Analysing ESG Materiality using theoretical and empirical data sources
July 18, 2023
Materiality was again in the spotlight recently with launch of IFRS sustainability disclosure standards S1 & S2. Whilst the standards were generally welcomed as a significant step in creating a global baseline for corporate sustainability disclosures grounded in the best practices of accounting standards it was seen as controversial in some quarters. In a large part this was due to its approach to dealing with materiality and more specifically ‘double materiality’ — something largely avoided in the standards according to critics. Proponents highlight the mantra that “what can be measured can be managed” when suggesting that focusing on single materiality / materiality should be the priority for disclosing organisations and is thus the focus of these standards. Double materiality encompasses not just sustainability risks which impact a company’s financial performance but the degree to which that company’s products and activities impact the wider world — a philosophy more adopted in European standards such as in the work of EFRAG.
How do investors identify and measure material issues however? What data is required and how easy is it to interpret this data?
First of all its important to note there is no single approach to materality, although there are some common features to different standards. According to https://www.materialitytracker.net/ “a typical financial threshold is 5–10 percent of earnings… between 5 and 10 percent requires judgement.”. Probability and severity feature strongly in materiality approaches as do baseline guidelines for particular industry sectors such as the SASB Materiality Map. Many standards contain a notion of materiality including SASB, SFDR, Climate Bonds and others which investors can use as a starting point for analyses. However this is largely theoretical work and challenging to apply in depth to individual companies materiality profiles at scale — across say a portfolio or in constructing a universe.
Some of the limitations of materiality guidelines are linked to the uniform nature of the guidance. Companies are free agents and can sometimes behave in unpredictable ways (e.g. Twitter) which can have a material impact on share prices difficult to account for in sector level guidance. Similarly data is often not granular enough to be sufficient for materiality guide rails — for instance industry sector standards (e.g. TRBC, NAICS, GICS, RBICS etc) often group companies into broad categories which make comparability of companies impossible from a materiality perspective (e.g. Food & Tobacco). SASB in particular uses its own industry sector standard the Sustainable Industry Classification System (SICS) which is linked via its materiality map to a hierarchy of material sustainability issues. Finally whether an issue is material or not from an investor perspective is subject to differing research and opinions — its subjective and ultimately proven or disproven by a share price change or event (e.g. cancelling a dividend).
Fortunately computational techniques continue to emerge which can help investors empirically understand material issues within their investment universe or portfolio. Modern ESG platforms are beginning to extend traditional industry sector data with more meaningful contextual data such as products, processes and activities and their respective ESG risks. Knowledge Graphs, such as our www.responsiblecapital.io ESG platform enable this top down perspective. Combined with the power of Large Language Models (LLMs) we are now also able to summarise huge volumes of research, news and opinion into Knowledge Graph concepts and connections useful for financial analysis. Features of other standards such as SFDR PAIs and the EU Taxonomy technical screening criteria can also be useful here.
Existing semi & unstructured disclosures such as SEC financial disclosures play a significant role in helping both equity and fixed income investors to identify material issues for portfolio companies. AI and Natural Language Processing (NLP) techniques can be used very effectively to identify and summarise these for particular companies — for instance by extracting financial provisions for specific litigation risks from 10Ks (e.g. Forever Chemicals provisions) or processing at scale Risk Factors disclosures in bond prospectuses (S1 etc.).
News, thankfully, is much more widely read than financial disclosures and thus more a significant data source in analysing materiality. It is largely untapped by ESG investors given challenges around volume, noise, multiple languages and the varied credibility of sources. Fortunately we have developed a data product to enable investors to incorporate news flow analysis from over 1M sources into ESG Integration at a sector, thematic and corporate level — Responsible Capital News Screen. By extracting structured data from news such as ESG concepts, companies, locations and sentiment investors can help cut through the noise to identify material issues, measure company exposure and the degree to which exposure is changing over time. We can perform analyses such as ranking the top 10 companies globally linked to a particular issue such as water pollution and also which locations the news is linked to in order to identify litigation risks from varying strength of legislation across different legal jurisdictions for water regulations for example.
Ultimately taking a comprehensive approach to analyse materiality relies on combining both theoretical data from standards bodies such as in our attached work combining SASB’s materiality map and empirical data such as news. Our Responsible Capital product enables users to do just that and seamlessly analyse the two facets side by side. The below images show two analyses created using a year’s worth of ESG news — the first predominantly blue matrix shows the volume of news linked to SASB’s Ecological Impacts risk category for each top level SICS industry group. As you can see whilst there are some correlations where SASB identify issues that also commonly feature in news stories such as Ecological impacts associated with Extractives & Mineral Processing there are some surprising areas where news volumes exist for materiality matrix combinations not explicitly referenced by SASB such as Ecological Impacts and Finance or Consumer Goods. Vice versa there are also SASB referenced combinations for which there is little material event related news flow such as Land Use And Ecological Impacts linked to the Food & Beverage sector.
It is even more revealing when sentiment is overlayed instead of news volumes. For example here interestingly we see that Environmental Impacts of Project Development are very negatively referenced in news flow linked to the Transportation sector yet are not acknowledged as a SASB material issue. Similarly Ecological Impacts linked to Finance and Infrastructure are both very negatively talked about yet not identified by SASB as material. Interestingly some SASB identified intersections such as Food & Beverage + Land Use And Ecological Impacts and Renewable Resources & Alternative energy + Ecological Impacts of Project Development are talked about positively in the news. This is likely because although these material risks exist for these sectors press coverage typically focuses more on the positive contributions of these sectors to these issues rather than the negative impacts through for instance rare earth supply chains.
Of course analysing news for material exposures at sector level can only get you so far and most investors will be more interested in individual company exposures to material issues. All of the analysis and tools mentioned above we can apply at company level to identify material issues to aid investment research, risk management and the development of targeted engagement strategies.
We hope you’ve found this article useful, thought provoking and interesting! Do drop us a comment or note if you are interested in discussing further.