Making sense of divergent third-party ESG data ratings

Key points

• Third-party environmental, social and governance (ESG) ratings play an important role in asset management, but we believe their application needs to be better understood.
• We believe that simply taking third-party ESG ratings at face value could result in the mispricing of securities or the inappropriate inclusion of securities in sustainable portfolios.
• Newton welcomes a recent paper on the topic from the Centre of Endowment Asset Management at Judge Business School, Cambridge University. It is as part of a growing body of literature showing the disparity in ESG data by highlighting inconsistencies and biases across data providers – in terms of both the inputs to the ratings and the divergence of outputs.
• We believe that asset managers need to build capacity to manage and interpret the data that they buy.

Third-party environmental, social and governance (ESG) ratings play an important role in asset management, but we believe their application needs to be better understood. Newton has long held the view that external ESG ratings or scores should not be used in isolation when assessing companies, but as an input into the process. At Newton, our consideration of ESG issues more broadly is part of a multi-dimensional research approach that integrates financial analysis, thematic trends, macro-economics, investigative research and valuation considerations as appropriate.

Monitoring risks

We believe sustainability is a fundamental part of the mosaic of issues required to fully appreciate the material risks and opportunities influencing the securities and instruments in which we invest on behalf of our clients. External ESG ratings provide an opinion that is independent of the asset manager, and can therefore function as a counterweight to, or sounding board for, internal views. The ESG rating datasets also offer wide coverage across a universe, which can be helpful in monitoring ESG risks within portfolios, as well as providing an indicative view that supports the screening for new investment ideas.

Nonetheless, it is important to acknowledge that the information on which third-party ESG scores are based is subjective and dependent upon often opaque methodologies. It is based on unaudited data, and is proprietary to the data vendor. We believe that simply taking third-party ESG ratings at face value could result in the mispricing of securities or the inappropriate inclusion of securities in sustainable portfolios. Indeed, an asset manager’s choice of one ESG vendor over another could strongly determine the investment outcome.

Paper highlights data disparity

With this in mind, we find a recent paper distributed by the Centre of Endowment Asset Management (CEAM) at Judge Business School, Cambridge University helpful.[1] The November 2020 paper, ‘Divergent ESG Ratings’, is authored by Professor Elroy Dimson (CEAM), Professor Paul Marsh, and Dr Mike Staunton (London Business School). The authors point to this lack of consistency in showing substantial disparity in the rankings of 878 US companies across two leading ESG rating providers. Inconsistency is further illustrated in the authors’ in-depth analysis of the ratings of six large US companies which diverge significantly across three ESG rating agencies in terms of their overall ratings, and even more so across the pillars that comprise those overall ratings.

An implication of this disparity is that asset managers need to build capacity to manage and interpret the data that they buy. This requires a dedicated ESG data focus and function, supported by a data strategy that aims to plug gaps or weaknesses in available data. The data that lies behind headline ESG ratings is often the most useful aspect of the datasets that are available. For us, the most effective method is when this underlying data can be selectively and carefully integrated from multiple ESG providers in a way that makes the most sense for the asset manager.

Of course, not all the data within an ESG rating will be relevant or aligned with the fundamental view of the asset manager or the interests of its clients, and it is up to asset managers themselves to identify the datapoints that are most relevant to them. This is a difficult exercise, even when working with the raw data that underpins many ESG Ratings. Data consistency is a challenge here too, since the same data point can vary across providers, owing to different methodologies for data collection and estimation. We have seen considerable variation in carbon-emissions data, for instance, across three different providers when comparing data for individual companies.

Growing push for ESG ratings data regulation

Newton welcomes the CEAM paper as part of a growing body of literature that shows the disparity in ESG data by highlighting inconsistencies and biases across data providers – in terms of both the inputs to the ratings and the divergence of outputs. We believe that this type of research may be increasingly useful for asset managers, companies and ratings agencies as securities regulators in several jurisdictions – including the European Union and India – have been pushing to regulate the ESG ratings market. It is also helpful as asset managers strive for greater transparency in their sustainable investment processes and the data that underpins them, and for enhancements to their investment decision-making.

To find out more on this topic, you can download the CEAM ‘Divergent ESG Ratings’ paper.

[1] Elroy Dimson, Paul Marsh and Mike Staunton, ‘Divergent ESG Ratings’, The Journal of Portfolio Management, November 2020, 47 (1) 75-87.

Authors

Alexander Parkinson

Co-head of Responsible investment data integration