Replication package for “Rule-of-Law Digital Government and Corporate Social Responsibility Governance in Digital Firms: Evidence from China”
Description
This replication package contains the data and Stata code required to reproduce the empirical results reported in the manuscript “Rule-of-Law Digital Government and Corporate Social Responsibility Governance in Digital Firms: Evidence from China.” The package includes (i) the core firm-year dataset used as the merge backbone, (ii) auxiliary raw component datasets used to construct controls and additional measures, (iii) datasets for robustness checks, heterogeneity analyses, mechanism tests, and instrumental-variable specifications, and (iv) a master Stata do-file that performs the merges, data cleaning (including winsorization and variable generation), and estimation. Key external data sources include CSMAR (China Stock Market & Accounting Research Database), Huazheng ESG ratings (official website: https://www.chindices.com/esg-ratings.html), Bloomberg ESG score (https://www.bloombergchina.com/solution/sustainable-finance/ ), and official Chinese statistical yearbooks (National Bureau of Statistics of China: https://www.stats.gov.cn/ ). Some inputs (e.g., policy indicators and selected historical exposure measures) are hand-collected from official policy documents and historical materials; the resulting variables are provided in the deposited datasets. A detailed README is included in the files and documents the contents, variable construction notes, and reproduction steps.
Files
Steps to reproduce
Download all files in this dataset and place them in a single folder (recommended). Open Stata (Stata 15 or later recommended) and set the working directory to the folder containing the files. Install required user-written packages if needed (e.g., reghdfe, ivreghdfe, winsor2, coefplot) via SSC. Run the master script: do "Economic Modelling – Replication Code.do" The script (i) merges the raw component datasets, (ii) performs data cleaning and variable construction (including winsorization and derived indicators), and (iii) reproduces the regression outputs corresponding to the tables and figures in the paper. Output files (logs/tables/figures) are saved to the directories specified at the top of the do-file.