The hunt is on for more independent sources of information about Environmental, Social and Governance performance. Given the obstacles presented by the coverage, comparability, and reliability of companies’ self-reporting on their own Environmental, Social and Governance performance, investors and other stakeholders are beginning to seek out more alternative and independent sources of information. This in turn is driving greater innovation when it comes to ESG data collection and analytics. The first category of this data derives from advances in physical sensors - which power the Internet of Things, satellite imagery, and remote sensing capability more generally. For example, satellite images of flaring at oil and gas company operations can be used to detect and ultimately eliminate this wasteful and polluting phenomenon, which is a prominent source of greenhouse gas emissions. Meanwhile newer satellites equipped with high-resolution imaging technology can be used to detect concentrated sources of methane emissions from livestock - a not-insignificant cause of harmful global warming - and problematic areas of fossil fuel extraction. And, small robotic sensors can be deployed to patrol sewage systems near corporate facilities, and identify disease outbreaks or even opioid use among employees. The second category of ESG information results from advances in information and communications technology platforms such as e-government and social media sites. Scraping information from publicly-available government databases can help track down filed lawsuits, workplace injuries, reports of environmental pollution and other indicators of ESG risk. Meanwhile social media services like Twitter, LinkedIn, and Glassdoor (as well as blogs and traditional news sources) are available for data scraping and natural language processing that can identify the prevailing sentiment about firms among current, prospective and former employees, customers, and investors. While it is not a source of information per se, the analytics that become possible when combining organic data sources like those mentioned here with self-reported information from firms can be invaluable. Machine learning algorithms may be able to identify and predict ESG risks using an array of physical, social media-based, and reported data with greater reliability than any single data source. This can help investors and other stakeholders maximize the benefit of available ESG information, and identify the social and environmental hazards most likely to occur.
ESG Data Collection