How to estimate whether preferential trade agreements contribute to international environmental impact shifting. A new methodology and empirical illustration for Switzerland

We present a new methodological approach for estimating the international relocation of environmental impacts of consumption through trade liberalisation. Our approach reconstructs changes in import flows after the entry into force of a preferential trade agreement (PTA) while taking into account different dynamics of trade diversion and trade growth. Our empirical application focuses on whether and how much the entry into force of 25 PTAs affected greenhouse gas (GHG) emissions embodied in Swiss imports between 2000 and 2018. The analysis is based on a unique dataset for more than 7'000 products that includes information on whether and how much of a given product was imported under a particular PTA or other trade rules. We find that the relocation of embodied GHG emissions attributed to the 25 PTAs is relatively small compared to the overall increase in imported GHG emissions. Most of the additional relocation of GHG emissions via trade diversion between 2000-2018 took place via imports under other trade rules. These findings indicate that the marginal effect of additional PTAs only has a small effect on overall embodied emissions in imports, at least when it comes to GHG emissions and as long as these PTAs have a similar scope as prior trade liberalisations. Given the widespread controversies over the environmental implications of PTAs, our approach provides a useful template for ex-post assessments of such implications. The results of such assessments can help stakeholders and policymakers in designing evidence-based policies to identify and mitigate the "outsourcing" environmental impacts of consumption that may result from PTAs.