Virtual Water Resource Utilization Efficiency and Influencing Factors in China's Tertiary Industry: An Input-Output and Stochastic Frontier Analysis

This study aims to investigate the virtual water utilization efficiency and its influencing factors across 14 industries of China's tertiary industry during 2002-2020. By applying the input-output model, Shephard water distance function, and stochastic frontier analysis (SFA), this research incorporates multi-factor analysis with the total virtual water footprint as the water input indicator. Results show that the annual total virtual water footprint of the tertiary industry exhibited fluctuating changes, influenced by macroeconomic conditions, industrial structure, and water-saving policies. The overall virtual water utilization efficiency increased, though the growth rate decelerated, with significant disparities across industries: high-efficiency industries were concentrated in specific fields, while low-efficiency industries were mostly traditional service industries. Stochastic frontier analysis indicates that most estimation errors originated from the technical inefficiency term. Technical inefficiency analysis reveals that factors such as water resource endowment are significantly correlated with water-related technical inefficiency. This study provides a basis for deepening the understanding of water resource utilization in the tertiary industry and offers references for optimizing water resource management policies.