This thesis consists of three chapters, all of which focus on the rise of China as a quasi-natural experiment in order to assess the effects of foreign trade shocks on the political economy of trade policy and on the dynamics of labor markets and earnings inequality in Brazil. In the first chapter, we use evidence on the differential exposure across local labor markets to this China shock in order to estimate its effect on Brazilian labor markets outcomes, in particular on measures of income inequality. First, we find that the export demand shock has decreased wage inequality in the tradables sector, mostly through the between-firms component of wage dispersion, and provide evidence that this reduction seems driven by a change in wage-setting behavior of firms, and may be linked to a reduction in the wage premium of exporter firms. We then estimate a model based on Helpman et al. (2016), and explore sectoral differences in the foreign demand shock to run counterfactual exercises that support the hypothesis that this shock can explain part of the aggregate reduction in the exporter wage premium and in wage dispersion. In the second chapter, we develop a version of the dynamic trade model by Caliendo et al. (2019) in order to estimate the effects of the dual China shock on the sectoral dynamics of Brazilian employment. We show that both shocks lead to a contraction in most manufacturing sectors, and an expansion in most services sectors, but the general equilibrium effects of the shocks are modest, especially if compared to an alternative counterfactual in which Brazilian productivity in primary sectors increase. We then extend the model to include two types of labor, skilled and unskilled. Results also point to small distributional effects of the China shock, but consistent with reduced-form evidence obtained in Chapter 1. In the final chapter, we build a novel dataset on Brazilian trade associations’ characteristics in order to investigate whether industries with higher capacity of political organization are able to obtain more protection from foreign competitors. We use variation in import penetration as a measure of the need for trade protection, and address endogeneity on this measure by using an instrumental variables strategy based on the China import shock. Evidence suggests that industries with larger employer unions are able to obtain more protection, particularly through non-automatic licensing; the estimates suggest that this effect is higher when import penetration increases more intensely, which is interpreted as increased need for protective measures.