Scientific Beta: Do ESG scores have negative impacts?

January 16, 2024

By Erik Christiansen, ESG Investment Specialist at Scientific Beta

It is well known that ESG scores diverge widely across providers, putting their reliability into question. This divergence also reduces any positive real-world impact investment strategies might have that are built on such scores, as investors fail to focus their efforts on a common direction.

More recent research has shown that ESG scores actually have negative impacts, as they rely heavily on lofty corporate promises, which serve to hide current harmful corporate practices. And except for carbon emissions reduction targets, ESG promises are not realised by companies in the future either.

Can we make ESG scores converge?

Numerous academic papers have documented the divergence of ESG scores across providers. To some extent this lack of common assessments of companies’ ESG ‘performance’ reflects the confusion around what the different offerings seek to measure. Investors often have overlapping goals when defining their ESG strategies: avoiding complicity with practices or products they condemn, having a positive impact on social and environmental outcomes in the real world, while perhaps simultaneously hoping to avoid underpriced financial risks or uncover underpriced opportunities.

With investors often failing to clearly spell out their objectives, ESG data providers could be forgiven for lacking clarity in the purpose of their ESG scores: maintaining some ambiguity means the same product can cater to more prospective buyers. In particular, it is sometimes unclear which perspective of double materiality they try to shed light on: do they assess a company’s impacts on its stakeholders, or rather the financial risks/opportunities these stakeholder impacts generate for the company?

In a perfect polluter-payer world, where companies bear the full cost of their negative externalities (and reap the full benefits of the positive ones), both perspectives are equivalent. In the real world they are not.

So perhaps the divergence of ESG scores simply reflects the varying scopes of what they seek to assess? This would mean that the standardisation of definitions, for example through regulation, could lead to agreement among providers. Or, with enhanced transparency on raters’ definitions and methodologies, one might hope that the users of the scores could understand – and even themselves correct – the disagreements.

Unfortunately, academic research has dashed such hopes. Scores also diverge because of differences in data sources, the choice and (subjective) treatment of assessment criteria, as well as their weighting and aggregation methodologies.

Rating divergence dilutes investor impact

Investors seeking to influence companies and thereby to have an impact on real-world outcomes should seek synergies between investment decisions and engagement. These synergies require investment decisions to send clear, predictable, and consistent signals to company managers.

The less investors agree in their trades, risk/return assumptions and preferences, the lesser the impact: if investors share the same ESG goals but pull in different or even opposing directions, it seems intuitive that their efforts will be diluted and eventually cancel out.

Lofty promises hide adverse impact

Since companies who can convince the raters to agree on a good score are favoured with lower cost of capital (which is equivalent to higher stock prices and stock option values), they have a clear incentive to get a great score. And to compute ESG scores, the providers take both companies’ policies and their realised outcomes into account.

This is where recent research brings disturbing evidence. Companies can inflate their scores by making cheap, lofty promises. And the worse their practices are, the stronger the incentive to compensate with glossy policies. Perhaps the raters are justified in putting more emphasis on policies than on outcomes? In effect, outcome metrics are to some extent backward looking. At the very least there is a lag between when the data are published and applied, and when the events occurred.

If policies were a proper proxy for future outcomes, policy focused ESG scores would still be relevant to investors who seek to change the world, rather than simply buying companies that are already virtuous.

Glimmer of hope

There is a glimmer of hope, however, as there is one area where companies appear to carry through their promises, and that is carbon emissions reductions. In more recent years, reduction promises have predicted greenhouse gas emission scores that were later realised. In its latest progress reports, the Science Based Targets initiative has also shown that companies with approved emission reduction targets decarbonise their activities at a quicker pace than other firms.

The attention such climate claims receive, and the relative ease with which their realisation can be verified, mean that companies are warier of making empty promises. Among the multitude of ESG dimensions being analysed and scored, climate change is thus an exception. But an important exception.