Tesla vs Exxon? ESG ratings are like stock opinion reports more than fact. WSJ article unpicks methodology and score of 5 major companies and higlights the wide dispersion of results and different methodologies. (Summary outcomes above)
Like a stock report, you should understand the assumptions and methods to derive “Buy/Sell/Target Price” when utilising analysis. An investment analyst rating is an opinion, so is an ESG rating. The WSJ piece dissects why the ratings are so different. The components of methodology and weighting.
“The problem here isn’t the ESG ratings, but that they are used as though they were some sort of objective truth... they are no more than a series of judgments by the scoring companies...and investors who blindly follow their scores are buying into those opinions, mostly without even knowing what they are.”
“Investors should not treat ESG scores as settled facts to be used on their own, but as potentially worthwhile analysis that needs to be understood before being acted on. The thick ESG reports behind the scores offer useful detail about the policies and controversies around each business. But just as with financial accounts, investing without understanding is unlikely to deliver what you want.”
My Linkedin Post is over 10,000 views and 100 likes and counting, with many of the ratings providers (most who I know) providing comments.
From FTSE:
• The comparison to sell side research & buy/sell ratings is an interesting one. It should be valid but in reality the high correlation & herding that you see in sell side ratings is actually not seen in ESG ratings – as shown in the article.
• As far as I’m aware, no one in the ESG data / ratings market is claiming to provide some form of objective truth. All outputs are based on methodologies and as the article points out – it’s essential to consider how scores and ratings are derived and that as an end investor / user you are comfortable with this.
• For example, at FTSE we have always maintained separate datasets for assessing how a company operates (our ESG Ratings) and the products and services that companies manufacture / provide (our Green Revenues data). We have found that investors appreciate this distinction. Just because a company is contributing to the green economy does not mean that it is doing a good job of managing a range of operational ESG issues…Attempting to net these things off in a single number is problematic.
• Generally speaking, at FTSE we aim to be as transparent as possible – in particular with the issuers we rate e.g. in addition to sharing all of our data with each company in our universe as part of our annual research cycle we now provide (for free) a “Corporate Peer Comparison Tool” which allows companies to see both the scores & ratings (including for Green Revenues) that we have derived as well as comparison to their peers by sub sector, industry, country etc.
From Xi Li (of Active Ownership paper fame):
“Academics have criticized these ESG ratings for a long time.... not surprising. A better way to use these ratings is to compare it in a time-series manner (i.e. Tesla's rating last year vs. current year), not cross-sectionally (i.e. Tesla vs Exxon). However, even this may have problems, because these data providers' algorithm may change over time... “
From Mike Tyrell (of SRI-Connect)
It would indeed be an encouraging evolution if both suppliers and users of ratings came to regard them as opinions rather than in any respect as objective reality. However, I fear that we are still some distance from a situation in which asset managers buy / sell / focus on / ignore a stock bsed in a rating from one provider are treated with the same derision as they would be if they bought / sold a stock based on the recommendation from a single broker. ... and we should note that James Mackintosh's main concern seems not to be the fundamental active managers who can pick and choose ratings and research but the "billions of dollars of exchange-traded funds based on ESG indexes ... [and] ... fund managers [who] are being pushed to produce portfolios with better ESG ratings, encouraged by public mutual-fund ESG scores." There is, it seems to me, a real problem with the same research processes being used to produce pre-trade investment advice (counter-consensus opinion needed) and post-trade portfolio analytics (objective exposure data needed). I can't quite articulate the nature of the problem and I certainly can't see a business model that would fix it ... but I think it's one to watch. … He also comments back to FTSE… Aled Jones - I agree that no-one would openly claim 'objective truth' but I suspect that both sides are guilty of treating opinion as if it were this. There is equally a small number of research firms claiming "this is just the opinion of an analyst"
From Sustainalytics:
…Sustainalytics tries to differentiate itself from its competition by the transparency we provide to our clients. We are not a black box. That said, we don’t provide full methodological details publicly. … but do give more colour on the Tesla rating…Fully agree with your points and many of the points in the article. Sustainalytics agrees that ESG is a separate signal and one that needs to be interpreted alongside other financial information. Also, ESG Ratings and what they measure are not as standardized as traditional financial signals. In regards to Tesla, from our analyst: “Tesla’s management of these issues ranges from one extreme to the other. What it manages well – the carbon impact of its products – it manages very well, while on the other hand, its management of human capital and product governance risks reveals significant shortcomings. As a company committed to the production solely of electric vehicles and other products related to renewable energy, Tesla is the global leader among automakers when it comes to the carbon emissions of its fleet – no small achievement. However, the company has been involved in a steady stream of controversies related to the timely delivery of cars, the safety of its autopilot technology, and its management of its workforce, and despite such controversies, commitments to labour rights and programmes governing product quality are lacking.”
Some quant data looks at some of the ESG ratings sets are here (not peer reviewed): A look at using the Thomson Reuters Data set and by BAML can be found here.
Another lens in a more quant fashion can be found in this MSCI look through its own data and a return on capital lens. And Nordea Quant also look at it through the MSCI lens.
Other papers and thoughts to look at:
Why ESG might be so important in an intangible world
and an academic paper on Active Ownership and why and how it adds value.