The Intra-EU Value Chain: An Approach to Its Economic Dimension and Environmental Impact

economies Article The Intra-EU Value Chain: An Approach to Its Economic Dimension and Environmental Impact Óscar Rodil-Marzábal * and Hugo Campos-Romero   Citation: Rodil-Marzábal, Óscar, and Hugo Campos-Romero. 2021. The Intra-EU Value Chain: An Approach to Its Economic Dimension and Environmental Impact. Economies 9: 54. https://doi.org/ 10.3390/economies9020054 Academic Editor: Ralf Fendel Received: 20 February 2021 Accepted: 31 March 2021 Published: 7 April 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). ICEDE Research Group, Departamento de Economía Aplicada, Universidade de Santiago de Compostela, Praza do Obradoiro, 0, 15705 Santiago de Compostela, A Coruña, Spain; hugo.campos.romer[email protected] *Correspondence: oscar[email protected] Abstract: This paper aims to analyze the economic dimension and environmental impact of intra- EU value-added generation linked to global value chains (GVCs) through input-output analysis. For this purpose, information has been collected from TiVA (Trade in Value Added, OECD) and Eora databases for the years 2005 and 2015. From an economic perspective, the results point to a strengthening of the value-added generated within Factory Europe. From an environmental perspective, all EU28 members have reduced their exports-related impacts in intensity-emissions terms, but not all of them in the same degree. An approach to the environmental Kuznets curve (EKC) has also been carried out through a panel data model. The results show a positive impact of the participation in intra-EU value chain (Factory Europe) on CO 2 emissions per capita. Further, an inverted U-shaped curve for CO 2 emissions is found for the period 2005–15. In this sense, European economies with lower development levels (many Eastern and Southern countries) seem to be still on the rising segment of the curve, while the more developed ones seem to be on the decreasing segment. These results highlight the need to design global monitoring and prevention mechanisms to tackle growing environmental challenges and the need to incorporate specific actions associated with the GVCs activity. Keywords: EU28; intra-EU trade; global value chains; environmental impact; Factory Europe; environmental Kuznets curve; pollution haven hypothesis; circular economy 1. Introduction Throughout the world, society is increasingly aware of environmental problems and the need to adopt changes leading to a more sustainable way of life. In response to this need, governments have implemented various measures to address environmental issues. On a global scale, the United Nations agreed in 2015 on a plan to progress towards 17 sustainable development goals (SDGs), under the 2030 Agenda for Sustainable Development. In this respect, the environmental dimension is present in several of these SDG targets. Furthermore, the European Union (EU) is promoting the so-called circular economy (CE) strategy with the aim of minimising the generation of waste and making the best use of the available resources. In fact, since 2015 various CE measures have been proposed and implemented. However, none of them mentions the impact of international fragmentation of production on the environment despite the relevance of production linkages within the so-called Factory Europe, especially between EU members. Indeed, it is noteworthy that in the EU integration process intra-regional trade flows have always been much more intense than the extra-regional ones. Concerning international economy, the role of global value chains (GVCs) is particularly relevant as a way of organising production processes, not only by fragmenting them into different stages, but also by locating these stages in different countries. However, as Baldwin (2006,2012) suggests, the term “global” is relative since many of these value chains tend to have a strong regional orientation. Economies 2021,9, 54. https://doi.org/10.3390/economies9020054 https://www.mdpi.com/journal/economies Economies 2021,9, 54 2 of 17 According to this author, there are three large commercial regions in the World, namely Factory Europe, Factory Asia, and Factory America. Among these, Factory Europe is the most intensive in terms of intra-regional trade. In this regard, EU enlargements have usually been accompanied not only by an intensification of intra-EU trade but also by processes of intra-EU industrial relocation, as in the last 15 years with the Central and Eastern European Countries (CEECs) (Fritsch and Matthes 2017). It should be noted that while extra-EU exports of goods and services accounted for more than 14% of global exports in 2019, intra-EU exports accounted for more than 17% (Data extracted from the World Trade Organization database). Furthermore, participation in GVC also generates innovation processes due to the need of being competitive in international markets. A potential innovation pathway related to GVCs is the so-called environmental upgrading, which is defined as any improvement implemented by a firm that reduces its emissions and energy consumption. Therefore, it is necessary to include the environmental effects of intra-EU value creation processes related to GVCs participation in policies oriented to the promotion of the CE. This would allow a better assessment of these programmes and the consideration of measures to reduce the impact of trade on the environment. In this sense, the input–output (IO) methodology is useful for analyzing the environmental impact of trade. The aim of this paper is to analyze the economic dimension and environmental impact of the intra-EU value-added linked to GVCs by using an IO approach (value-added trade perspective). This approach focuses on the intra-EU value chain linked to both extra-EU and intra-EU exports. In this respect, the main research subject is twofold. Firstly, to analyze the changes in the EU28 fragmentation of production, with a special focus on the Central and Eastern European enlargement. Here, the underlying hypothesis concerns the economic dimension and the evolution of the intra-EU value chain. In this sense, the paper aims to test whether this regional value chain has strengthened in the analyzed period (2005–15), when the integration of Eastern European countries took place. Secondly, the paper aims to analyze the environmental impact associated with the intra-EU value-added creation processes along the regional value chain (Factory Europe). Here, the underlying hypothesis concerns the differences between the economic dimension (economic development; value-added exports) and environmental impact (CO 2 emissions) of the intra-EU value chain participation of European countries. In this respect, the paper aims to verify the existence of asymmetries between these two perspectives in view of some theoretical approaches, providing new insights and evidence to the economyenvironment link. In this sense, the main novelty of this paper is to bring a double perspective to the study of the environmental impact of intra-EU trade flows. First, it focuses on valueadded flows linked to the intra-EU fragmentation of production through the analysis of forward GVC participation. Second, it addresses the analysis of emissions associated with intra-regional flows in an advanced context of economic integration. In order to address these issues, an analysis of intra-EU value-added generation from a GVC approach (forward GVC participation) is carried out using IO methodology with environmental extension for the years 2005 and 2015. The selection of the year 2005 is justified by the fact that most CEECs joined the EU from 2004 onwards. In addition, a panel data estimation is performed in order to verify the fulfilment of the environmental Kuznets curve hypothesis. The statistical information used comes from Eora26 database, a multiregional IO (MRIO) model that includes 186 countries and 26 homogeneous sectors for all of them. Moreover, TiVA (Trade in Value Added, and Organization for Economic Co-operation and Development (OECD)) database is used to address specific trade issues in terms of value-added. Data collected at the World Bank (population, and GDP deflator) have also been used. The paper is structured as follows: Section 2introduces the main theoretical aspects related to GVCs and its environmental impact. Section 3presents the methodology frame- Economies 2021,9, 54 3 of 17 work employed in this paper, based on IO analysis. Section 4presents and discusses the main results. Finally, Section 5offers some conclusions and possible future research. 2. Literature Review Early work on the analysis of international flows and the environment was characterized by static analyses of the impact of trade on the environment as well as the impact of environmental policies on trade performance. This approach has led to uncertain and divergent results. In this line, some authors have found that the adoption of environmental protection policies worsens trade performance due to competitiveness losses. According to these findings, there seems to be a trade-off between trade and environmental gains (Siebert 1974 ,1977). For example, the recent study by Fang et al. (2019) finds that China’s regions with tighter environmental regulations have reduced their export dynamism. However, other proposals show opposite results by using alternative assumptions. For example, Pethig (1976) notes that specialization in emission-intensive goods can result in welfare losses and that the adoption of environmental protection measures can lead to strengthening a country comparative advantage. In turn, Stavropoulos et al. (2018) find an inverted U-shaped relationship between the rigidity of environmental regulation and industrial competitiveness, suggesting that the linkage between environmental regulation and trade performance is complex and non-linear. Ansari and Khan (2021) analyze the impact of trade openness on environmental quality (in terms of the ecological footprint) in a set of Asian countries between 1991 and 2016. The authors attempt to differentiate the three classic effects considered in the trade-environment binomial: scale, technical and composition effects. On the one hand, the results verify that the scale effect, referred to the volume of trade, deteriorates environmental quality. On the other hand, the technical and composition effects, referred to the production technology used, and the products manufactured, respectively, reduce environmental deterioration. These results are in line with the work of Antweiler et al. (2001) and Chou and Wang (2020), among others. Later theories such as the pollution haven hypothesis (PHH) or the environmental Kuznets curve (EKC) became increasingly relevant. On the one hand, the PHH establishes that in the presence of international disparities in environmental regulations and in the presence of free movement of capitals, industries affected by tight environmental regulation will relocate their environmentally harmful activities to countries with softer environmental standards. As with the early work on the trade–environment nexus, some authors found evidence that holds this hypothesis (Low and Yeats 1992;Lucas et al. 1992), while others pointed out that international differences in environmental regulation are not determinant to relocation (Dean 1992;Grossman and Krueger 1991). Despite these discrepancies, PHH is still the subject of much recent research. Some of these studies verify the fulfilment of this hypothesis under certain conditions or for specific sectors that are particularly sensitive to environmental regulations, such as electronic equipment (Copeland and Taylor 2004;Kellenberg 2009). Regardless of whether the offshoring decision is based on less restrictive environmental regulation, the overall impact on the environment is the one stated by the PHH (Guzel and Okumus 2020;Zhao et al. 2020). Manufacturing tasks generally tend to relocate to developing countries, which at the same time tend to have lower environmental constraints than developed countries. As a result, new producers are unlikely to adopt the same environmental and eco-design practices applied in the country of origin. The environmental damage resulting from trade is particularly relevant in those activities that are extremely fragmented and organized through the so-called GVCs, extensive international networks that involve the participation of several agents located in different countries for the manufacture of a product (Arndt and Kierzkowski 2001;Gereffi 2005, 2014). These large production networks involve large movements of goods, increasing the environmental impact of trade in two ways: production in countries with weaker Economies 2021,9, 54 4 of 17 environmental regulation (environmentally harmful production) and increasing flows of international transport, which involve emissions and materials use. On the other hand, the EKC states that as a country develops its impact on the environment increases up to a certain level of GDP per capita, from which further development will reduce environmental degradation. As with the PHH, some studies verify (Jalil and Mahmud 2009) while others are less conclusive about the fulfilment of the EKC (Ahmed et al. 2016;Destek et al. 2018;Kleemann and Abdulai 2013). It should be noted that not all production relocation processes take place from developed countries to Asia or other developing areas. The integration of Eastern European economies into the EU since 2004 has favored the relocation of certain activities from the Central European economies to the new Member States. However, this relocation has taken place more as outsourcing processes than through the establishment of subsidiaries (Marin 2006). Moreover, according to Geishecker (2006), most of the tasks initially offshored to Eastern Europe are labor-intensive, although more service-related tasks have also started to be relocated (Sass and Fifekova 2011). The work by Mandras and Salotti (2020) analyzes the potential effects of the eventual integration of the Balkan countries into the EU from an economic point of view. These authors find that integration has positive effects, but uneven across sectors. Particularly noteworthy is the potential outcome derived from participation in GVCs achieved through integration into the Union. This integration would mainly benefit participation in the regional chain of Factory Europe. However, these authors do not address issues arising from the possible relocation of production processes as a consequence of this integration. According to the PHH, since the early years of Eastern European economies’ integration into the EU to the harmonization of their environmental standards (Despite the number of member countries, there have been no relevant cases of non-compliance concerning European environmental regulation. According to Börzel and Buzogány (2019), the success in the harmonization of environmental legislation is mainly due to two factors. On the one hand, to the strategy broadly followed by European institutions to ensure compliance with the European common regulation. On the other hand, environmental policy is not particularly rigorous at the EU level and seeks to modify the legislation of each member state to avoid intra-EU irregularities), it could be expected a worse environmental performance of these countries compared to the rest of the EU. At the same time, according to the EKC, their trade integration into the EU should favor higher growth rates among Eastern European countries. Assuming that due to their low development within the EU context they are still in the rising section of the EKC, an increase in emissions generated by these countries may be expected as a result of their integration into the EU. In this sense, the work by Pablo-Romero et al. (2017) finds an inverted U-shaped curve—as established by the EKC—for the transport sector, which accounts for around a quarter of total emissions. Furthermore, these authors also state that the inflexion point has not yet been reached. The study by Leitão and Balogh (2020), focused on the agricultural sector, also finds that the higher growth and productivity rates in the sector lead to more environmental concerns. The work by Al-Mulali et al. (2015) obtains interesting results by finding a positive relationship between GDP growth and CO 2 emissions, while finding a relationship of opposite sign between trade openness and emissions. However, if a higher degree of trade openness induces higher increases in GDP, following this rationale international trade would generate an increase in emissions through increases in GDP. Shahbaz et al. (2019) estimate the fulfilment of the EKC hypothesis in Vietnam. First, they verify that factors such as energy consumption, economic structure, FDI and urbanization positively influence CO 2 evolution. Regarding EKC compliance, the study is non-deterministic, finding an N-shaped evolution in long-term projections of CO 2 emissions as a function of economic growth. The authors conclude that, since economic growth causes an increase in CO 2 emissions, the adoption of clean technologies should be considered by the government as a preventive policy measure. Economies 2021,9, 54 5 of 17 Regarding the Eastern European economies, there is a lack of literature analyzing whether this region has become a sort of pollution haven within the EU. From this scarce literature it is not possible to draw an unequivocal conclusion on this issue. For example, Martínez-Zarzoso et al. (2017) attempts to analyze the relationship between the level of stringency of environmental regulation and intra-EU trade flows, their results hardly support the PHH. However, other studies point to opposite conclusions. In this sense, Bagayev and Lochard (2017) find that as EU member states advance in environmental regulation, they also increase their imports of emission-intensive goods from both, Asia and emerging European countries. These authors also highlight the intensive use of fossil fuels in these regions: more restrictive environmental regulation in some territories appears to generate greater environmental impacts in others, hindering environmental improvement on a global scale. This result raises questions about the need for global coordination mechanisms to effectively achieve a sustainable future. The research conducted by Tsagkari et al. (2018) analyzes the link between the rapid economic growth of the Polish country following its trade opening and European integration process and the environmental effects linked to these processes. Their results point out to a significant increase in CO 2 emissions directly linked to the processes of integration and trade liberalization. Ho and Iyke (2019) analyze the relationship between trade openness and its effects on CO 2 emissions for a set of Central and Eastern European countries. Their results are consistent with the EKC since they find a negative relationship between emissions and trade openness—which is linked to economic growth—in the long term. These authors also point out that there is a limit to the reduction of environmental impacts from trade openness and therefore, beyond a certain point, the positive effects could be reversed. This paper aims to contribute to fill the gap in the literature by analyzing the environmental impact of intra-EU production linkages in a context of economic integration and international fragmentation of production. In this sense, it seeks to answer the research question of whether the integration of EU economies, especially in the case of Eastern European countries, has led to changes in the environmental impact (measured in CO 2 emissions) of their intra-EU linkages (measured in trade in value-added). 3. Data, Methods, and Model Specification The IO framework introduced by Leontief (1937,1951) allows the analysis of the intersectoral relations of an economy. Currently, there are some IO projects oriented to the analysis of international trade, especially useful for analyzing trade in value-added, of great utility in the GVC framework. The IO tables developed for this purpose include not only intersectoral linkages within an economy, but also those of each domestic sector with each foreign sector. These analytical tools are generally entitled MRIO models (Lenzen et al. 2012,2013;Timmer et al. 2015). The data for this paper were collected from two open-access sources. On the one hand, information concerning the value-added generation at the intra-EU level was obtained from TiVA database developed by the OECD. The specific information for the analysis of production chains was obtained by operating on the basis of the “forward participation in GVCs” indicator for the European level (This indicator provides the value-added that has been generated in an EU28 country (country A) and subsequently transformed and exported by another member country (country B) to another EU or non-EU country (country C), expressed as a percentage of total gross exports by country A. This is a measure of the complex production linkages within the EU28 regional value chain. The product of this indicator by the total gross exports results in the forward participation as an export flow). On the other hand, the data for the environmental analysis of intra-EU production linkages comes from the MRIO database Eora26—an open access database for researchers. Eora26 provides a set of harmonized MRIO tables for 190 countries and 26 sectors for the period 1990–2015. There is a set of satellite accounts accompanying the economic accounts, Economies 2021,9, 54 6 of 17 which provide information on sectoral emissions for each economy. In this work, CO 2 emissions have been selected due to the information availability for all European countries. In accordance with statistical availability, the period selected for the analysis is from 2005 to 2015—the last year available in Eora. In the following, the mathematical procedure for the calculation of the forward participation from the MRIO framework, measured not as a percentage, but as an export flow of value-added, is presented. The procedure for the transformation of export flows measured in monetary terms into trade flows measured in CO 2 emissions is also provided. As the analysis is carried out at a country level, the sectoral dimension is excluded in the formulation. For a set of c countries, Tc×c is defined as the intermediate transactions matrix, Fc×c as the final demand matrix, and Xc×1 as the total output vector. The basic IO entities are as follows: X=Ti +Fi;A=Tˆ X−1;L=(I−A)−1(1) where i is a unity vector used to perform the row or column sums required (In case a row-wise sum is required, i will be of c× 1 dimension while in case a column-wise sum is required, i will be of 1 ×c dimension). Ac×c represents the technical production coefficients matrix and ˆ X−1 c×c is a matrix whose diagonal contains the elements of the total output vector; the remaining elements are null—all elements with a circumflex accent represent the diagonalization of the original vector. Finally, Lc×c represents the Leontief inverse matrix and Ic×cthe identity matrix. Following Koopman et al. (2014), it is possible to obtain the total output by breaking down each element according to where it is actually absorbed: X∗=LF (2) where X∗ is a total output matrix of c×c dimension whose diagonal elements represent the total output generated in the economy of reference and absorbed in the same economy. The non-diagonal elements represent international trade. Let Xexp c×c be the matrix X∗ modified so that the diagonal elements have been removed. Each non-diagonal element of Xexp represents the gross exports from one country to another. A country’s total gross exports, EXPg c×1are defined as the row sum of Xexp. EXPg=Xexpi(3) In order to obtain the forward participation measured as an export flow of valueadded it is first required to obtain the value-added coefficients, v1×c from the technical production coefficient matrix. v=u−iA (4) where u1×c is a vector composed of “ones”. The product of the value-added coefficients by the total output gives the value added generated by each economy, which is equivalent to GDP. The forward participation in the GVCs of value-added generated within the EU28 and subsequently exported by any member country to any country in the world, FPEU c×1 , is obtained as follows: FPEU =ˆ vBEUiXexpi(5) In the expression 5, BEU represents the inverse Leontief matrix modified so that it only includes the external interrelations of the EU28 member countries. To obtain the forward share in terms of emissions, the vector of total CO 2 emissions, E1×c is used as a starting point. Emission intensity, e1×c, is defined as the ratio of CO2emissions to total output: e=Eˆ X−1(6) The variable e provides information on the CO 2 emissions emitted in units of mass for each monetary unit produced. In this regard, this variable is used in the same way as Economies 2021,9, 54 7 of 17 an output multiplier. Therefore, to obtain the emissions linked to domestic value-added exports, µc×1, it needs to proceed as follows: µ=ˆ evFPEU (7) where ev is the element-wise product between the vectors e and v , being ˆ ev its diagonalization. In addition to the descriptive analysis, in order to achieve robust results, an econometric model is proposed to test the EKC hypothesis, using country-level panel data for EU28 member states from 2005 to 2015. The proposed model specification is as follows: ln(CO2)it =β0+β1GDPit +β2GDP2 it +β3DVAFPit +εit (8) where the subscripts “i” and “t” refer to the cross-section (countries) and time-period (years) units, respectively; ln(CO 2 ) is the dependent variable and refers to the natural logarithm of CO 2 emissions per capita; GDP is the real GDP per capita, GDP 2 its square, and DVAFP is the domestic value added per capita oriented to forward participation in the intra EU value chain (Factory Europe). According to the EKC hypothesis, the sign of β1 parameter is expected to be positive, as CO 2 emissions should increase with the production level proxied by real GDP per capita; also, the sign of β2 parameter is expected to be negative. If the hypothesis is verified, higher income countries should have a better environmental performance (measured by CO2emissions per capita) than the lower income ones. The turning point can be obtained as a derivative of expression (8) with respect to GDP and equating the derivative to 0. dln(CO2) d(GDP)=β1+2β2GDP =0 GDP =−β1 2β2 (9) It should be noted that, unlike other empirical models that analyze the relationship between economic development and CO 2 emissions, a variable related to the forward participation in intra-EU value chain have been included to test the influence of intra- EU production linkages over CO 2 emissions. In this sense, and in line with the PHH, the relocation of activities within Factory Europe may be leading to an increase in CO 2 emissions per capita of the countries involved. Therefore, the sign of β3 parameter is expected to be positive, since CO 2 emissions should be increasing as countries, many of them Eastern European, increase their participation in the intra-EU value chain. Data used have been collected in the databases listed above (TiVA, Eora26). Table 1 shows the main statistics of the variables included in the regression. Table 1. Summary statistics of the variables. Variable Mean Median S.D. Min Max Ln(CO2) 2.03 2.03 0.396 1.31 3.33 GDP 2.68 ×1042.04 ×1041.80 ×1043.52 ×1039.62 ×104 GDP21.04 ×1094.18 ×1081.41 ×1091.24 ×1079.26 ×109 DVAFP 1.78 ×1031.08 ×1032.31 ×103123 1.50 ×104 Note: “Ln(CO 2 )” refers to the CO 2 emissions per capita (in log), “GDP” refers to the real GDP per capita, and “DVAFP” refers to the domestic value-added per capita oriented to intra-EU forward participation in GVC. Source: Own elaboration based on Trade in Value Added (TiVA), Eora26, and the World Bank. Regarding the model estimation, the use of the Pooled Ordinary Least Squares (POLS) method by a log-level equation is used to avoid multicollinearity problems, therefore a unit change in the independent variables is expected to change de dependent variable by 100 ·βj percent. The model estimation results using GRETL software are presented and discussed in Section 4.2. Economies 2021,9, 54 8 of 17 4. Results and Discussion 4.1. Descriptive Analysis: Facts and Trends Based on the formulation and data indicated in the methodological section (Section 3) , the changes in intra-EU production linkages by country are analyzed, considering the valueadded generated and exported in the European regional value chain. In this respect, first, a comparison is made between forward participation rate in 2005 and 2015, corresponding with the stage of integration of the CEECs into the EU. As can be seen in Figure 1, most EU countries have increased their level of forward participation in GVCs throughout intra-EU production linkages as a percentage of their total gross exports (extra and intra-EU exports) between 2005 and 2015. Moreover, it is important to note that intra-EU exports account in many cases for 60–70% of member states’ external trade. Economies 2021, 9, x FOR PEER REVIEW 8 of 18 fore a unit change in the independent variables is expected to change de dependent variable by 100 · βj percent. The model estimation results using GRETL software are presented and discussed in Section 4.2. 4. Results and Discussion 4.1. Descriptive Analysis: Facts and Trends Based on the formulation and data indicated in the methodological section (Section 3), the changes in intra-EU production linkages by country are analyzed, considering the value-added generated and exported in the European regional value chain. In this respect, first, a comparison is made between forward participation rate in 2005 and 2015, corresponding with the stage of integration of the CEECs into the EU. As can be seen in Figure 1, most EU countries have increased their level of forward participation in GVCs throughout intra-EU production linkages as a percentage of their total gross exports (extra and intra-EU exports) between 2005 and 2015. Moreover, it is important to note that intra-EU exports account in many cases for 60–70% of member states’ external trade. Figure 1. Intra-EU forward global value chains (GVC) participation as a percentage of total gross exports. EU28 countries. 2005 and 2015 See Appendix A for country codes list. Source: Own elaboration based on TiVA. In this sense, it is possible to establish that a large part of the external activity of EU countries is organized through regional value chains, and not only in the form of global chains, in the so-called Factory Europe. From Figure 1 it can be seen that intra-EU value creation processes have strengthened over the period under analysis. There are other two important regions where intra-regional trade stands out: Factory Asia and Factory America. The existence of these areas of high intra-regional trade intensity implies, in terms of exports-related emissions, that most of the direct environmental effects of foreign trade remain within the region where the trade flows originate. In this case, Factory Europe stands out the other two factories (Asia and America), showing a share of intra-regional exports above 50% of total EU28 exports, as can be seen in Figure 2. Aut Bel Cze Dnk Est Fin Fra Deu Grc Hun Irl Ita Lva Ltu Lux Nld Pol Prt Svk Svn Esp Swe Gbr Bgr Hrv Cyp Mlt Rou EU28 0 2 4 6 8 10 12 14 16 18 0 2 4 6 8 10 12 14 16 18 Intra-EU forward GVC participation (%) 2015 Intra-EU forward GVC participation (%) 2005 Figure 1. Intra-EU forward global value chains (GVC) participation as a percentage of total gross exports. EU28 countries. 2005 and 2015 See Appendix Afor country codes list. Source: Own elaboration based on TiVA. In this sense, it is possible to establish that a large part of the external activity of EU countries is organized through regional value chains, and not only in the form of global chains, in the so-called Factory Europe. From Figure 1it can be seen that intra-EU value creation processes have strengthened over the period under analysis. There are other two important regions where intra-regional trade stands out: Factory Asia and Factory America. The existence of these areas of high intra-regional trade intensity implies, in terms of exports-related emissions, that most of the direct environmental effects of foreign trade remain within the region where the trade flows originate. In this case, Factory Europe stands out the other two factories (Asia and America), showing a share of intra-regional exports above 50% of total EU28 exports, as can be seen in Figure 2. Although the share of intra-European trade has declined over time, it still accounts for more than half of the international trade of most member countries. Furthermore, it should be noted that, despite the increasing insertion of European exports in international markets, intra-EU value creation has deepened, increasing the internal production chains of the Factory Europe, as already pointed out in Figure 1. Economies 2021,9, 54 9 of 17 Economies 2021, 9, x FOR PEER REVIEW 9 of 18 Figure 2. EU28 exports (extra and intra) and intra-EU forward participation. 2005 and 2015 (The value data are in millions of US$, the percentage data are calculated out of the total EU28 exports). Source: Own elaboration based on TiVA. Although the share of intra-European trade has declined over time, it still accounts for more than half of the international trade of most member countries. Furthermore, it should be noted that, despite the increasing insertion of European exports in international markets, intra-EU value creation has deepened, increasing the internal production chains of the Factory Europe, as already pointed out in Figure 1. Before analyzing the environmental impact of trade in value added from domestic sources, it is worth analyzing the changes in the emissions intensity of the member states in the period considered. Emissions intensity is the ratio between the volume of emissions and total output, which results in tonnes of CO2 per thousand dollars of total output (Figure 3). Figure 3 shows some discrepancies in the evolution of emissions intensity. In particular, countries with a higher initial emissions intensity also show a higher percentage reduction over time. These countries are mainly represented by the Eastern European economies. For the other member states, the reduction in emissions intensity is less relevant. Behind the initial discrepancy in emissions intensity and its evolution over time lies the integration of the above-mentioned economies into the EU. The integration process requires the harmonization of regulations, including, to some extent, environmental issues. The adoption of European environmental standards significantly contributes to the reduction of the emission intensity in Eastern European economies. In this sense, these results show a reduction in the disparities between member countries in terms of CO2 emissions linked to international trade. However, this indicator does not allow to observe the extent to which Eastern European countries have attracted the most polluting tasks of Factory Europe. 59.6% 53.2% 40.4% 46.8% 12.4% 13.1% 0 500 1000 1500 2000 2500 3000 3500 4000 2005 2015 Intra-EU exports Extra-EU exports Intra-EU forward participation Figure 2. EU28 exports (extra and intra) and intra-EU forward participation. 2005 and 2015 (The value data are in millions of US$, the percentage data are calculated out of the total EU28 exports). Source: Own elaboration based on TiVA. Before analyzing the environmental impact of trade in value added from domestic sources, it is worth analyzing the changes in the emissions intensity of the member states in the period considered. Emissions intensity is the ratio between the volume of emissions and total output, which results in tonnes of CO 2 per thousand dollars of total output (Figure 3). Economies 2021, 9, x FOR PEER REVIEW 10 of 18 Figure 3. Emissions intensity (tonnes of CO2 per thousand dollars) of EU28 countries. 2005 and 2015. Source: Own elaboration based on TiVA and Eora26. Moreover, it is possible to analyze the comparative evolution of emissions per capita and GDP per capita, as Figure 4 shows. Although the verification of the EKC is not the focus of this paper, this comparison can provide new insights on this relationship. Figure 4. Evolution of GDP per capita (constant 2005 US$) and CO2 emissions in EU28 countries, 2005–15. See Appendix A for country codes list. Source: Own elaboration based on TiVA, Eora26 and the World Bank. According to Figure 4, most European economies have decreased their GDP per capita (constant prices) while reducing their emissions per capita (third quadrant). However, the integration of the Eastern economies has not led all EU countries to behave in the same 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Bulgaria Estonia Romania Poland Slovak Republic Czech Republic Lithuania Croatia Slovenia Hungary Greece Latvia Cyprus Malta Portugal EU Spain Germany Finland Belgium Austria Italy Netherlands United Kingdom Luxembourg Denmark France Ireland Sweden 2005 2015 Aut Bel Bgr Hrv Cyp Cze Dnk Est Fin Fra Deu Grc Hun Irl Ita Lva Ltu Lux Mlt Nld Pol Prt Rou Svk Svn Esp Swe Gbr EU28 -6 -5 -4 -3 -2 -1 0 1 2 -4 -3 -2 -1 0 1 2 3 4 5 6 CO2per capita, cumulative variation rate 2005-2015 GDP per capita (constant prices), cumulative variation rate 2005–15 Figure 3. Emissions intensity (tonnes of CO 2 per thousand dollars) of EU28 countries. 2005 and 2015. Source: Own elaboration based on TiVA and Eora26. Economies 2021,9, 54 16 of 17 Ansari, Mohd Arshad, and N. A. Khan. 2021. Decomposing the Trade-Environment Nexus for High Income, Upper and Lower Middle Income Countries: What Do the Composition, Scale, and Technique Effect Indicate? Ecological Indicators 121: 107122. [CrossRef] Antweiler, Werner, Brian R. Copeland, and M. Scott Taylor. 2001. Is Free Trade Good for the Environment? American Economic Review 91: 877–908. [CrossRef] Arndt, Sven W., and Henryk Kierzkowski. 2001. Fragmentation: New Production Patterns in the World Economy. Oxford: Oxford University Press. Atici, Cemal. 2009. Carbon Emissions in Central and Eastern Europe: Environmental Kuznets Curve and Implications for Sustainable Development. Sustainable Development 17: 155–60. [CrossRef] Bagayev, Igor, and Julie Lochard. 2017. EU Air Pollution Regulation: A Breath of Fresh Air for Eastern European Polluting Industries? Journal of Environmental Economics and Management 83: 145–63. [CrossRef] Baldwin, Richard. 2006. Multilateralising Regionalism: Spaghetti Bowls as Building Blocs on the Path to Global Free Trade. The World Economy 29: 1451–518. [CrossRef] Baldwin, Richard. 2012. Global Supply Chains: Why They Emerged, Why They Matter, and Where They Are Going. CEPR Discussion Papers DP9103. Washington, DC: Center for Economic and Policy Research (CEPR). Börzel, Tanja A., and Aron Buzogány. 2019. Compliance with EU Environmental Law. The Iceberg Is Melting. Environmental Politics 28: 315–41. [CrossRef] Chou, Haifeng, and Qi Wang. 2020. Empirical Study on the Environmental Effect of Global Value Chains: Based on Scale, Technique and Composition Effects. Paper presented at 2020 International Conference on social and Human Sciences (ICSHS 2020), Camberra, Australia, March 7–8. Copeland, Brian R., and M. Scott Taylor. 2004. Trade, Growth, and the Environment. Journal of Economic Literature 42: 7–71. [CrossRef] Dean, Judith M. 1992. Trade and the Environment: A Survey of the Literature. WPS966. Washington, DC: The World Bank. Destek, Mehmet Akif, Recep Ulucak, and Eyup Dogan. 2018. Analyzing the Environmental Kuznets Curve for the EU Countries: The Role of Ecological Footprint. Environmental Science and Pollution Research 25: 29387–96. [CrossRef] [PubMed] Dogan, Eyup, and Roula Inglesi-Lotz. 2020. The Impact of Economic Structure to the Environmental Kuznets Curve (EKC) Hypothesis: Evidence from European Countries. Environmental Science and Pollution Research 27: 12717–24. [CrossRef] Duan, Yuwan, Ting Ji, and Tuotuo Yu. 2021. Reassessing Pollution Haven Effect in Global Value Chains. Journal of Cleaner Production 284: 124705. [CrossRef] European Commission. 2020. Leading the Way to a Global Circular Economy: State of Play and Outlook. SWD (2020) 100 final. Luxembourg: Publications Office of the EU. Fang, Jiayu, Cenjie Liu, and Chao Gao. 2019. The Impact of Environmental Regulation on Firm Exports: Evidence from Environmental Information Disclosure Policy in China. Environmental Science and Pollution Research 26: 37101–13. [CrossRef] Fritsch, Manuel, and Jürgen Matthes. 2017. Factory Europe and Its Ties in Global Value Chains. GED Focus Paper. Cologne: Bertelsmann Stiftung. Geishecker, Ingo. 2006. Does Outsourcing to Central and Eastern Europe Really Threaten Manual Workers’ Jobs in Germany? The World Economy 29: 559–83. [CrossRef] Gereffi, Gary. 2005. The Global Economy: Organization, Governance, and Development. The Handbook of Economic Sociology 2: 160–82. Gereffi, Gary. 2014. Global Value Chains in a Post-Washington Consensus World. Review of International Political Economy 21: 9–37. [CrossRef] Grossman, Gene M., and Alan B. Krueger. 1991. Environmental Impacts of a North American Free Trade Agreement. NBER Working Papers 3914. Cambridge: National Bureau of Economic Research, Inc. Guzel, Arif Eser, and ˙ Ilyas Okumus. 2020. Revisiting the Pollution Haven Hypothesis in ASEAN-5 Countries: New Insights from Panel Data Analysis. Environmental Science and Pollution Research 27: 18157–67. [CrossRef] [PubMed] Ho, Sin-Yu, and Bernard Njindan Iyke. 2019. Trade Openness and Carbon Emissions: Evidence from Central and Eastern European Countries. Review of Economics 70: 41–67. [CrossRef] Jalil, Abdul, and Syed F. Mahmud. 2009. Environment Kuznets Curve for CO 2 Emissions: A Cointegration Analysis for China. Energy Policy 37: 5167–72. [CrossRef] Kellenberg, Derek K. 2009. An Empirical Investigation of the Pollution Haven Effect with Strategic Environment and Trade Policy. Journal of International Economics 78: 242–55. [CrossRef] Kleemann, Linda, and Awudu Abdulai. 2013. The Impact of Trade and Economic Growth on the Environment: Revisiting the Cross-Country Evidence. Journal of International Development 25: 180–205. [CrossRef] Koopman, Robert, Zhi Wang, and Shang-Jin Wei. 2014. Tracing Value-Added and Double Counting in Gross Exports. The American Economic Review 104: 459–94. [CrossRef] Leitão, Nuno Carlos, and Jeremiás MátéBalogh. 2020. The Impact of Intra-Industry Trade on Carbon Dioxide Emissions: The Case of the European Union. Agricultural Economics 66: 203–14. [CrossRef] Lenzen, Manfred, Keiichiro Kanemoto, Daniel Moran, and Arne Geschke. 2012. Mapping the Structure of the World Economy. Environmental Science & Technology 46: 8374–81. [CrossRef] Lenzen, Manfred, Daniel Moran, Keiichiro Kanemoto, and Arne Geschke. 2013. Building Eora: A Global Multi-Region Input–Output Database at High Country and Sector Resolution. Economic Systems Research 25: 20–49. [CrossRef] Economies 2021,9, 54 17 of 17 Leontief, Wassily W. 1937. Interrelation of Prices, Output, Savings, and Investment. The Review of Economics and Statistics 19: 109–32. [CrossRef] Leontief, Wassily. 1951. The Structure of American Economy, 1919–39: An Empirical Application of Equilibrium Analysis. New York: Oxford University Press. Low, Patrick, and A. Yeats. 1992. Do “Dirty” Industries Migrate? In International Trade and the Environment. Edited by Patrick Low. World Bank Discussion Papers 159. Washington, DC: World Bank, pp. 89–103. Lucas, Robert E. B., David R. Wheeler, and Hemamala Hettige. 1992. Economic Development, Environmental Regulation, and the International Migration of Toxic Industrial Pollution: 1960–88. WPS1062. Washington, DC: The World Bank. Mandras, Giovanni, and Simone Salotti. 2020. An Input–Output Analysis of Sectoral Specialization and Trade Integration of the Western Balkans Economies. Economies 8: 93. [CrossRef] Marin, Dalia. 2006. A New International Division of Labor in Europe: Outsourcing and Offshoring to Eastern Europe. Journal of the European Economic Association 4: 612–22. [CrossRef] Martínez-Zarzoso, Inmaculada, Martina Vidovic, and Anca M. Voicu. 2017. Are the Central East European Countries Pollution Havens? The Journal of Environment & Development 26: 25–50. [CrossRef] Mosconi, Enrico Maria, Andrea Colantoni, Filippo Gambella, Eva Cudlinová, Luca Salvati, and Jesús Rodrigo-Comino. 2020. Revisiting the Environmental Kuznets Curve: The Spatial Interaction between Economy and Territory. Economies 8: 74. [CrossRef] Pablo-Romero, Maria Pablo, Luís Cruz, and Eduardo Barata. 2017. Testing the Transport Energy-Environmental Kuznets Curve Hypothesis in the EU27 Countries. Energy Economics 62: 257–69. [CrossRef] Pethig, Rüdiger. 1976. Pollution, Welfare, and Environmental Policy in the Theory of Comparative Advantage. Journal of Environmental Economics and Management 2: 160–69. [CrossRef] Sass, Magdolna, and Martina Fifekova. 2011. Offshoring and Outsourcing Business Services to Central and Eastern Europe: Some Empirical and Conceptual Considerations. European Planning Studies 19: 1593–609. [CrossRef] Shahbaz, Muhammad, Ilham Haouas, and Thi Hong Van Hoang. 2019. Economic Growth and Environmental Degradation in Vietnam: Is the Environmental Kuznets Curve a Complete Picture? Emerging Markets Review 38: 197–218. [CrossRef] Siebert, Horst. 1974. Environmental Protection and International Specialization. Review of World Economics 110: 494–508. [CrossRef] Siebert, Horst. 1977. Environmental Quality and the Gains from Trade. Kyklos 30: 657–73. [CrossRef] Stavropoulos, Spyridon, Ronald Wall, and Yuanze Xu. 2018. Environmental Regulations and Industrial Competitiveness: Evidence from China. Applied Economics 50: 1378–94. [CrossRef] Timmer, Marcel P., Erik Dietzenbacher, Bart Los, Robert Stehrer, and Gaaitzen J. de Vries. 2015. An Illustrated User Guide to the World Input–Output Database: The Case of Global Automotive Production. Review of International Economics 23: 575–605. [CrossRef] Tsagkari, Marula, Alexis Gaona, Juan-Felipe Gonzalez, and Jaakko Järvinen. 2018. The Evolution of Carbon Dioxide Emissions Embodied in International Trade in Poland: An Input-Output Approach. Environmental & Socio-Economic Studies 6: 36–43. [CrossRef] Zhao, Guomei, and Cenjie Liu. 2020. Carbon Emission Intensity Embodied in Trade and Its Driving Factors from the Perspective of Global Value Chain. Environmental Science and Pollution Research. [CrossRef] Zhao, Xiaomeng, Chuanjiang Liu, Chuanwang Sun, and Mian Yang. 2020. Does Stringent Environmental Regulation Lead to a Carbon Haven Effect? Evidence from Carbon-Intensive Industries in China. Energy Economics 86: 104631. [CrossRef]