Ensaios sobre política monetária, choques de oferta e dinâmica inflacionária.

Abstract

Inflation is one of the main variables in modern macroeconomic analysis, as central banks around the world aim to maintain price stability. Following the high inflation observed during the coronavirus pandemic, the challenge of dealing with large supply shocks posed inflationary difficulties not seen by central banks since the oil shocks of the 1970s. In this context, this thesis proposes three empirical exercises related to this theme. In the first article, a nonlinear monetary policy rule is estimated considering different inflation regimes. Monthly data from January 2003 to December 2021 and econometric models with threshold effects in the presence of endogenous regressors were used. The results indicate a smoothing process in the conduct of Brazilian monetary policy, showing that economic activity and the exchange rate are not statistically significant in the reaction function of the Brazilian central bank during the period studied, regardless of the inflation regimes analyzed. In the low inflation regime, the effect of the deviation of inflation expectations from its target is statistically null, revealing a preference of the Brazilian central bank for price stability. In the second article, a case study on the truck drivers' strike that took place in Brazil from May 21 to 30, 2018, is proposed to investigate how the degree of tradability of inflation subgroups is affected by a supply shock. Using monthly data from June 2017 to December 2019 and a dynamic difference-in-differences model applicable in event studies, the results indicate that tradable sub-items respond more intensely than their non-tradable counterparts, exhibiting a persistence of nine months. Additionally, sectoral analysis highlights the groups Food and Beverages, Housing, Household Articles, Transportation, and Personal Expenses as the most affected by the truck drivers' strike, with price effects persisting for eight months. Finally, the third essay investigates whether the asymmetry of relative prices serves as a good alternative proxy for modeling the effects of supply shocks in the Hybrid New Keynesian Phillips Curve (HNKPC), using monthly information from January 2003 to December 2023 and the generalized method of moments robust to heteroskedasticity and autocorrelation (GMMHAC). Additionally, robustness exercises were conducted in subsamples indicative of different macroeconomic conditions. The results confirm that the asymmetry of relative prices emerges as a good supply shock variable in the NKHPC and that its effect tends to be greater in scenarios of higher economic uncertainty. Inflation inertia proved significant, with effects amplified in scenarios of greater instability. Lastly, the forward-looking component of inflation shows statistical significance only in an environment of greater macroeconomic stability.