Gravity model of timber trade from Latin America

This study examines the determinants of timber trade flows from 18 South and Central America countries to global markets over the period 1999--2018 by estimating a gravity model for trade. We focus on industrial roundwood, the dominant timber product in international trade. Given the high prevalence of zero trade flows within our import-export matrix, we employ a Negative Binomial hurdle gravity model estimated following the Bayesian paradigm. The resultant posterior distribution is approximated through the Integrated Nested Laplace Approximation. Moreover, in order to assess the impact on trade dynamics of economic, geographic, and environmental factors, we focus on country-specific random effects rather than fixed effects---a choice particularly natural within a Bayesian setting. Our study contributes to the literature by incorporating a novel statistical methodology for gravity models and by providing insights into Latin America's bidirectional timber trade. The policy implications highlights the importance of managed forest activities and integrated land-use policies are crucial for strengthening Latin America’s role in the global timber trade while mitigating deforestation and supply risks from climate-related shocks. Additionally, economic stability and well-designed trade agreements can enhance trade flows, ensuring that timber trade supports both economic development and forest conservation.