Abertura comercial e seus impactos na pobreza: uma análise em painel dinâmico para os estados brasileiros

Abstract

This study analyzes the nonlinear effects of trade openness on the proportion of poor individuals across Brazilian states over the period 2012–2023. To this end, we use annual data for the Brazilian federative units and adopt a dynamic panel framework with an endogenous threshold, following Seo and Shin (2016), which allows for the simultaneous estimation and testing of nonlinear and/or asymmetric effects in these relationships. The main hypothesis tested is that trade openness affects poverty in Brazilian states in a nonlinear manner, conditional on an optimal level of income inequality. The linearity test confirms the presence of a threshold effect associated with the Gini index, beyond which the impacts of economic variables change significantly. In regimes of lower inequality, both trade openness and economic growth exert stronger effects on poverty reduction, whereas under higher inequality these effects are substantially attenuated, reinforcing that the effectiveness of growth and trade openness in reducing poverty must be aligned with policies aimed at reducing income concentration.