External Demand Shocks, Export Intensity, and Productivity: Sectoral Evidence for the Brazilian Industry (2015–2019)
international trade; productivity; export intensity; learning-by-exporting; Brazilian industry.
This dissertation investigates the relationship between sectoral export intensity and
productivity in the Extractive and Manufacturing Industries of Brazil over the period
2015–2019. Grounded in International Trade Theory, the New Trade Theory, and
heterogeneous-firm models à la Melitz, the study engages with the learning-by-exporting
(LBE) literature, which distinguishes self-selection effects from potential productivity gains
arising from export experience. Drawing on international and Brazilian evidence based on
microdata, which points to robust self-selection and heterogeneous learning across firms, the
research shifts the focus to the sectoral level, seeking to verify whether changes in export
intensity leave an identifiable “signature” in the trajectory of aggregate productivity. To this
end, a balanced sectoral panel is constructed for 27 activities in the Extractive and
Manufacturing Industries (CNAE 2.0, two-digit), with annual information on real value
added, employment, capital stock, exports, measures of labor productivity and total factor
productivity (TFP), as well as an external demand shock indicator in a shift-share format. The
empirical strategy treats sectoral export intensity as an explanatory variable with an
exogenous component derived from these external demand shocks and estimates
contemporaneous and lagged responses of productivity. The central objective is to distinguish
transient gains associated with scale, capacity utilization, and reallocation from those
compatible with persistent learning, as well as to explore differences between extractive and
manufacturing sectors. The dissertation thus aims to contribute to the reconciliation between
theory, microeconomic evidence, and aggregate outcomes, providing inputs for the design of
export-promotion policies from a sectoral perspective.
Keywords: