Breaking the Perception of ESG Data Estimations
Estimations can help address ESG data challenges to unlock value for the private markets
In recent years ESG has been a fixture in the headlines, with heightened interest from investors and growing concerns about the potential for greenwashing. The need for reliable data has become increasingly paramount. Nevertheless, a notable obstacle persists in the form of limited available ESG data—a challenge that can be effectively addressed through the use of estimates.
Despite the widespread acceptance of estimations in other areas of investment analysis, such as Time Value of Money or Risk vs Return Tradeoffs, there remains a hesitancy when it comes to the use of ESG data estimates.
Understanding the potential value of estimated ESG data requires education on the process of deriving these estimates and how they can illuminate critical sustainability factors. These estimations support a range of use cases including investment management and can initiate meaningful conversations, address major data gaps, and ultimately empower investors to make better-informed decisions.
The Need for Estimations
Private markets are increasingly recognized for their potential role in the journey to address environmental and social challenges. However, they still lag behind public markets, particularly in terms of sustainability data disclosures. And, even when data is reported, it may not always be comprehensive, unbiased, or computed in a way that allows for easy comparisons.
Reliance solely on reported data can introduce biases, due to imbalances between resources. For instance, certain regions such as Asia-Pacific exhibit considerably lower disclosure rates in comparison to regions like Europe. When looking at one of the most commonly reported metrics globally, Scope 1 emissions, the available data reported for this metric in this region stood at 23% in 2021, contrasting with regions like Europe where this figure reached 38%¹. Similarly, during the same year, small-cap companies reported only 24% of their Scope 1 data, while mid-cap companies reported 56%, and large-cap companies led the way with an 82% reporting rate². This lack of comprehensive reporting stems from various challenges, including limited resources, lack of expertise in sustainability topics, and competing priorities.
Estimates therefore become a critical tool to fill these gaps, enabling investors to gain a better understanding of the overall sustainability performance of their investments and, in turn, make informed decisions that promote sustainable practices.
SFDR Case Study – Acknowledging the Need for Estimations
The regulation recognizes the vital role that reliable estimations can play. The EU Sustainable Finance Disclosure Regulation (SFDR) requires financial market participants to make a number of sustainability-related disclosures. For those with more than 500 employees – under Article 4 of SFDR – it is mandatory to report annually at entity level their performance on up to 18 mandatory Principal Adverse Impact metrics (PAIs) and two from a set of 46 further PAIs for all their investments³.
However, for cases where direct reporting from investee companies is not available, SFDR allows the use of “best efforts” “by carrying out additional research, cooperating with third-party data providers or external experts, or making reasonable assumptions.” This approach acknowledges the challenges in calculating complex sustainability metrics while emphasizing the use of the best available resources including estimates in specific situations⁴.
Deriving the Most Reliable Estimates
In the complex landscape of sustainability data, Clarity AI stands out by providing reliable estimates. Our approach combines advanced machine learning (ML) and artificial intelligence (AI) to build robust models, essential for investors seeking comprehensive sustainability insights. These models are the collaborative creation of data science and sustainability experts, crafted using state-of-the-art techniques and a solid infrastructure. This approach ensures not only the generation of sound estimates but also the ability to validate and explain them.
A distinctive feature of our ML models is their ability to understand complex, non-linear relationships between various company characteristics and their environmental performance. These models adeptly analyze how factors like business type, location, financial performance, size, and broader macroeconomic conditions interplay, offering a nuanced understanding of sustainability metrics.
Data quality is essential in our process. The only way to create sound models is to feed them with high-quality information. That is why we implement rigorous methods to collect high-quality, reliable reported data, complemented by advanced ML algorithms like anomaly detection and classification methods to eliminate unreliable samples. This meticulous approach ensures accuracy and scalability in our data processing and allows us to use the right data to train our models so that they can learn meaningful signals.
By merging this carefully curated reported data with our reliable estimates, we can derive industry benchmarks, equipping investors with data they can trust to fill data gaps and model the performance of private companies that do not report.
How ESG Data Estimates Drive Positive Change
Investors embracing estimates in their reporting showcase a forward-looking dedication to fostering sustainable and transparent practices while implementing the tools to enable meaningful conversations and engagement with the companies they invest in. As methodologies and standards evolve, estimations offer a pragmatic solution to bridge disclosure gaps. Ultimately, the use of ESG estimates promotes a culture of continuous improvement and better ESG practices in private markets.
It is in that vein that Clarity AI has partnered with eFront, a part of BlackRock, to integrate its leading estimations and insights, delivered via industry-level benchmarks, directly into investors’ workstreams through the eFront Insight platform. Leveraging these industry benchmarks, eFront’s General Partner (GP) and Limited Partner (LP) clients can generate an initial baseline of the sustainability performance of their investments. These advanced technology-based capabilities allow for GPs and LPs to fulfill key requirements, such as regulatory disclosures, while working towards comprehensive data disclosure.
This baseline also empowers participants to assess their portfolios, prioritize engagement with specific investments, and gain a unified view of their portfolio.
In the ever-evolving landscape of ESG investing, estimations play a pivotal role in unlocking the potential of private markets to drive positive change and promote sustainable practices. Embracing estimations is not just a practical choice; it’s a strategic move towards a more responsible and transparent investment future.
¹This analysis was performed using a dataset of 16,183 companies with available 2021 Scope 1 emissions data. The composition of the sample is: 92% public companies, 3.6% private companies, 3.6% public investment firms and 0.8% other entities.
²Small-cap companies are defined as those with a market capitalization below 2,000 million USD, medium-cap companies have a market capitalization ranging from 2,000 to 10,000 million USD, and large-cap companies have a market capitalization exceeding 10,000 million USD.
³N.b. in their recent Final Report on draft Regulatory Technical Standards, the ESAs are suggesting a number of changes including the introduction of further mandatory PAIs.
⁴See for example a recent ESMA publication on the use of estimates
