In the first two parts of this paper, we discussed the degree of heterogeneity that exists between the scores of different ESG rating providers and, in a second step, the importance of combining these different sources to improve the estimation of an ESG market factor.

Read part 1 on data heterogeneity

Click here

Read part 2 on the existence of an ESG risk factor

Click here

In this third and last part, we want to explore what benefits, both conceptual and practical, may result from the use of alternative data in addition to existing fundamental data. More concretely, nascent data based on sentiment analysis, estimated from recent artificial intelligence techniques, seem to be particularly interesting thanks to their complementary characteristics. This approach is part of the same informational complementarity objective that we have discussed so far and shows how the concept of “universal” ESG involves the measurement and analysis of different more directly observable variables.

Highlights

  • Sentiment analysis based on the use of different artificial intelligence techniques is conceptually complementary to fundamental ESG analysis methods
  • Empirically, an ESG sentiment metric built from the analysis of various written sources, behaves as a fundamental score, and provides an additional level of information
  • The use of this type of data is therefore part of an informational logic that makes it possible to estimate a less biased ESG factor
Written in March 2022

This material is provided for informational purposes only and should not be construed as investment advice. The views and opinions expressed may change based on market and other conditions. There can be no assurance that developments will transpire as forecasted. Actual results may vary.

Seeyond is an affiliate of Natixis Investment Managers
Seeyond - Registered Office: 43 Avenue Pierre Mendes France – 75013 Paris – France
A limited company with capital of € 4 963 183 - Regulated by AMF under n° GP 17000034
Company Trade Registration (RCS) Number 525 192 720 Paris