MULTIVARIATE ANALYSIS OF THE FISCAL STRUCTURE OF BRAZILIAN STATES: AN APPLICATION OF PRINCIPAL COMPONENT ANALYSIS (PCA)

Abstract

This study analyzes the fiscal structure of Brazilian states in 2024 using Principal Component Analysis (PCA) processed in the R environment (v. 4.5.2). The primary objective was to identify latent patterns of financial autonomy, expenditure rigidity, and investment capacity. The methodology employed indicators of tax burden, investment rates, personnel expenditure, and transfer dependency, sourced from SICONFI and IBGE. Results indicate that two components explain 86.60% of the total variance: PC1 (52.91%), capturing the Autonomy vs. Structural Dependency dimension, and PC2 (33.69%), associated with Tax Pressure and Investment. Factor score analysis enabled the proposal of a management typology, contrasting states of balance and flexibility (e.g., SP and ES) with those characterized by vulnerability and rigidity (e.g., AP and AC). The study confirms theories regarding Brazilian budgetary rigidity, highlighting that federal transfer dependency and payroll costs are the primary fiscal bottlenecks, particularly in the North and Northeast regions. It concludes that fiscal autonomy is a decisive factor for public management modernization and the transition to performance-based controllership models.