In our previous blog posts, we have talked about scepticism around responsible and sustainable investing as well as how it has been impacted by wider trends in the investment world, such as the popularity of passive investing and emergence of ETFs.
We also covered issues such as worries around sacrificing returns, but what happens when we get slightly more granular with ESG factors and the sustainable research process?
‘Big Data’ has undoubtedly shaken up numerous global industries, from logistics to retail to media and many more. Behavioural data analytics are increasingly being used for R&D, sales and marketing, and other business functions and processes. Investment is of course no different, and quant tools have played a vital role in it for many years. It should come as no surprise then, that the whole spectrum of responsible investing (from ESG to impact investing) is also very reliant on data, including company disclosures, voting records. Some say ESG-specific performance metrics are still in their infancy – quite a contrast to the financial statements and annual reports that analysts have incorporated into their research process for years. However, there is a universe of tools out there to help professional investors get their ESG data.
ESG Research and analysis is likely to comprise some form of analysis of regulatory filings, and public data, as well as company reports. In addition to these resources, sustainability research firms and ratings agencies such as Sustainalytics are becoming a major force in the market. MSCI compiles its own ESG ratings, and agencies such as Vigeo Eiris are providing in-depth data to professional investors around the world.
In the Netherlands, institutional asset manager APG, a subsidiary of the country’s National Civil Pension Fund (the pension fund for government and education employees) recently announced the use of artificial intelligence to identify innovative opportunities for investing in companies that contribute to solving climate change, education and healthcare issues among others.
In Japan, Hiro Mizuno, executive managing director of the country’s Government Pension Fund (who is also the world’s largest investor), gave a keynote presentation at a conference emphasising the potential of AI and machine learning in mining large amounts of environmental and similar data to make evidence available to investors in real time.
AI-powered analytics could be particularly useful in impact investing, where measurement of impact and relating outcomes to particular goals is crucial. Measuring third party data (as opposed to self-reported), US-based Distilled Analytics says it has the ability to measure the non-financial risks and impacts of particular investments through AI-driven research in addition to human analysis.
Of course, we have barely begun to explore the full potential of artificial intelligence, and many ESG integration processes use a combination of quantitative and qualitative factors – one frequently used approach is to benchmark on quantitative factors and then overlay qualitative or ‘subjective’ factors – this is often to ensure that the investments reflect the beliefs or values of the investor(s), though this is in no way guaranteed. The parts of the ESG research process that are driven by formulas or ratings, and the extent of subjective judgement applied by analysts will of course vary with every asset manager. While there may be companies that are more or less universally viewed as ‘bad actors’ in either E, S, or G, for some (such as Tesla, for example) their rating is hotly contested. This means that qualitative judgement is welcome – especially if analysing whether a specific investment has had a particular impact or outcome.
For our Responsible Investment Review report, we have interviewed over 60 experts, including around the topic of tools and methodologies fund managers rely on to construct ESG funds, and what investors who are keen on aligning their investment with their values of beliefs might expect. For more on benefits and downsides of different ESG research processes and to find out more about the report, please get in touch with the team by email here.