Parcelling virtual carbon in the pollution haven hypothesis Luis Antonio López, Guadalupe Arce, Jorge Enrique Zafrilla ⁎ Universidad de Castilla-La Mancha Facultad de Ciencias Económicas y Empresariales, Plaza de la Universidad n. 1, 02071 Albacete, Spain abstract article info Article history: Received 3 May 2013 Received in revised form 4 May 2013 Accepted 5 May 2013 Available online 13 May 2013 JEL Classifications: F14 Q56 L9 Keywords: Pollution haven hypothesis Global value chain Input–output Spain‐China trade The methodology proposed in this paper allows us to parcel the pollution haven hypothesis (PHH) into a bi-regional input–output framework to analyse whether the specialisation of countries in different stages of production and/or in final goods trading generates an increase or a decrease in global emissions as a con- sequence of international trade. We apply the model to the Spain–China trade relationship as it existed in 2005, finding a PHH of 29,667 KtCO 2 . If this trade had not existed (so each country had met its demand for intermediate and final goods), global emissions would have been reduced by these 29,667 KtCO 2 . Of this PHH, 43.5% corresponds to imports of final goods; 32.4% is related to imports of intermediate goods for the last stage of production; the remainder, 24.1%, is caused by global value chains (GVC) between the countries. Only 3229 KtCO 2 of PHH emissions are linked to domestic emissions from the sector in which the imports are produced; the rest is explained by domestic linkages or successive rounds of domestic production, which supports the existence of an indirect PHH. Together with a trade growth in the last years, the fall of trade bar- riers would have implied a transformation of global production chains that have boosted global emissions. © 2013 Elsevier B.V. All rights reserved. 1. Introduction From the point of view of the global economy, the most efficient framework is to produce goods where, direct and indirectly, the produc- tion pollutes the least, where an environmental comparative advantage exists (Peters and Hertwich, 2008). The pollution haven hypothesis (PHH) occurs when a reduction in trade barriers increases trade and subsequent emissions because firms seek to exploit other comparative advantages (Copeland and Taylor, 2004), like low salaries and energy costs, the lack of environmental regulations, etc. A branch of the litera- ture has studied the existence of PHH using econometric regression models (Antweiler et al., 2001; He, 2006), 1 and another branch of the literature has addressed the importance of evaluating this hypothesis using the input–output framework through virtual carbon in exports minus virtual carbon avoided by imports (Ackerman et al., 2007; Chen and Chen, 2011; Dietzenbacher and Mukhopadhyay, 2007; Peters et al., 2007; Zhang, 2012). International trade has increased from 5.5% to 21% of worldwide GDP from 1950 to 2007 (WTO-UNEP, 2009). Environmentally, the effect is such that 26% of CO 2 emissions linked to production in the world economy in 2008 are internationally traded (Peters et al., 2011). The methodology of emissions balance, production-based emissions minus consumption-based emissions, shows how this effect differs between developed and developing/emerging countries (Chen and Chen, 2011; Peters and Hertwich, 2008; Peters et al., 2011; Peters et al., 2012). De- veloped countries have a deficit and developing countries have a sur- plus; the former import energy and CO 2 intensive goods directly or indirectly in exchange for environmentally-friendly exports and have a higher level of consumption or a deficit in trade (Davis and Caldeira, 2010; Steen-Olsen et al., 2012). Nevertheless, it is not possible to use the emissions balance to know if international trade increases or de- creases global emissions because the aggregation of worldwide trade and emissions balances for all countries is always zero, whereas for two regions, the trade balance influences the emissions balance. Jakob and Marschinski (2012) show that the sign of the emissions balance de- pends on the trade balance, the energy intensiveness of the economy, emissions intensiveness, and, finally, the comparative advantages de- rived from the specialisation processes in the countries. The PHH measure proposed by Dietzenbacher and Mukhopadhyay (2007) isolates the effects of comparative advantages in the evolution of emissions in international trade through the emissions induced by Energy Economics 39 (2013) 177–186 ⁎ Corresponding author. Tel.: +34 967 599 200x2183. E-mail addresses: Luis.LSantiago@uclm.es (L.A. López), Guadalupe.Arce@alu.uclm.es (G. Arce), Jorge.Zafrilla@uclm.es (J.E. Zafrilla). 1 Antweiler et al. (2001) propose a model that allows decompose trade effects on pollution into scale, technique and composition effects without confirming the exis- tence of PHH; they conclude that international trade has a positive influence on the en- vironment and SO 2 emissions decrease by 1% when considering 44 countries during the 1971–1996 period. Furthermore, He (2006) proposes a simultaneous equation model to study the relationship between foreign direct investment (FDI) and SO 2 emis- sions and confirms the existence of PHH in China during the period 1994–2001. Specif- ically, the author finds that a 1% increase in FDI increases SO 2 emission by 0.09%, which provides evidence for the existence of PHH. queryIn the literature some difficulties have been found to confirm the PHH, however, pollution haven effect (PHE), defined as the effect produced by the tightening of environmental legislation on location deci- sions of companies, has been amply confirmed (Ederington et al., 2005; Kellenberg, 2009; Levinson and Taylor, 2008; Wagner and Timmins, 2009). 0140-9883/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.eneco.2013.05.006 Contents lists available at SciVerse ScienceDirect Energy Economics journal homepage: www.elsevier.com/locate/eneco