Finanças públicas no Brasil: regras fiscais, sustentabilidade da dívida pública e vulnerabilidade fiscal

Abstract

Chapter 1 - Brazil’s New Fiscal Framework: Evidence from a Dynamic Stochastic General Equilibrium (DSGE) Model. This project consists of three essays on public finance. The first essay investigates how macroeconomic variables and the trajectory of public debt respond to the proposed implementation of a fiscal rule along the lines of the New Brazilian Fiscal Framework. To this end, a New Keynesian model incorporating features of the Brazilian economy is calibrated and simulated within a Dynamic Stochastic General Equilibrium (DSGE) framework. The results indicate that, in the absence of a fiscal rule, government spending and public debt follow an upward trajectory. Conversely, the implementation of a fiscal rule that limits government spending leads to a downward debt trajectory, highlighting the positive impacts of fiscal control on debt sustainability and trajectory. Chapter 2 - Sustainability and Public Debt Limits: Application of a Fiscal Fatigue Model to Brazilian States. The second essay analyzes the fiscal sustainability of Brazilian states from 2001 to 2020, employing various econometric methodologies alongside a Fiscal Fatigue Model. The study empirically confirms the fiscal fatigue hypothesis for Brazilian states, using a quadratic reaction function that enables the estimation of debt limits and fiscal space for each state. The results reveal heterogeneous fiscal scenarios: while some states, such as Minas Gerais, Rio de Janeiro, Rio Grande do Sul, and São Paulo, exhibit unsustainable debt trajectories and negative fiscal space, other states display higher debt limits and a gradual reduction in debt over time, supported by low growth-adjusted interest rate differentials. Additionally, Laffer curves were estimated for the most fiscally distressed states, indicating that, except São Paulo, the other states are already operating near the optimal tax rate, thereby limiting room for further tax increases. These findings offer valuable insights for state-level fiscal management by providing indicators for debt monitoring, budget margin analysis, and economic risk assessment, while also informing fiscal policy decisions by accounting for states’ adjustment capacity in both debt and revenue dimensions. Chapter 3 - Fiscal Vulnerability Index and Public Debt Forecasting for Brazilian States. The final essay develops a Fiscal Vulnerability Index to assess the fiscal fragility of Brazilian states, based on indicators of indebtedness, solvency, liquidity, budgetary rigidity, and financial autonomy, through Principal Component Analysis. Additionally, it applies machine learning techniques to forecast public debt. Results indicate that methods such as Lasso, Ridge, and combinations with non-linear techniques outperform traditional debt forecasting approaches. The Fiscal Vulnerability Index and Own Source Revenue emerged as significant predictors, validating the Index as a reliable tool for identifying fiscal risks. This study contributes to fiscal governance by providing innovative indicators and methodologies that enhance the forecasting, monitoring, and management of the Brazilian states’ public debt.