Quantitative Evaluation of Budget Performance Management Policy: Evidence From China
利用LDA主题模型、自然语言处理和PMC-TE指数模型，分析了2001至2024年间2501项中国预算绩效管理政策，揭示了九类政策工具及其演化阶段，为政策制定者优化资源配置提供参考。
ABSTRACT The quantitative assessment of budget performance management policies is vital for improving the scientific rigor of policy formulation, enhancing the practicality of policy implementation, optimizing public resource distribution, and advancing the refinement of the national governance system. This research, which analyzes 2501 Chinese budget performance management policies from 2001 to 2024, utilizes latent Dirichlet allocation (LDA) topic model, natural language processing (NLP), text encoding techniques, and the PMC‐TE index model to quantitatively evaluate the effectiveness and evolutionary trends of China's budget performance management policies at various developmental stages. The result reveals that China's budget performance management policies can be grouped into nine primary policy tools. Notably, financial budget management, resource allocation, and operation management are the most frequently employed and effective policy tools. These policy tools are further divided into three policy domains: financial strategy and social development, financial supervision and innovative management, and financial governance and reform. Furthermore, the historical evolution of China's budget performance management policy tools displays a four‐stage development pattern: foundational and preliminary development, rapid expansion and deepening, peak and adjustment, and optimization and transformation. These findings offer a novel theoretical framework for understanding the progress and patterns of the evolution of China's budget performance management policies.