The Use of Data Envelopment Analysis to Assess the Efficiency of Public Security Spending in Brazil from 2015 to 2024: A Dynamic and Multi-Stage Application.
Public security; Efficiency analysis; Data Envelopment Analysis.
Given Brazil’s high crime rates—which demand more effective public policies to combat and prevent crime and violence in order to ensure citizens’ safety and the protection of individual rights—this study proposes assessing the efficiency of public security spending using Data Envelopment Analysis (DEA). To this end, the 27 federative units (states and the Federal District) are considered as decision-making units (DMUs). Data were collected from the Brazilian Public Security Yearbook, covering the period from 2015 to 2024. Total public security expenditure and the size of the police force in each federative unit are used as inputs, while outputs are derived from the ratio between population and the number of intentional violent deaths, defined as a Safety Index (SI). The study is expected to reveal substantial heterogeneity across federative units, identifying a set of efficient units (on the efficiency frontier) that can serve as benchmarks for the others. It also aims to estimate, for inefficient units, potential improvement targets for the SI through the reduction of waste and/or a more effective combination of spending and personnel, highlighting slacks and feasible adjustment goals. Additionally, the analysis is expected to map temporal patterns (2015–2024), examining whether efficiency improved or deteriorated over time and whether some units remained persistently inefficient, suggesting structural issues in management and resource allocation. As a contribution, the study seeks to provide evidence-based inputs for policymaking by indicating opportunities to optimize spending, prioritize strategies, and allocate resources more efficiently, ultimately strengthening planning and improving public safety outcomes.