Purpose This study aims to identify whether increases in trademark activity reshape countries' trade structure, specifically whether they raise or lower concentration across products and markets, and determine through which side of the bilateral relationship (origin vs destination) these effects operate. Design/methodology/approach We assemble a bilateral country-pair panel (1995–2022) linking trademark activity (domestic and Madrid System filings) to trade concentration measures (Herfindahl–Hirschman Index (HHI); Theil indices decomposed into product and market components). To address endogeneity, we instrument trademarks using exogenous variation from the Madrid System's expansion and fee structure; specifications include origin, destination and year fixed effects with clustered standard errors. Findings Inbound (destination-side) trademark activity causally increases trade concentration, with robust effects on market concentration (imports and exports) in non-high-income destinations. Instrumental variables (IV) estimates are larger than ordinary least squares (OLS), indicating OLS understates the true effect. Product-level diversification effects under OLS attenuate under IV, suggesting that trademarks reallocate market shares rather than expand product variety. Origin-side effects are generally weak or insignificant. Research limitations/implications Instrument strength varies across income subsamples; while robust overall and for developing destinations, some high-income estimates have weaker first stages. Future work could integrate product-level brand intensity and firm-level outcomes to refine mechanisms. Originality/value This study provides the first causal evidence that trademarks rather than patents systematically increase trade concentration via demand-side (destination) channels, leveraging the Madrid System as an instrument and decomposed Theil measures to pinpoint mechanisms.
Tharwat et al. (Tue,) studied this question.