Key points are not available for this paper at this time.
The article aims to assess the effectiveness of financial expenditures in higher education for 31 provinces in China, namely overall, technical and scale efficiency, and to determine whether the returns to scale are increasing or decreasing for each province. The main method used is the three-stage DEA analysis / Data Envelopment Analysis, and software DEAP2.1 and Frontier 4.1 are used for modelling. Financial expenditures on higher education form an input parameter in the DEA model. The output parameters (determining the efficiency of higher education) are the number of students, the number of full-time teachers and the new value added to fixed assets in higher education institutions. The modelling was carried out for 2008-2013 and 2014-2018 (the COVID-19 pandemic is not considered to avoid uncommon trends and unpredictable disturbances in the model). A comparison of the analysis results for these periods allowed the identification of new trends and patterns. The study takes into account three main groups of parameters: 1) endogenous (inefficiency of internal management processes); 2) exogenous (gross regional product per capita of each province in China; the extent of university clustering, which is determined by political and market decisions and does not depend on educational institutions; and the degree of fiscal decentralisation, which is estimated in the article as the ratio of local and state budget expenditures per capita, taking into account the distribution of total budget expenditures for each district and the total population of the respective territory); 3) random disturbances. At the first stage of the study, a BCC model focused on inputs (to assess the technical efficiency of financial expenditures on higher education in different provinces) was built. It does not consider exogenous factors and random disturbances. In the second stage, SFA (stochastic frontier analysis) was applied to account for the impact of all three groups of factors (endogenous, exogenous and statistical noise) on the efficiency of financial expenditures in higher education for each province in China. In the third stage, the adjusted and standardised input and output parameters are incorporated into the BCC models, and the actual value of efficiency caused by exogenous parameters and the influence of random factors is determined. A comparison of the results of each modelling stage for each stage in the study and each province in China allowed the establishment of how the influence of the determinants of higher education financing efficiency changed in chronological and spatial contexts.
Yu et al. (Wed,) studied this question.