Using multiple data sources to evaluate a company’s environmental, social, and governance (ESG) performance has become commonplace for investment professionals. As ESG issues are becoming increasingly important to asset owners, regulators, and investors, incorporating multiple sources for ESG information offers a means to support performance evaluations—despite the lack of standardized disclosures, transparent scoring methodology, and consistent data availability.
Unfortunately, much of the information used by major ESG data providers is based on the same company disclosures. Although analytics providers have different ways of addressing the challenges associated with inconsistent disclosures and aggregating the data into scores, what companies say about themselves is an important input.
For a complete view, investors must consider information from other sources such as news articles, government regulatory findings, and reports by non-governmental organizations (NGOs). To this end, several data providers apply artificial intelligence (AI) technology to analyze these sources and extract meaningful data to incorporate into ESG scores.
1. Positive and negative views of a company.
Every company is incentivized to present itself in the best light. Because disclosure requirements around ESG issues are still emerging, companies have considerable discretion to omit unflattering facts from their ESG reports. News articles and regulatory filings, on the other hand, cover a range of issues that stakeholders consider relevant—without a company’s inherent conflict of interest.
2. Real-time monitoring and alerts on ESG issues.
When an ESG-relevant issue arises, media coverage demonstrates (in real time) the concerns of stakeholders, including workers, customers, neighbors, regulators, and shareholders. ESG scores based on external signals incorporate the latest information. In many cases, local sources are the first to capture positive or negative information related to corporate impacts on the surrounding community, environment, and employees. ESG datasets incorporating news are most likely to pick this up if they capture sources in the local language. On the other hand, most company-provided ESG data appears in once-a-year disclosures. ESG data powered by AI can highlight significant events as they happen, triggering alerts for portfolio managers and signals for quantitative trading models.
3. Timely updates to the ESG factors stakeholders see as most important for a company.
In today’s complex world, what matters most to investors, customers, communities, and regulators changes constantly. One event—a product recall, a labor crisis, or an environmental scandal—can prompt stakeholders to rethink how they evaluate ESG performance. Traditional ESG rankings based on annual corporate disclosures generally have a fixed framework for determining what issues to include and how to weight them. By contrast, systems that monitor external sources can identify and send alerts when topics become more and less important to stakeholders. These signals can dynamically change the weightings for factors used to evaluate a company or an industry.
4. Signals for quantitative models.
For quantitative investors, ESG data from external sources can improve the absolute return of their models. Using reliable and timely data to quantify ESG information from unstructured text yields a strong stock selection factor that is additive to performance. By comparison, scores based on corporate disclosures generally change on an annual basis, which is often too slow for quantitative strategies.
5. Intel on private companies.
Private markets have seen significant growth in the last 10-15 years as institutions turn to alternative assets, including private equity, to chase higher returns. These same institutions are weaving ESG into investment processes, both to satisfy regulations like the EU’s Sustainable Finance Disclosure Regulation (SFDR), and to meet client demand. 75% of alternative fund investors surveyed by EY said that their scrutiny of alternative managers’ ESG policies has increased over the past couple of years. Although some private companies disclose ESG data, disclosure mechanisms don’t compare with public companies. ESG scores that don’t rely on annual public disclosures can pick up important information about private company ESG behavior by incorporating news and publications that are specifically focused on private markets.
6. Insights into how a company’s business affects its reputation.
A key technique for analyzing external data is sentiment analysis, which uses AI to determine if given text has a positive or negative view of the company. Companies that have engendered significant positive sentiment around their ESG activities may see benefits in customer perception, brand loyalty, and in raising capital. Negative sentiment may well make doing business harder. If there is a significant gap between the external view of a company and its self-reported performance, it may signal that a company is greenwashing, in which case traditional ratings deserve greater scrutiny.
Data from outside-in sources is most useful when integrated with an investment firm's core workflow. The completeness and consistency of information is a challenge for all types of ESG data. Adding additional perspectives can help fill gaps and provide a more balanced view. One common approach is creating one or more internal scores that combine multiple data sources to reflect the firm’s ESG priorities.
Mackenzie Hargrave contributed to this article.
This blog post is for informational purposes only. The information contained in this blog post is not legal, tax, or investment advice. FactSet does not endorse or recommend any investments and assumes no liability for any consequence relating directly or indirectly to any action or inaction taken based on the information contained in this article.