ESG ratings may be too simplistic

The decision to remove Tesla from the S&P 500 ESG index earlier this year struck an odd note to many. Its products have been a catalyst in electrifying transport. Yet it was excluded due to its shakier record on labour rights, the greening of its production facilities, and issues with its battery supply chains. These factors weakened its overall ESG rating on which the index decision was based.

Tesla board member Hiromichi Mizuno, a pioneer of ESG investing in Japan, accused the ratings provider of giving too much weight to negative impacts and not enough to positive ones. There may be some truth to that, but it is also clear that a single, data driven ESG rating cannot and should not try to be everything at once.

Multiple ratings

At a minimum, there may need to be two distinct ratings: one to reflect a company’s positive impact and another its negative externalities. Attempting to capture both in one score dilutes the informational value of the final rating – as the positives and negatives inevitably cancel each other out, resulting in a rating that fails to represent either and can’t be relied upon to guide capital allocation decisions.

One option is to focus the ESG rating on the negative impact a company has and use another system, such as the United Nation’s sustainable development goals (SDGs) framework, to assess its positive business activities, e.g., products designed to tackle climate change. Tesla, for example, naturally has a high SDG score because its revenue comes from selling electric vehicles and green tech, both crucial elements in the push to reach net zero. That doesn’t mean its lower overall ESG rating isn’t important but, taken together, both ratings give a clearer picture of where its impact lies and can help with capital allocation.

If an investor’s goal is to allocate capital to climate solutions, then the SDG rating should be the preferred driver. If the investor has no preference over the product or services a company produces but wants to invest only in companies that behave in an environmentally and socially responsible manner, then the ESG rating should dominate. In practice, the two ratings are used together, with investors targeting certain impacts with the SDG rating while maintaining minimum ESG standards by setting a threshold for the ESG rating.

Weighting the numbers

There are technical reasons why current ESG ratings don’t fully capture a company’s level of sustainability. One is that providers often combine varying E, S and G scores into a headline rating, assigning a weight to each pillar and averaging them out. This approach works better the closer it gets to the two ends of the ESG spectrum. For example, a very high ESG rating usually indicates that a company performs well across all three pillars, and vice versa.

For most companies that fall in the middle, however, the headline ESG rating could be misleading. A company with poor ‘E’ practices could still be included in an ESG fund if its environmental score is sufficiently smoothed by an above-average performance on social and governance goals. This explains why fossil fuel companies with little interest in the energy transition may score unexpectedly highly on ESG, even as a company like Tesla falls down the rankings, and why investors need to pay close attention to ESG fund holdings.

It also matters whether companies are scored on a relative or absolute basis. Relative scoring is essentially a ranking within an often questionable ‘peer group’ rather than a true assessment of sustainability. It can lead to a company that performs poorly on ESG on an absolute basis getting a top rating because its peers are doing even worse. Relative scores may also change not because of anything the company does but simply because the peer average has changed, perhaps due to new additions or reclassifications.

At a portfolio level, such an approach can be meaningless. A portfolio of best-in-class coal mining companies, for example, judged by relative ratings, could appear better on ESG than one with an average range of finance companies.

Data versus analysis

Even if investors use two ratings and apply them on an absolute basis, they may still not tell the whole story, especially where ratings are data-driven or have a formulaic qualitative overlay. To get a complete picture requires a deeper type of company-specific analysis.

This is not always possible for ESG analysts who run the quant models and apply the overlays, as they generally lack the depth of knowledge required to make sense of the ESG data in the context of a company’s business. It is however often possible for seasoned, bottom-up fundamental analysts familiar with ESG methodologies, who meet with companies regularly and can gain first-hand insights into a company’s ESG practices and plans through these meetings.

Today’s quantitative ESG ratings are designed in a way that frequently exposes them to being misunderstood or misapplied, and one system may rate a firm entirely differently to another. This presents challenges for investors seeking to assess sustainability, build portfolios and measure outcomes. Over time, we expect to see further standardisation and more complementary use of different rating types, though qualitative insights will remain essential. Integrating ESG into the investment process isn’t so much a destination as a journey – one that gets better as more people undertake it.