ENGO Participation and Firm Emissions: Evidence from China’s PITI Program

This dataset is constructed to examine the role of environmental non-governmental organization (ENGO) participation in mitigating governance distortions under China’s decentralized environmental regulatory system. The main research hypothesis is that third-party information disclosure can alleviate the principal-agent problem between central and local governments, thereby improving environmental outcomes.
Using the introduction of the Pollution Information Transparency Index (PITI) as a quasi-natural experiment, the dataset enables empirical analysis of its impact on firm-level pollutant emissions. The key outcome variables include chemical oxygen demand (COD) and sulfur dioxide (SO2) emissions.
The dataset combines firm-level, city-level, and policy-level information to construct a firm-year panel. Firm-level data include financial characteristics and pollution emissions. City-level data capture economic conditions, industrial structure, fiscal capacity, and environmental characteristics. Policy variables identify cities subject to environmental governance initiatives such as PITI, low-carbon, and two-zone pilot programs.
The data allow researchers to examine how ENGO participation influences environmental performance. Empirical results based on this dataset show that information disclosure is associated with significant reductions in firm-level COD and SO2 emissions. Mechanism analysis indicates that this effect operates through strengthened enforcement incentives (reflected in reduced rent-seeking behavior) and the upgrading of regulatory instruments (reflected in increased pollution levy collection), both of which contribute to stricter local environmental regulation.
Heterogeneity analysis suggests that the effects are more pronounced among firms with higher administrative rank, larger size, higher pollution intensity, and greater tax contributions. Further decomposition analysis indicates that emission reductions arise from both cleaner production processes and enhanced end-of-pipe treatment.
The dataset is compiled from multiple sources, including the Annual Environmental Survey of Polluting Firms (AESPF), the Annual Survey of Industrial Firms (ASIF), CSMAR, WIND, and various Chinese statistical yearbooks. Policy variables are constructed from official government documents.
This dataset can be used to replicate all empirical results in the associated research and to support further studies on environmental regulation, information disclosure, and firm behavior in developing economies. Detailed instructions for data processing and replication are provided in the accompanying Readme file.