This is a post made by our CEO Shai Hill on LinkedIn on 10th March 2023.
The large legacy ESG ratings brands use a vast amount of estimated data.
That's how they are able to cover >10,000 companies, including emerging market companies that don't actually publish any ESG data.
There are 3 problems here:
- 𝗧𝗵𝗲𝘆 𝗱𝗼𝗻'𝘁 𝗺𝗮𝗸𝗲 𝗰𝗹𝗲𝗮𝗿 𝘄𝗵𝗮𝘁 𝗰𝗼𝗺𝗽𝗮𝗻𝘆 𝗱𝗮𝘁𝗮 𝗶𝘀 𝗮𝗰𝘁𝘂𝗮𝗹 𝗮𝗻𝗱 𝘄𝗵𝗮𝘁 𝗶𝘀 𝗲𝘀𝘁𝗶𝗺𝗮𝘁𝗲𝗱.
Depending on your subscription, you can sometimes click through to a CO2 emissions number, but you won't know whether it is an actual value disclosed by the company, or a value estimated by that ratings firm.
- 𝗧𝗵𝗲 𝘄𝗼𝗿𝗱 '𝗲𝘀𝘁𝗶𝗺𝗮𝘁𝗲𝗱' 𝘀𝘂𝗴𝗴𝗲𝘀𝘁𝘀 𝗮𝗻 𝗮𝗻𝗮𝗹𝘆𝘀𝘁 𝗺𝗶𝗴𝗵𝘁 𝗵𝗮𝘃𝗲 𝘀𝘁𝘂𝗱𝗶𝗲𝗱 𝘁𝗵𝗮𝘁 𝗰𝗼𝗺𝗽𝗮𝗻𝘆 𝗮𝗻𝗱 𝗶𝘁𝘀 𝗶𝗻𝗱𝘂𝘀𝘁𝗿𝘆 𝗮𝗻𝗱 𝗺𝗮𝗱𝗲 𝗮𝗻 𝗶𝗻𝗳𝗼𝗿𝗺𝗲𝗱 𝗰𝗼𝗺𝗽𝗮𝗻𝘆-𝘀𝗽𝗲𝗰𝗶𝗳𝗶𝗰 𝗲𝘀𝘁𝗶𝗺𝗮𝘁𝗲.
In reality, although their precise methodology is typically opaque, the estimated value is just an average, calculated from that company's regional and sectoral peer group. That's why we call it a 'guesstimate'.
As I warn investors, it's like hiring an analyst after you were reassured that they got 70% in their final mathematics exam. You then learn that actually, they never showed up for that exam and this score was in fact a class average.
You can surely appreciate that if they had sat the exam, their score might have been very different from 70%.
3. 𝗔 𝘀𝗲𝗿𝗶𝗼𝘂𝘀 𝗽𝘂𝘀𝗵𝗯𝗮𝗰𝗸 𝗺𝗶𝗴𝗵𝘁 𝗯𝗲 𝗯𝗲𝗴𝗶𝗻𝗻𝗶𝗻𝗴 𝗳𝗿𝗼𝗺 𝗮𝘀𝘀𝗲𝘁 𝗼𝘄𝗻𝗲𝗿𝘀.
These are the institutions who ultimately own the capital that asset management firms invest (on their behalf).
Many asset owners trust the large legacy brands in ESG ratings; in fact some demand that their asset managers use one of them.
But many never realised how many of the 'reassuring' grades they review each quarter are based on guesstimated data.
Within the UK, some asset owners (public pension fund trustees) are even receiving advice that relying on estimated ESG data might constitute a breach of fiduciary duty.
What do we think?
We only use company-disclosed data behind any fundamental analysis we do - and if the company hasn't disclosed data that a recognised framework like the SASB Standards would expect to be disclosed, we make that clear and mark the company down for it.
Many of you might disagree that this is the best approach. But if you are going to use estimated data - you should at least know it's estimated data.
𝗕𝘂𝘁 𝘄𝗵𝗮𝘁 𝗱𝗼 𝘆𝗼𝘂 𝘁𝗵𝗶𝗻𝗸? When reviewing ESG data and scores for a company, do you see estimated ESG data as necessary gap filling?
We invited readers of this post to give their thoughts on the question below:
How comfortable are you with estimated ESG data?
Want to share your thoughts? Email us via email@example.com or fill in this form HERE to book in a demonstration on our own approach to ESG analysis.
Call us on 020 3478 1144.