Produto interno bruto, inflação e desemprego e suas relações de causalidade com variáveis de execução orçamentária de Estados brasileiros

Abstract

In recent decades, the fiscal crisis has worsened in several Brazilian states. Many of them have a debt of billions of reais, making it urgent that this fiscal problem is tackled on many fronts. In this context, it is important to understand that the macroeconomic variables interact with the variables related to budget execution in cause and effect relations, enabling the economic development or even degradation of the population's living conditions. In this sense, this paper aims to contribute to the confrontation of this crisis and to obtain the fiscal balance by clarifying how these relations occur and proposing an econometric model that makes it possible to make current revenue forecasts for the states analyzed. The general objective of this research was to analyze the causalities between the macroeconomic variables GDP, inflation and unemployment and the budget execution variables of the Brazilian states. For this, for the period 2008-2 to 2015-2 were identified the amounts referring to the revenues and expenses of four Brazilian states (Minas Gerais, Pernambuco, Rio Grande do Sul and São Paulo), as well axs the values referring to the variables GDP, inflation and unemployment relative to the territories of these political entities. Granger's causality test was used to discover existing causal relationships. Among the main results, it was found that, according to this test, the Percentage Change of GDP has a unidirectional causal relationship with the variables used to represent budget execution. It has been found to cause Total and Current Revenue, Total Expense, and Budget Execution and Current Budget Execution Quotients. It was found that the Percentage Change of GDP has a positive relationship with Total and Current Revenue and Total Expenditure, that is, whether there is an increase or reduction of that also vary in the same direction. Regarding inflation, represented by the Extended Consumer Price Index (IPCA), it was found that, according to Granger's causality test, it causes the Capital Revenue and the Budget Execution Quotient, also in a positive relationship. one way. The causalities between income and expenditure were also analyzed and a two-way causal relationship was evident in almost all relationships. Finally, the econometric model was developed using the vector autoregressive model (VAR). For the choice of the model, only the variables that in the Granger causality test were shown to cause current revenue were used. The choice of model was made using model selection criteria such as R2, adjusted R2, Akaike, Schwarz and Hannan-Quinn information criteria. All tests were performed using panel data. Tests for the construction of the model were performed using Gretl software. The other tests were performed with Eviews software.