Arrecadação tributária do ICMS no Estado do Ceará: uma análise da capacidade preditiva

Abstract

The present dissertation aims to verify the predictive capacity of a set of variables regarding the growth of the Tax on the Circulation of Goods and Services (ICMS) in the state of Ceará, Brazil. A multiple regression model was used for this purpose, and six variables were selected to assess their predictive power concerning this tax revenue. Based on the analysis of the tax composition in the state, with the main segments being industry, commerce, and electricity, the following variables were selected: Industrial Production in Ceará, Nominal Revenue from Retail Trade, and Electricity Consumption in the state. Additionally, Federal Revenue, Inflation (based on the IPCA), and Exports were also included. The estimations derived from the research, based on these statistical series with monthly data from 2000 to 2011, as well as projections carried out for the period from March to August 2011, demonstrate that while some variables exhibit significant predictive power, others are not useful for estimating ICMS revenue in Ceará. The findings of this study may support revenue forecasting and contribute to improving the state's budget planning.