Data for Environmental Decentralization and Environmental Pollution in China: A Dynamic Game and Spatial Panel Analysis
Description
This is data for Environmental Decentralization and Environmental Pollution in China: A Dynamic Game and Spatial Panel Analysis.
Files
Steps to reproduce
Data for this study were obtained from publicly available official statistical sources. Provincial-level environmental pollution indicators (industrial wastewater, waste gas, smoke, dust, sulfur dioxide, and solid waste emissions) were collected from the China Environmental Yearbook and China Environmental Statistical Yearbook. Environmental decentralization indicators were constructed using personnel data from the China Environmental Protection Statistical Yearbook, supplemented with provincial population and GDP data from the China Statistical Yearbook. To address missing values, linear interpolation was applied, and all monetary variables were deflated to constant 2000 prices using provincial CPI indices. The environmental pollution index was constructed via the entropy-weighted method, while environmental decentralization indicators were calculated following the approach detailed in Yi and Yu (2023), with additional adjustments for GDP and population heterogeneity. A new alternative decentralization measure based on sub-provincial environmental staff data was also developed for robustness checks. All econometric analyses were performed using Stata 17.0. Static and dynamic panel regressions (including system GMM) were estimated using built-in and user-written commands (xtreg, xtabond2). Spatial panel models (Spatial Durbin Model) were implemented using the xsmle command, with three spatial weight matrices (contiguity, geographic distance, and economic distance) constructed in Stata. Numerical simulations of the three-agent dynamic game model were conducted using Mathematica 11.3 to solve for optimal decisions and perform comparative static analysis. The full Stata and Mathematica code for data construction, estimation, and simulation is available from the corresponding author upon reasonable request.
Institutions
- Central University of Finance and EconomicsBeijing, Beijing