Is Brazil’s premature deindustrialization a case of Dutch disease? Comment on Nassif, Feijó and Araújo (2026)
Main Article Content
Abstract
Nassif, Feijó and Araújo (2026) test whether Brazil’s premature deindustrialization is a case of Dutch disease. We identify econometric problems that call into question their conclusions: the dependent variable does not measure deindustrialization; the OLS equation omits lagged dependent variables, generating autocorrelation that invalidates inference; trend and aggregate demand controls are absent; the trade liberalization variable is circularly constructed with a wrong-signed coefficient; REER and FIN remain endogenous; the VEC neither resolves identification problems nor is ordered consistently with the causal argument; and independent replication is precluded by inaccessible data. Their central claims require a revised specification.
JEL Classification: O14; O54; C22; C26; C32.