Artificial intelligence and, specifically natural language processing (NLP), could enable investors to access new data sources that will help assess the ESG quality of issuers, especially in emerging markets.
“Unstructured data provides access to new types of information, but it also provides colour and perspective to data already in hand,” said Caroline Le Meaux, global head of ESG research, engagement and voting, at Amundi.
This is based on new research from the asset manager, in conjunction with the IFC, illustrating the value of unstructured data as a source of under-utilised insights into corporate ESG performance.
Such data could also help issuers make sure the data available are more efficiently communicated and used, and therefore might combat what Le Meaux described as “ESG reporting fatigue”.
“It is furthermore a way to limit biases due to discrepancies in the means dedicated to corporate social responsibility communication. NLP is therefore particularly useful in the emerging market space,” she added.
Seeking new data sources
The ratings of ESG data providers for both developed and EM issuers have come under criticism for using different definitions of materiality as well as varying indicators, weights and scoring methodologies to calculate ESG ratings.
At the same time, there is scepticism among investors and other market participants about data providers that use different methods to source raw data and to calculate proxies to estimate missing information.
The new report by Amundi and the IFC – Artificial Intelligence Solutions to Support Environmental, Social, and Governance Integration in Emerging Markets – reveals the potential of algorithms such as ESG NLP to analyse vast amounts of unstructured text from sources.
“Today, because of regulation differences, disclosure is different from one region to another,” Le Meaux said. “There are nevertheless data sources that are underused and could, if correctly analysed, strengthen our ESG assessment.”
These include environmental and social impact assessments, multilateral development bank project disclosures, non-governmental organisation and civil society organisation reports, social media, industry research and regulatory reports. Analysis by document type and in combination enables the sentiment profiles of different sources of information to be compared.
Key data attributes
Both the quality and traceability of data are also key parts of effective ESG analysis.
This requires access to raw data. “[This] is essential for a sound and fruitful dialogue when it comes to engaging with companies to discuss their views and to influencing and encouraging their adoption of best practices and their positive impact on key societal issues such as the Sustainable Development Goals,” said Le Meaux.
“Having access to good quality data with a large breadth, on a large number of issuers will improve the quality of Amundi’s ESG analysis as well as the equal treatment of issuers,” she added.
It might also help address an emerging misperception among investors that ESG strategies are pandemic-proof. This has arisen in line with the tendency of ESG strategies to overweight sectors that have proven resilient and thrived in the past 12 months, such as healthcare and technologies, and underweight industries impacted negatively by Covid-19, such as transport, energy and materials.
Investing in resources
To support the demand for ESG investment opportunities, research also shows that asset managers have beefed up their capabilities in this area.
On average, ESG teams at the 30 largest fund houses have grown more than 230% between 2017 and 2020. This highlights the extent to which the industry is expanding the analytical capacity to service the demand for ESG integration.
This article first appeared in ESG Clarity Asia.
