Answers are grounded in this paper and cite their sources.
Using Poisson pseudo-maximum likelihood gravity models with exporter-year, importer-year, and directed-pair fixed effects, we relate UN Tourism visitor and tourist flows to importer-reported UN Comtrade imports of an author-defined 30-code cultural-goods basket, 2007–2021. The tourist coefficient is 0.180 (SE = 0.058, N = 26,339; 0.233 including HS 490700) and stays significant at the 5% level in most robustness checks. On matched 2007–2020 observations, it exceeds that for total imports by 0.200 (SE = 0.085, p = 0.018; Holm-adjusted p = 0.074), and no basket’s Holm-adjusted difference from total or non-cultural imports is significant at the 5% level. The tourist coefficient falls to 0.108 (SE = 0.073; 0.190 including HS 490700) without mobility values above the 99.9th percentile and to 0.096 (p = 0.053) with directed-pair trends, whose dynamic version links future arrivals to trade. For the narrowest UNESCO-consistent basket, it is 0.067 (SE = 0.052). The visitor coefficient (β = 0.173, SE = 0.084, N = 20,598) is −0.055 where both directions share a classification basis and 0.473 elsewhere (tourists: 0.318 and 0.049); this split changes measurement and the pairs observed. The two samples share 54 of 167 economies; on their small common support, neither coefficient differs significantly from zero.
No takes yet. Share an insight, caveat, or question.
Gao et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: