Does international trade drive CO2 emissions at the U.S. state-level?

Whether and how trade impacts carbon emissions in Global North nations like the United States (U.S.) remains contested. Prior analyses focus on nation-states and rarely examine this relationship in subnational contexts. To fill this gap, I test a range of theories related to the trade-emissions nexus using U.S. state-level panel data from 2002–2019 and the half-panel jackknife fixed effects estimator. I examine whether trade intensity in the aggregate and delineated by industrial sector is associated with industrial CO2 emissions per capita and industrial emission intensity. I also employ moderation analysis to test whether the impact of trade is moderated by scale, compositional, and technique effects emphasized by the treadmill of production, ecological modernization, and factor endowment perspectives. I find that trade intensity in the aggregate is not associated with state-level emissions or emission intensity, but when disaggregated, it is associated with increased emissions and emission intensity in states more reliant on agriculture, forestry, fishing, hunting, and mining exports. Trade is also moderated by compositional and scale effects, where trade effects are greater in states with more capital-intensive economies that experience higher rates of economic growth—supporting arguments made by the treadmill of production and factor endowment perspectives.