Aumentar receitas ou cortar gastos? Discutindo o nexo entre receitas e despesas do governo central brasileiro^♦

Abstract

The aim of this study is to analyze the short and long term intertemporal relationship between spending and revenue variables of the Brazilian federal government. Using monthly data 1997 to 2015 the study applied cointegration techniques and estimates asymmetric error correction models (ECM-TAR and ECM-MTAR). As a robustness analysis a simulation of shock was performed on variable revenue and expenditure of the central government. The results show that the government respects its budget constraint only for the period from 1997 to 2013. During this period there is evidence, in the short term, of a one-way causal relationship type tax-spend, where a tax reduction generates an increase in spending and the long-term causal relationship is bidirectional. Could not unlike previous studies, identify evidence of a causal relationship of spend-tax type. The results also disproved the traditional interpretation that the way to control spending is to reduce government revenues. Overall, the results show that by incorporating the analysis of the years 2014 and 2015 the central government fails to respect their budget constraint and cannot establish a causal relationship between income and government spending.

Keywords
Revenue; Expenditure; Fiscal illusion; Error Correction Models