU air quality legislation, — complemented by Directive 2004/107/ EC — set legally binding limits for ground-level concentrations of outdoor air pollutants. Also, the National Emissions Ceiling Directive (NEC Directive), adopted on 23 October 2001, sets upper limits for each Member State for the total emissions in 2010 of the four main pollutants responsible for acidification, eutrophication and ground-level ozone pollution. Additionally, ozone pollution is considered a global problem as it has been dealt with by the Convention on Long-range Transboundary Air Pollution. In fact, the Hemispheric Transport of Air Pollution -HTAPTask Force has recently shown that intercontinental transport contributes significantly to air pollution in terms of ozone emissions concentration. Therefore, the level of ozone pollution at any given location depends not only on local and regional sources, but also on sources from other continents (UNECE, 1999; 2007). Effective international commitments must be supported by rigorous measures of pollutant emissions by country. This idea was set out in Article 4 of United Nations Framework Convention on Climate Change (UNFCCC) (UN, 1992) within the context of global warming. This article establishes that countries have both common and different responsibilities regarding this issue. Countries put pressure on global ozone emissions concentrations not only because of their domestic production but also because of their domestic consumption. However, actual international ozone emissions commitments are based on the production perspective (emissions inventories) and not on the consumption-based perspective, that is, the emissions embodied in the consumption of imported and domestically produced goods and services by the residents of one specific country. Some previous papers consider one or both of these two perspectives for analyzing the emissions embodied in the trade of one country or area , i.e. Weber et al. (2008), Marin et al. (2012), Homma et al. (2012), Dieztzenbacher et al. (2012) or Sato (2012), but they are mainly focused on the greenhouse gas emissions or just on CO2 emissions. However, as far as we know, there is no previous literature analyzing the ozone precursor emissions from these two perspectives for the case of Spain. 4 As pointed out recently by the European Environment Agency (EEA,2014), further studies are needed to properly allocate the air emissions produced in the EU to EU and non-EU countries. In fact, the health effects of ozone precursor emissions are becoming one of the priority issues in the governments’ agenda (WHO, 2007). Effects on agriculture are also of interest for recent literature (see Kumar et al., 2011 for California-USA, Oksanen et al., 2013 for India and Yi et al., 2016 for China). The aim of this paper is to analyze the ozone precursor emissions by sectors for Spain in 2009, embodied in Spanish international trade with i) the EU and ii) the rest of the world (non-EU countries). The interest in focusing on Spain is twofold, Spain is the EU Member State with the highest level of non-methane voltaic organic compounds emissions (one of the main ozone precursor) and Spain did not comply with the emission ceiling for this pollutant in 2010, despite the EU commitments on ozone precursor emissions set out in the National Emissions Ceilings Directive (NEC Directive). The results will lead us to provide some policy recommendations focused on the most polluting sectors from the Spanish international trade perspective. As the literature suggests, multi-regional input-output models (MRIO) are the most appropriate tool for the calculation to account for emissions to air, either from the production perspective or from the consumption perspective (Miller and Blair, 2009, Wiedmann, 2009, Zhou and Imura, 2011, Su and Ang, 2014, Zhang et al., 2015, Hong et al, 2016, Brizga et al. 2016 and Jiang et al., 2016). To the best of our knowledge, the only previous paper on ozone precursor emissions using a multi-regional input output model is Arto et al. (2012). This paper contributes to the literature in three areas. Firstly, the paper is focused on the Spanish economy, offering more detailed information about sectors and countries involved in the ozone precursor emissions embodied in Spanish international trade. Secondly, the most recent available version of environmental accounts of World Input-Output Database (WIOD) has been used, i.e. 2009 (Arto et al., 2012, used the year 2008). And thirdly, some general policy recommendations are provided based on the results. The article is structured as follows. Section 2 explains the methodological approach and briefly describes the database, i.e. WIOD. The results are reported by type of ozone precursor gas and by area in Section 3. The results are discussed in Section 3. The conclusions and recommendations for policy analysis are given in Section 4. 5 2. Methodology 2.1 Input-output analysis Input-output analysis (IOA) revolves around the so called input-output tables, which reflect the supply and demand of the economy in terms of products, industries and final users. IOA is a useful method to better understand and account for the links between consumption and production sectors (Leontief, 1970). By using the so-called Leontief quantity model (Rueda-Cantuche, 2010), the total output of an economy X can be broken down into final and intermediate demand, as indicated in (1): X AX Y= + (1) where X is the total industry output vector for n industries (n x 1); Z = AX is a matrix describing the intermediate products of industries; A is a matrix (n x n) of input-output coefficients showing the inputs needed per unit of output by each industry; and Y stands for a final demand matrix (n x 1) showing the consumption, investment and exports of all goods and services. Within this framework, industry-by-industry IO tables from the WIOD database have been used (Dietzenbacher et al., 2013) with the same number of industries and commodities (n). Reordering (1), it yields: ( ) 1 X I A Y LY − = − ⋅=⋅ (2) where I is the identity matrix (n x n) and (I-A) -1 is the so-called Leontief inverse matrix, L (n x n), which shows the total requirements of the economy for the production of goods and services to satisfy a certain level of final demand. Leontief-style IOA accounting has become an increasingly active area of research for a variety of environmental indicators, including CO2 (Kanemoto et al., 2014; Chen & Zhang, 2010). The Environmental Input-Output (EIO) model allows us to analyze the link between emissions, productive sectors and the final demand. By multiplying both 6 sides of equation (2) by the emission coefficients (C), which are the emissions per unit of output, the following equation results: ( ) 1 c C I A Y CLY − = ⋅ − ⋅= ⋅⋅  (3) where c is an n x 1 vector representing the total emissions sector and 𝐶 󰆹 is a diagonal matrix n x n that represents the emission coefficients of the economic sectors. These emission coefficients have been calculated as the total emissions of each industry over the total output. 2.2 Multi-regional input-output analysis The multi-regional input-output analysis is based on a set of interconnected input-output tables of various countries (Miller & Blair, 2009). While equation (3) refers to one single country with n industries, similar equation for a three-region model with n industries in each region, namely: Spain (u), rest of the EU (r) and rest of the world (w) is shown as follows: ˆ00 ˆ 00 ˆ 00 uu ur uw u u uu ur uw ru rr rw ij r r ru rr rw i js wu wr ww w wu wr ww w C c LLL YYY c C L L L Y Y Y CLY c LLL YYY C           = ⋅ ⋅ =⋅⋅                  (4) i being the region where the intermediate products have been produced; j the region that has imported the intermediate products and has incorporated them into final production and s the region where the final products are consumed. The equation (4) is redefined allowing for a fully-fledged decomposition of the final demand by region. 7 ˆˆˆ ˆˆˆ ˆˆˆ ˆˆˆ ˆˆˆ ˆˆˆ uu ur uw uu ur uw u uu u ru u wu u ur u rr u wr u ru rr rw ru rr rw r r uu r ru r wu r ur r rr r wr wu wr ww wu wr ww ww uu w ru w wu w ur w rr w wr CLY CLY CLY CLY CLY CLY c c CLY CLY CLY CLY CLY CLY cCLY CLY CL Y CLY CLY CL Y        = +             ˆˆˆ exp exp ˆˆˆ exp ˆˆˆ exp uu ur uw uuu uuwurwuww urw ru rr rw u r r ruwrrwrww r r w uww wu wr ww wr w wuwwrww ww CLY CLY CLY gdom g g C L Y C L Y C L Y gimp gdom g gimp g gdom CLY CLY CL Y +              + =++                 (5) This equation (5) allows us to calculate the following figures. a) u u gdom is equal to the emissions embodied in Spanish final and intermediate products that are consumed by Spanish residents and is calculated as follows: u uu ur uw u uuuuruuwu gdom C L Y C L Y C L Y=++   (6) b) u r gexp is equal to the emissions embodied in Spanish final and intermediate products that are consumed by EU countries and is calculated as follows: expu uu ur uw r uururruwr g CLY CLY CLY=++   (7) c) u w gexp is equal to the emissions embodied in Spanish final and intermediate products that are consumed by non-EU countries and is calculated as follows: exp u uu ur uw w uuwurwu ww g CLY CLY CLY=++   (8) Therefore, the total emissions produced in region u, Spain, (PEu) are: exp exp u uu u u rw PE gdom g g= ++ (9) Additionally, equation (6) allows us to calculate the following items: d) u r gimp is equal to the emissions embodied in EU final and intermediate products that are consumed by Spanish residents and is calculated as follows: 8 u ru rr rw r ruurrurwu gimp CLY CLY CLY=++   (10) e) u w gimp is equal to the emissions embodied in non-EU countries final and intermediate products that are consumed by Spanish residents and is calculated as follows: u wu wr ww wwuuwruw wu gimp CLY CLY CL Y=++   (11) Then, the emissions footprint of region u, Spain, (EFu) is calculated as the total emissions embodied in all final and intermediate products consumed by Spanish residents: uuu u urw EF gdom gimp gimp= ++ (12) Furthermore, the difference between equations (9) and (12) is the so-called emission trade balance (ETB) which is the difference between the emissions actually produced in Spain (9) and the Spanish footprint (12). As mentioned before, the ETB is nothing more than the difference between the emissions embodied in exports (EEE) minus the emissions embodied in imports (EEI). exp exp ( ) exp exp uuu uuu u uu u r w u r w uuuu r wr w ETB PE EF gdom g g gdom gimp gimp g g gimp gimp EEE EEI =−= + + − + + = =+−−=− (14) A positive ETB value means that the EEE are larger than the EEI and, therefore, the country's emissions due to its national exported production are greater than the country's emissions provoked by its imported production. Meanwhile, a negative ETB value means that the emissions provoked by the national exported production are lower than the emissions embodied in imported production. Moreover, the total emissions embodied in trade for region u, Spain, (TETu) allow us to calculate the total emissions involved in the international trade of one region (u), including those from exports and imports of final and intermediate products. The TET allows us to determine the global emissions generated by the international trade of one country or area. The TET can be calculated as follows: 9 exp exp uuuu u r wr w TET g g gimp gimp EEE EEI=+++=+ (15) 2.3. Database The data used in this paper come from the World Input-Output Database (WIOD), as described in Dietzenbacher et al. (2013) and Timmer et al. (2015). This is a free database financed by the European Union and developed with the aim of analyzing the effects of globalization on trade patterns, environmental pressures and the socioeconomic development of a large group of countries. The data include world input-output tables for the 27 European Union countries and 13 other non-EU economies and also the corresponding national IO tables. The WIOD currently covers the period 1995-2013 and includes 35 industries and 59 commodities. However, data on energy and emissions have not been updated up to 2013 yet so the analysis has been carried out for the most recent year, i.e. 2009 that were available the environmental data. In fact, the precursor gases of tropospheric ozone considered have been: nitrogen monoxide (NOX), non-methane voltaic organic compounds (NMVOC), methane (CH4) and carbon monoxide (CO). 3. Results The analysis has been carried out in detail for 35 sectors, distinguishing the emissions embodied in the Spanish trade with the EU-27 and non-EU countries separately and considering the four ozone precursors (see Tables 1 and 2). The non-EU countries considered have been Australia, Brazil, Canada, China, India, Indonesia, Japan, Korea, Mexico, Russia, Turkey, Taiwan and United States. Insert here Table 1 Insert here Table 2 16 In light of the major findings mentioned above, the results are discussed by sectors. Firstly, the Agriculture sector is considered apart from the rest due to the importance for the Spanish economy. Then, the rest of the sectors are discussed distinguishing on one hand, those that are now implementing several measures for reducing emissions and on the other hand, those without current policy measures and therefore, some policy recommendations are provided. In both of cases, paper focus on new measures mitigation oriented. The Agriculture sector is one of the main polluting sectors considering the results for the four ozone precursors. This sector contributes notably to the increase in the emissions embodied in Spanish trade (TET) with EU and non-EU countries. In fact, the results show that the emissions embodied in Spanish final and intermediate products of the Agriculture sector exported (EEE) are higher than those imported from the EU countries (EEI) and therefore the ETB is positive. In the case of EU countries, the Agriculture sector contributes most of the emissions embodied in Spanish trade (TET) for NMVOC, CH4 and NOX. However, when non-EU countries are considered, the emissions embodied in the final and intermediate products of the Agriculture sector consumed by Spanish residents (EEI) are higher than for those produced in Spain and exported (EEE) to non-EU countries (with the exception of NMVOC emissions). These results show that the Agriculture sector should be the main focus of Spanish environmental policies. Firstly, the agricultural sector is the main methane emitter due to livestock farming. Mitigation policies in the agricultural sector would not only impact on emission levels but also on guaranteeing food resources to people. Additionally, it should be borne in mind that critical levels of O3 could damage plants and lead to a reduction in agricultural crops yields. For example, China -the main emitter globallyhas recently developed well-oriented measures in the agricultural sector to mitigate methane emissions. In particular, cleaner production technologies have been widely promoted, encouraging the improvement of farming practices, rational application of fertilizer and the adoption of diversified crop systems (Gan et al., 2011). Therefore, the Chinese Twelfth (2011-2015) and Thirteenth (2016-2020) Five-Year Plans have focused efforts on such good practices. Actions in the Agriculture sector are also necessary to avoid NOx emissions, as the results show. In response to the problems caused by nitrogen emissions from 17 agriculture, the European Union adopted the Nitrate Directive in 1991 and many countries started to develop polices to reduce nitrogen emissions. Taxes on fertilizers were used in some countries but were finally abandoned, as was the case for Austria, Finland and Norway. Although taxes might be an effective instrument in reducing nitrogen emissions, the problem of nitrate pollution from agriculture is a classic example of how difficult it is to address such diffuse pollution through ‘first-best’ instruments. Taxing nitrate itself is not an efficient solution since the problem which the instrument seeks to address – emissions to various media – is related to the application of nitrate in rather complex ways: the method of cultivation, the crop being cultivated (and the timing thereof), the type of soil and the weather. Instead of taxes, alternative policies have been implemented in some countries such as codes of good practice in the use of fertilizers. These codes are oriented to mitigate not only NOX emissions but also other ozone precursors, such as methane, NMVOC and CO. For example, the adoption of these codes of environmental practices has been a pressure measure in several industries such as Chemical sector (King and Lenox, 2000). Therefore, this practice should be considered by the Spanish Government due to the Spanish agricultural sector appearing as a key sector in the results obtained in Section 4. Additionally, there are other effective tools that can also exert pressure via consumers and the general public such as certification (for internal processes) and eco-labels (for final products). The adoption of these practices serves as a way to legitimize externally the firms' environmental behavior. When these tools are implemented, the group of leader companies are motivated to certificate their processes or/and final products and therefore, the rest of competitors try to imitate them in a short/medium term. These tools are not only effective in the case of the Agriculture sector but also in other sectors such as chemicals, basic metals or transport sector. Any case, literature highlighted the difficulty of creating self-regulation without explicit sanctions (King and Lenox, 2000). The EU authorities did not regulate the use of pesticide products from an environmental protection perspective until 2009 (European Union, 2009). However, these measures did not enter into force in Spain until 2012 (Spanish Royal Decree, 1311/2012). This regulation establishes the way that fertilizers should be handled and mixed in order to minimize emissions. Additionally, this regulation obliges the producers to manage the containers as if they were hazardous waste and details how they should discharge 18 polluted water resulting from washing the containers. Furthermore, the regulation limits the sale of these products only to those producers that exhibit the handler card and the treatments must be recorded in the exploitation register of each farm. Also, there is a system of inspection and sanction in the event that producers do not follow the regulation. Therefore, Spanish legislation is now going in the right direction considering the importance of the agricultural sector in terms of ozone precursor emissions as the results show in this paper. After recommending the measures mentioned above, a reflection on its feasibility should be done. Exports from agriculture sector shared 10% of total export up to 2013 - last available data (INE, 2016a and ICEX, 2012). This sector employed 4.3 % of total Spanish workers at the end of 2015 (INE, 2016 b) and contributed to 2.3 % of GDP up to 2014 (INE, 2016c). The profitable contribution of agriculture products to Spanish exports/imports balance could replace or even obscure the environmental objective of reducing the negative emissions of this sector. Transport sector (mainly inland and air transport), energy sector and refinery industry might receive great attention from Spanish authorities although mitigation policies linked with their activities were put in force in the past. Regarding policy agenda for the next years some policy recommendations are exposed below. Other sectors like Community services and the basic metals sector received less attention in the past. We refer to them at the end of this section. The Inland transport sector is relevant for NOx emissions, and, to a lesser extent, the air and water transport sector. In the case of inland transport, the replacement of traditional materials by lightweight polymers in automobile construction reduces the fuel consumption per kilometer and therefore the emissions. Also, the inclusion of chemical additives in fuels such as ad-blue (i.e., urea) can improve efficiency and reduce the generation of pollutants. Ferrón-Vilchez et al. 2015) conducted a mini case study focused on a Spanish road freight transport industry. They founded that managers decided to introduce ad-blue to reduce NOx in the exhaust system of trucks benefiting from economies of scale in purchasing this component internalisating environmental costs. In fact, thanks mainly to chemicals, a current car emits just 10% of the pollutants emitted in 1950 (Interempresas, 2009). 19 Together with enhancing materials used in a vehicle’s design, lower levels of NOX emissions generated by road transport could be achieved using two pillars. The first is the right use of NOX absorber devices, mainly in cars powered by diesel. Such devices allow the reduction of oxides of nitrogen (NO and NO2) emissions from a lean-burn internal combustion engine by means of adsorption. Although these devices are becoming common in developed countries they are not generally in widespread use globally. The second pillar consists of increasing the use of electric vehicles or hybrid vehicles (powered with electricity, petrol and natural gas or hydrogen), all of which are lower emitters. Air transport should also be analyzed for its contribution to NOx emissions. Aircraft emit gases and particles directly into the upper troposphere and lower stratosphere, at altitudes located between 9 km and 13 km, altering their composition and concentration of gases. The harmful effects of these emissions have largely been studied as one of the ozone precursors (MAGRAMA, 2015) promoting global warming and photochemical smog (Berend, 2015). Improvements in some of the structural components of aircraft have shown that it is possible to reduce NOX emissions by 50%. This has happened with the SMART Fixed Wing Aircraft (SFWA) program aimed at developing new products related to the wings of the aircraft (CSJU, 2008). The leading aircraft manufacturers - Boeing Co. and Airbus Grouphave begun to evolve their most successful models in that direction -B737 (Max version) and A320 (a Neo version) (Airbus Group, 2015). The reduction of NOX emissions associated with air transport would be greater if, in addition to the introduction of this improvement to the wings of aircraft, airports imposed stricter emissions requirements on the aircraft that operate out of them. Another sector that might receive political attention is the energy sector. The Electricity sector is particularly relevant for the NOx emissions involved in Spanish trade. Due to foreign energy dependency and in order to avoid larger NOX emissions, higher shares of non-fossil fuel sources in the energy matrix might be a solution. For many countries, mainly in the EU and of course in Spain, this would imply a more intensive use of renewable energy sources due to legal decisions against nuclear power deployment. However, this is not the only recommendation in the field of energy policy. Literature shows that there is room for enhancing energy efficiency in terms of energy conversion 20 from primary to secondary energy. This is true not only for developed countries but also for developing ones (Chikkatur, 2008; Remme et al., 2011). The Refinery industry is also revealed as another driver of NMVOC and CO emissions. This industry has already shown significant results in reducing emissions of some greenhouse gases. This has been the case of SOX for example by the installation of desulphurization devices in refinery plants. These plants recover the sulphur and convert it into a co-product that is then sold to the copper industry, and used for the extraction of this mineral by the solvent extraction method (SX-EW). In the case of emissions of ozone precursor agents, recommendations for the refining industry should go in two directions; improving energy efficiency and changing the type of fuel used. In the case of Spain, since the 1990s, refiners have reduced energy consumption by 16% (Spanish Environment Department, 2004). However, the most important change is the replacement of the use of liquid fuels by gaseous fuels. In traditional technology, the gases generated in the refining process are used as fuel in the internal processes of industrial plants. Now the technology allows the refining industry to use the fuel gas generated in the refining oil process for powering the turbines and electricity without using water. Therefore, it is now possible to capture and reduce the residual fuel gas by using it as a by-product. By combining the refinery fuel gas reformulated without steam or water, the new technology reduces emissions of NO and NO2 among other gases (Interempresas, 2015). This rute should become a priority action. The Community services sector should receive policy attention because it not only provokes the main methane emissions besides the Agriculture sector, but also other relevant pollutant emissions such as landfill gas (LFG). This gas is created from the decomposition of organic matter in landfills for municipal solid waste (MSW). This gas consists of approximately 99% CH4 and CO2, and a small amount of different organic compounds to methane (NMVOC). Therefore, the policy regulation should be oriented to capture, convert and use this gas as an energy source. For example, the use of LFG helps to reduce odours and other hazards associated with LFG emissions and helps prevent the methane from migrating into the atmosphere. It also prevents the local smog increasing and mitigates global climate change. 21 The Basic Metals sector is relevant when considering the CO emissions embodied in Spanish trade with the EU. In fact, this industry includes two important activities: cement and lime production. Cement is a basic material for building and civil engineering construction and lime and its derivatives are used as a binder in building and construction. This explains why the construction sector appears as an important emitter of NMVOC. As suggested by JRC (2013), some policy recommendations could be put into force. Starting with the cement industry, a technique to reduce the industry's energy use and emissions, expressed per unit mass of cement product, is to reduce the clinker content of cement products. This can be done by adding fillers and additives, for example, sand, slag, limestone, fly ash and pozzolana, in the grinding step. This technique would reduce emissions to air. In Europe, the average clinker content in cement is 80–85%. Additionally, the electrical energy use can be minimized in the Basic metal sector through the installation of power management systems and the utilization of energyefficient equipment such as high-pressure grinding rolls for clinker comminution and variable speed drives for fans as well as, in some cases, replacing old raw material mills with new mills. These power management systems could result in the saving of resources and reduction of emissions and waste. The clinker burning process is usually optimized to reduce the heat consumption, to improve the clinker quality and to increase the lifetime of the equipment (the refractory lining, for example) by stabilizing process parameters. Reduction of emissions, such as NOx and dust, is a secondary effect of this optimization. Focusing now on the Lime industry, because of the wide range of exhaust gas conditions, a variety of dust collectors are used, including cyclones, wet scrubbers, fabric filters, electrostatic precipitators and gravel bed filters. For example, typical cyclones remove about 90% of the dust from lime kilns. Therefore, dust collectors should be recommended. All previous policy recommendations would impact on the supply side of markets, forcing industries to adopt mitigation practices. But the demand side of the market might also receive attention in order to contribute to mitigate the ozone precursor emissions. Obviously, action is needed to enhance the general knowledge about the problems caused by ozone, for example the ozone precursor footprint could be included 22 in food products labels in order to drive consumer decisions and change their patterns towards cleaner products. 5. Concluding remarks The main conclusions that can be drawn from the analysis of the results are the following: a) The Agriculture sector seems to be the most relevant one considering the total emissions embodied in Spanish trade within the EU. This sector contributes 75%, 58% and 26% of CH4, NMVOC and CO emissions respectively. Despite of this, mitigation measures feasibility would depend on the way in which a debate about environmental versus economic objectives is solved. b) In terms of the ETB, Spanish trade with the EU shows that the NMVOC and CO emissions embodied in Spanish exports are much higher than those embodied in Spanish imports from this area. Again, the Agriculture sector seems to be related to these emissions, as is the Basic metal sector. Current regulation in relation to the use of pesticide products in Spain seems to deal with this problem. However, more active environmental regulations focused on the Agriculture sector are needed, such as the introduction of codes of good practice in the use of fertilizers and the promotion of cleaner production technologies. c) The most important sector that contributes to the total emissions embodied in Spanish trade within the non-EU area is the Coke, Refined Petroleum sector which accounts for 59% and 53% of CO and NMVOC emissions respectively. Additionally, the Agriculture sector is also relevant in terms of CH4 (44%) and NOX (25%) emissions. In terms of the ETB, Spanish trade with the non-EU area shows that Spanish regulation should focus on the national consumption pattern in order to reduce Spanish pressure on ozone precursor emissions worldwide, especially in the case of CO and CH4 emissions. Certification or eco-labels mentioned have room for further deployments in Spain. 23 Acknowledgements The first two authors are very grateful for the financial support received from Project SEJ-132 by the Andalusian Regional Ministry of Innovation, from the Roger Torné Foundation through the Chair on Energy and Environmental Economics at the University of Seville, and from the ECO2014-56399-R Project of the Spanish Ministry of Economy and Competitiveness. They also acknowledge the funding provided by the Universidad Autónoma de Chile (Chile) and from the project Nº 018/FONDECYT/16 of Chile’s Department of Education. The authors are also grateful for the useful comments and anonymous reviewers from the Journal of Cleaner Production. The views expressed in this article are the sole responsibility of the authors and should not be attributed to the European Commission or any of its services. References Airbus Group (2015). New Sharklets and A320 Neo. Último Acceso: 16/febrero/2015. http://www.airbus.com/aircraftfamilies/passengeraircraft/a320family/technology-andinnovation Arto, I.; Genty, A.; Rueda-Cantuche, J.M.; Villanueva, A. and Andreoni, V. (2012). Global Resources Use and Pollution, Volume 1 / Production, Consumption and Trade (1995-2008). JRCc scientific and policy reports. European Commission. Berend, N. (2015). Contribution of Air Pollution to COPD and Small Airway Dysfunction. Official Journal of the Asian Pacific Society of Respirology; 2. Accessed November. Brizga, J; Feng, K. and Hubacek, K. (2016). Household carbon footprints in the Baltic States: A global multi-regional input–output analysis from 1995 to 2011. Applied Energy, In Press, Corrected Proof, Available online 17 February 2016.Chen, C.Q. & Zhang, B. (2010). Greenhouse gas emissions in China 2007: Inventory and input-output analysis. Energy Policy, 38: 6180-6193. Chikkatur, A.P. (2008). A resource and technology assessment of coal utilization in India; Pew Center on Global Climate Change Working Papers: Arlington, United States. Dietzenbacher, E., Pei, J., and Yang, C. (2012). Trade, production fragmentation, and China's carbon dioxide emissions. Journal of Environmental Economics and Management 64 (1), pp. 88–101. Dietzenbacher, E., Bart, L. Stehrer, R., Timmer, M. and Vries, G. (2013). The construction of World Input-Output Tables in the WIOD Project, Economic Systems Research, 25(1), pp. 71-98. EEA (2014). European Environment Agency. Air quality in Europe — 2014 report. Denmark. European Union (2009). Reglamento (CE) n.º 1107/2009, del Parlamento Europeo y del Consejo, de 21 de octubre de 2009, relativo a la comercialización de productos 24 fitosanitarios y por el que se derogan las Directivas 79/117/CEE y 91/414/CEE del Consejo, y la Directiva 2009/128/CE, del Parlamento Europeo y del Consejo, de 21 de octubre de 2009, por la que se establece el marco de la actuación comunitaria para conseguir un uso sostenible de los plaguicidas. Ferrón-Vilchez, V., De la Torre Ruiz, J.M. and Ortiz de Mandojana (2015). How much would environmental issues cost? The internalisation of environmental costs in the European transport industry. N. Environmental Engineering and Management Journal, Vol. 45(9), 2149-2162. Homma, T., Akimoto, K. and Tomoda, T. (2012). Quantitative evaluation of time-series GHG emissions by sector and region using consumption-based accounting. Energy Policy, Volume 51, pp. 816-827 Hong, J; Shen, G.Q.; Guo, S; Xue, F. and Zheng, W. (2016). Energy use embodied in China׳s construction industry: A multi-regional input–output analysis. Renewable and Sustainable Energy Reviews, 53, 1303-1312. INE (2016 a). Mercancías importadas y exportadas por sección, capítulo arancelario y año. http://www.ine.es/jaxi/Tabla.htm?path=/t41/a121/a1998/l0/&file=x10031.px&L=0 INE, (2016 b). Encuesta de población activa. http://www.ine.es/dyngs/INEbase/es/operacion.htm?c=Estadistica_C&cid=1254736176 918&menu=resultados&idp=1254735976595 INE, (2016 c). Contabilidad nacional de España. http://www.ine.es/jaxi/menu.do?type=pcaxis&path=%2Ft35%2Fp008&file=inebase&L =0 Instituto Español de Comercio Exterior (ICEX); La empresa exportadora española 2007-2010 http://www.icex.es/icex/es/navegacion-principal/todos-nuestros-servicios/informacionde-mercados/estudios-de-mercados-y-otros-documentos-de-comercioexterior/4591630.html Jiang, W.Q.; Li, J.S; Chen, G.Q; Yang, Q; Alsaedi, A; Ahmad, B. and Hayat, T. (2016). Mercury emissions embodied in Beijing economy. Journal of Cleaner Production, 129, 15, 134-142. Kanemoto, K; Moran, D.; Lenzen, M. and Geschke, A. (2014). International trade undermines national emission reduction targets: New evidence from air pollution. Global Environmental Change, 24, 52-59 King, A. A., Lenox, M. J. (2000). Industry self-regulation without sanctions: The chemical industry's responsible care program. Academy of Management Journal, 43(4), 698-716. 25 Leontief, W. (1970). Environmental repercussions and the economic structure: an inputoutput approach, Review of economics and statistics, 52 (3), pp. 262-271. Marin, G; Mazzanti, M. and Montini, A (2012). Linking NAMEA and Input output for ‘consumption vs. production perspective’ analyses: Evidence on emission efficiency and aggregation biases using the Italian and Spanish environmental accounts. Ecological Economics, 74, 71-84. Miller, RE. and Blair, PD. (2009). Input-output analysis, Foundations and Extensions, 2ª edición. Cambridge University Press, Cambrigde. Oksanen, E; Pandey, V; Pandey, A.K., Keski-Saari, S; Kontunen-Soppela, S. and Sharma, C. (2013). Impacts of increasing ozone on Indian plants. Environmental Pollution, 177, 189-200. Rypdal, K; Berntsen, T; Fuglestvedt, J.S.; Aunan, K.; Torvanger, A.; Stordal, F; Pacyna, J.M. and Nygaard, L. 2005. Tropospheric ozone and aerosols in climate agreements: scientific and political challenges. Environmental Science & Policy, 8 (1), 29-43. Rueda-Cantuche, JM. and Amores, AF. (2010). Consistent and unbiased carbon dioxide emission multipliers: Performance of Danish emission reductions via external trade, Ecological Economics, 69, pp. 988-998. Sato, M. (2012). Embodied Carbon in Trade: A Survey of the Empirical Literature. Centre for Climate Change Economics and Policy Working Paper no. 89 UNECE (1999). Gothenburg Protocol to Abate Acidification, Eutrophication and Ground-level Ozone. United Nations Economic Commission for Europe (UNECE) Convention on Long-range Transboundary Air Pollution (LRTAP). www.unece.org/env/lrtap/multi_h1.htm (accessed 18 March 2010). Interempresas (2009). Available at http://www.interempresas.net/Quimica/Articulos/27913-La-quimica-en-Espana-reduceun-63-por-ciento-sus-emisiones.html) Accesed november, 2015. Interempresas (2015). Available at http://www.interempresas.net/Quimica/Articulos/137456-Cepsa-aumenta-eficienciareduce-emisiones-refineria-Gibraltar-San-Roque-mediante.html) Accessed november, 2015. JRC (2013). JOINT RESEARCH CENTRE. Institute for Prospective Technological Studies. Sustainable Production and Consumption Unit. European IPPC Bureau . Best Available Techniques (BAT) Reference Document for the Production of Cement, Lime and Magnesium Oxide. Available at: http://eippcb.jrc.ec.europa.eu/reference/BREF/CLM_30042013_DEF.pdf.Accessed november, 2015