Modelos para previsão de receitas tributárias: o ISSQN e o IPTU das cidades mais populosas da Bahia

Abstract

Brazilian legislation establishes fiscal balance between public revenues and expenses. The collection of public revenue is essential for the strategic planning of budgets in municipalities. Law 4,320/1964, the Federal Constitution of 1988 and the Fiscal Responsibility Law/2000 form the triad that originated the guidelines for budget planning in Brazilian public finances. Consubstantiated by regulatory laws, the Municipal Audit Courts regulate local standards for compliance with legislation, among them, the requirement to forecast revenue and fix expenses, which has encouraged scientific discussions about the quality of forecasts and the methodologies that can be used for public budget forecasting. The objective of the study is to verify the efficiency gain in forecasting revenues from municipal taxes, ISSQN and IPTU, using Box-Jenkins, Exponential Smoothing and Artificial Neural Networks approach methodologies, as there is a lack of literature that specifically addresses these taxes. Data from the period from 2006 to 2022, from municipalities in Bahia with a population of more than 100 thousand inhabitants, were analyzed using Time Series methodologies: Box-Jenkins, Exponential Smoothing and Artificial Neural Networks approaches. The results of the analyzes revealed that the forecasts using the econometric methods discussed were mostly superior to the municipalities' forecasts, which tends to improve the quality and reliability of municipal revenue forecasts. Another relevant finding refers to the quality and reliability of the information provided by municipalities. The levels of error and the recurring repetition of forecast values in consecutive years in some municipalities indicate concern regarding the quality of the data being published. The results suggested that there is an improvement in forecasts if econometric techniques are used. It is up to managers and technicians to evaluate which methodology suits the characteristics of their municipalities.