Answers are grounded in this paper and cite their sources.
Replication study finds similar results in estimating supply curve parameters using gravity model of bilateral trade, suggesting implications for economic analysis.
This contribution replicates a key result of Allen, Arkolakis, and Takahashi (AAT, Journal of Political Economy, 2020), who estimate the supply curve in a cross section of countries using the estimated fixed effects of a demand-side gravity model of bilateral trade as dependent variable. The replication of their preferred specification delivers almost identical estimation results. However, AAT do not consider that the dependent variable of the supply function is a vector of estimated parameters, the fixed effects of the demand-side gravity model. The present contribution proposes a two-step estimation approach that takes into account the variation induced by the estimation of the fixed effects in the first step and leads to consistent estimates of the standard errors. Moreover, the estimates of the supply curve parameters turn out as highly sensitive to the specification of the demand-side gravity model, limiting the applicability of AAT’s approach.
No takes yet. Share an insight, caveat, or question.
Michael Pfaffermayr (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: