Research

Agricultural trade is widely considered as a key margin of climate adaptation, as it allows regions facing large climate damages to agriculture to substitute imports for domestic production. However, this view abstracts from an important feedback effect: the reallocation will change regions’ incentive to mitigate. I establish this mechanism analytically, then quantify it in a multi-region integrated assessment model, calibrated to 13 regions, in which trade, structural change, and regional climate policy are jointly endogenous. The results show that this feedback is quantitatively important. Holding the temperature path fixed, all 12 regions that suffer net climate losses gain from eliminating agricultural trade frictions; however, once emissions, temperature, and mitigation policy respond endogenously, only five do. Regions that shift further out of agriculture cut their optimal carbon taxes by more, then cumulative global emissions through 2100 rise by 4.50 percent. The pattern reflects an asymmetry: adaptation is local, but mitigation has global effect, so lower climate exposure reduces a region’s local return of mitigation but not its global value. Agricultural trade liberalization therefore cannot be evaluated separately from the climate policy it induces.

The EU CBAM is a long-term climate policy, yet its long-run global effects remain insufficiently understood. This paper develops a forward-looking, multi-region dynamic general equilibrium model with endogenous investment and energy transition, sector-level bilateral trade, and climate–economy feedbacks. Allowing productive capital and generation capacity to adjust endogenously increases cumulative emissions reductions through the end of the century by about one-third, mainly because CBAM slows future capital and capacity accumulation rather than inducing stronger reallocation toward cleaner sectors. Revenue use is therefore central to policy design: lump-sum transfers mainly redistribute welfare, while renewable-investment subsidies generate much larger emissions reductions even without learning-by-doing—especially when directed to exporting regions—with learning-by-doing strengthening these gains further, but also producing uneven regional welfare effects.

Trade protection can support domestic clean-energy manufacturing, but the same products are also capital goods needed for decarbonization. This paper studies that tension in the context of U.S. tariffs on imported solar modules. The central mechanism is that tariffs expand domestic module production while raising the cost of renewable investment, slowing clean-capacity accumulation and increasing emissions. Because countries adjust climate policy and the effects spill across trade partners and the global climate, the domestic gains from protection must be weighed against delayed decarbonization and cross-country welfare losses. The results show that U.S. welfare is hump-shaped in the tariff: moderate protection can be beneficial, but higher tariffs increasingly undermine the energy transition.

This paper combines individual-level panel data from the China Family Panel Studies with county-level weather data to examine how rising temperatures affect the skill premium. A 1°C increase in annual average temperature lowers the skill premium by approximately 1 percentage point and is accompanied by a 2.4-percentage-point shift in employment from the tertiary to the secondary sector. The results suggest that temperature-induced structural transformation toward more climate-exposed and less skill-intensive sectors is an important mechanism.

This paper develops and empirically tests a bundled utility maximization framework for household decisions to discard multiple near-expiration food items. We find significant, non-monotonic bundle effects: after controlling for food attributes, the additional utility is positive for discarding one item or all items, but negative for discarding exactly two. These findings imply that food-waste interventions targeting individual products may mispredict household behavior unless they account for the combinations of foods households face.

This project develops an algorithm for solving non-cooperative climate policies in dynamic, multi-region integrated assessment models with bilateral trade and forward-looking investment.

This project studies how international capital flows, clean and dirty energy investment, and technology spillovers affect optimal climate policy across developed and developing regions.