Discussion

The authors opened the discussion by addressing two points raised by Gregory Mankiw in his discussion. First, they argued that their model allows for sizeable long-run effects. They explained that an adjustment of border taxes leads to an appreciation of the US dollar in their model. Because the United States is a net debtor in its own currency, this appreciation leads to a negative valuation effect, corresponding to a transfer from the United States to the rest of the world of roughly 16% of US gross domestic product. This valuation effect possibly outweighs the short-term benefits of border adjustment. Second, the authors challenged the idea put forth by Mankiw that there is a clear dichotomy between short-run and long-run effects when it comes to trade policy. The authors insisted on the importance of political economy considerations in the short run, which may prevent the adoption of desired tax changes and the realization of long-run benefits. The authors next replied to questions from both discussants regarding the role of dollar pricing and the importance of Calvo pricing in the context of a large tax policy change. They pointed out that dollar pricing plays a key role in the failure of the Lerner symmetry in their model. The adoption of a border tax on imports has two effects on US consumer prices. The direct effect raises these prices, while the indirect effect reduces them by leading to an appreciation of the US dollar. In the presence of dollar pricing, the direct pass-throughof a tax is full,whereas the short-run pass-through of the exchange rate to consumer prices in the United States is low. This asymmetry is responsible for the failure of the Lerner symmetry. The authors noted that dollar pricing is consistent with recent evidence: the dollar appreciated and then depreciated by 10%–12% while border prices remained roughly unchanged. They provide a rationale for dollar pricing as an equilibriumphenomenon. International firms decide to set prices in US dollars because US inputs