pass-through do câmbio para a inflação na economia brasileira (2003-2018): modelos ardl

• Ano de defesa: 2020
• Autor(a) principal: Sousa Jr, Valdecy Caetano de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Economia
• Palavras-chave em Português: pass-through; taxa de câmbio; Inflação; ARDL; N-ARDL; exchange rate; Inflation; CNPQ::CIENCIAS SOCIAIS APLICADAS::ECONOMIA::TEORIA ECONOMICA::ECONOMIA GERAL; CNPQ::CIENCIAS SOCIAIS APLICADAS::ECONOMIA::METODOS QUANTITATIVOS EM ECONOMIA::METODOS E MODELOS MATEMATICOS, ECONOMETRICOS E ESTATISTICOS
• Link de acesso: https://repositorio.ufu.br/handle/123456789/29005; http://doi.org/10.14393/ufu.di.2020.53

The goal of this dissertation is to investigate the relationship between changes in the exchange rate and inflation for Brazil from 2003 to 2019. Other than the estimation of the exchange rate pass through the work also investigates the existence of asymmetries in the pass through, searching for evidences of different magnitudes for appreciation and depreciation of the exchange rate. To achieve this goal we estimate four linear models (ARDL) and four nonlinear models (N-ARDL) using as dependent variables the official inflation index (IPCA) and the general price index (IGP-DI), while for the exchange rate we use the nominal and the real effective exchange rate. The model also uses as control variables the oil price, the degree of trade openness and the GDP gap. All the estimated models found the existence of cointegration and after this we estimated the long run coefficients, the short run dynamics and the error correction mechanism (ECM). The results indicate that changes in the exchange rate have a statistically significant impact in the inflation rate with a 1% variation in exchange rates increasing the IPCA and IGP-DI by 0.05 pp and 0.18 pp, respectively. For the asymmetries of the results concluded for the IPCA, which depreciate exchange rates and impact inflation more strongly than the appreciations (0.08 pp and 0.03 pp for nominal exchange rate variation and 0.11 pp and 0.03 pp for nominal effective exchange rate variation) for IGP-DI or coefficients of variation (0.17 pp and 0.18 pp for nominal exchange rate depreciation and recovery, respectively; and 0.14 pp and 0.29 pp for depreciation and depreciation rate) nominal exchange, respectively) and there is no evidence of such asymmetries. For robustness purposes the work also uses additional Granger causality tests for inflation, changes in the nominal and the nominal effective exchange rates. The results confirm the hypothesis that changes in the exchange rate Granger causes inflation (IPCA and IGP-DI).