Research and projects

As production becomes increasingly fragmented across borders, a growing share of carbon emissions is generated upstream and embodied in intermediate inputs traded along global value chains. Yet little is known about how firm-level sourcing decisions shape these emissions or how trade policy alters them. This paper investigates how carbon emissions are embedded in firms’ buyer–supplier production networks and how tariffs reshape these networks and the emissions embodied in firms’ global value chains. Using unique production network data for US firms combined with firm-level measures of direct and value-chain emissions, I first document that more productive firms source from more productive and cleaner suppliers, and that firms cleaner in-house also operate cleaner supply chains. Exploiting the US–China trade war as a quasi-natural experiment, I then show that tariff-exposed firms significantly reorganize their supplier networks, reallocating sourcing away from Chinese suppliers toward domestic and third-country alternatives. This geographic reallocation reveals a fundamental decoupling: the new suppliers are more productive but also more emissions-intensive, simultaneously upgrading supply chain quality while worsening its environmental footprint. Guided by these findings, I develop and calibrate a model of endogenous firm-to-firm network formation with emissions, in which the within-country link between productivity and cleanliness breaks down when trade policy forces reallocation across countries with different emissions technologies. Counterfactual simulations quantify the environmental cost of tariff shocks and explore implications for trade policy as environmental policy.

Keywords: Heterogeneous firms, emissions, production networks, supply chains, trade policy, tariffs. 

This paper examines the impact of input trade liberalization on firms’ incentives to alter the green composition of their trade. Using the Colombia-US Free Trade Agreement (FTA) as a quasi-natural experiment, we investigate the impact of input tariff reductions on Colombian firms’ sourcing decisions and import-related emissions. The research draws on a unique firm product-level dataset for Colombia, alongside data on the emissions embedded in firms’ imported inputs. The analysis reveals a pre-existing environmental bias in trade policy, where lower input tariffs were applied to emission-intensive products before the implementation of the FTA. Due to the FTA’s tariff harmonization, Colombian firms importing from the US benefit from greater tariff reductions on less emission-intensive inputs, prompting a reconsideration of their product portfolios. We find that tariff differentials prompt firms to increase the use of liberalized greener inputs and subsequently alter the emission intensities of their imports. Specifically, higher tariff differentials on less emission-intensive inputs incentivize firms to increase their use, leading to a notable shift in input composition. Over time, we observe a decline in emissions at the firm level, coupled with a reduction in the import share of emission-intensive products.

Keywords: Heterogeneous firms, trade liberalization, environment, GHG emissions.

Over the past decades, environmental policies have become an increasingly prominent tool of regulation worldwide. While designed to promote environmental protection, such measures often raise compliance costs for both domestic producers and foreign exporters. At the same time, as import tariffs have fallen, attention has shifted toward domestic regulations, in the form of non-tariff measures, as alternative trade barriers. In this context, environmental measures may not only pursue legitimate environmental goals, but also tilt competition toward domestic producers. This paper aims to investigate the trade effects of environmental non-tariff measures, focusing on their impact on domestic firms’ import patterns. Using firm-level import data for Belgium and a developed dataset of restrictive E-NTMs, we assess whether environmental measures impact import flows and alter sourcing decisions of Belgian firms. Our findings show that affected Belgian firms adjust their sourcing decisions in response to E-NTMs on their imported inputs. Specifically, extra-EU imports decline by 13.1%, while intra-EU imports increase by 9.1%. indicating a reallocation and sourcing shift toward EU suppliers. This shift coincides with changes in revealed comparative advantage, with EU members strengthening their export position in affected products and non-EU suppliers losing market share. The decline in extra-EU imports appears to be largely driven by countries without similar environmental policies in place, underscoring the importance of relative compliance costs.

Keywords: Non-tariff measures, heterogeneous firms, environment.