Polítca monetária e câmbio : efeitos sobre preços desagregados em um modelo favar para o Brasil

This paper investigates the effect of monetary and exchange rate shocks on disaggregated prices of the Brazilian Consumer Price Index (IPCA), from 1999 to 2011. We analyze the results of a factor-augmented vector autoregressive model (Favar), which are presented by different levels of aggregation. We estimate the model using Bayesian techniques, and construct impulse-response functions using sign restrictions over the responses of macroeconomic variables. The main results were: i) 65.9% of the surveyed prices at the sub-items level fell after a monetary shock and 50.7% rose after exchange rate's shock; ii) only 2.6% of the sub-items (weight of 1.6% of the index) showed the price puzzle for monetary shocks and 2.3% (weight of 0.5% of the index) for exchange rate shocks; iii) macroeconomic shocks are more persistent than series-specific shocks; iv) the answers are different according to the sector considered.