Visualizing environmental management: Corporate environmental images information disclosure and idiosyncratic risk

Enterprises pay more attention to the online visual communication of environmental information. Images significantly contribute to attracting the attention of stakeholders and enhancing the readability and under-standability of information. This study collected 20,699 images of environmental information published on social media by environmentally sensitive Chinese enterprises from 2016 to 2020. Large-scale automatic image content analysis technology based on a computer vision algorithm was used to mine the disclosure characteristics of the environmental images. We further used a regression analysis to explore the effect of environmental images in-formation disclosure on idiosyncratic risk. Our results show that enterprises are more willing to publish envi-ronmental management awareness images than renewable energy, environmental protection facilities, environmental certificates, and greening images. Enterprises with higher idiosyncratic risk and negative envi-ronmental events may disclose more environmental image information for impression management. The regression results show that environmental image disclosure can reduce idiosyncratic risk. However, this negative correlation no longer exists when idiosyncratic risk is high and negative environmental events occur. In conclusion, this study proposes a new perspective for examining the characteristics and effects of environmental information.