Countries that price carbon more stringently than their trading partners face carbon leakage and costs for their consumers. This paper develops a quantitative general equilibrium trade model with heterogeneous firms, input-output linkages, and emissions from fossil fuel use to study unilateral carbon pricing and the instruments that accompany it. Firm heterogeneity opens two channels: carbon pricing shifts entry and market size across sectors that differ in their economies of scale and selection effects, and it reallocates market shares toward more productive and cleaner firms. I estimate the scale and selection elasticities with administrative German firm data and apply the model to EU climate policy. Firm heterogeneity raises the consumption loss of carbon pricing by a factor of four relative to a model with constant returns to scale, almost entirely through economies of scale, and in combination with input-output linkages it raises carbon leakage. The same forces make output-based rebates and border adjustments more effective against leakage and create scope for recycling carbon revenues toward an industrial policy, which cuts the consumption loss by around 80%. In contrast, the reallocation of market shares across firms plays a minor role for emissions.

This paper investigates the role of firm heterogeneity in environmentally extended new trade models, contrasting Eaton-Kortum and Melitz models to Armington and Krugman models. We show that when emissions per sales are constant across firms -- a standard assumption in the literature -- all four models predict identical emission responses. However, when emissions per quantity are constant across firms, this equivalence breaks. We propose a generalized framework that nests both assumptions. Calibrating the model with multiple industries and estimating the key elasticity between emission intensity and productivity using German firm-level data, we find that firm heterogeneity considerably raises emissions from trade liberalization.

This paper revisits the exporter’s environmental premium (EEP) by incorporating emissions embodied in domestically and internationally sourced intermediate inputs. Combining administrative firm-level data and customs records for German manufacturers with an environmentally extended input-output table and fuel specific emission factors, we document three stylized facts: (i) embodied emissions account for over half of firms’ total emissions; (ii) exporters’ production involves disproportionately more embodied emissions, particularly through international sourcing; and (iii) once embodied emissions are considered, the EEP reverses: exporters appear cleaner based on production-related emissions alone, but dirtier in total emissions. We rationalize these patterns in a sourcing model and test its predictions using a shift-share IV strategy based on foreign demand shocks. Export expansion lowers the production-related emission intensity without affecting total emissions, underscoring the role of sourcing in shaping firm-level environmental outcomes. These findings highlight the importance of accounting for embodied emissions when evaluating the welfare and environmental consequences of trade liberalization. 

Carbon pricing policies are usually combined with compensation for exposed firms to prevent adverse competitiveness effects. In cap-and-trade systems, this carbon cost compensation mostly occurs through free allocation of emission permits. Using an administrative panel of German manufacturing firms, this paper investigates how free allocation in the European Union Emissions Trading System affects firms’ competitiveness and their incentives to reduce emissions. Leveraging a reform of free allocation rules in a continuous difference-in-differences design, we find that that a reduction of freely allocated emission permits decreased firms’ emission intensity. Our results suggest that this decrease is driven by energy efficiency improvements instead of outsourcing of emission intensive production. On the other hand, we do not find statistically significant effects on firms’ employment, sales, value added, investments and exports – indicating that the reduction in free permits did not reduce firms’ competitiveness.