Pillar 5
Tech and Data for Sustainability
SGFIN leverages AI, machine learning and advanced analytics to strengthen sustainability reporting and impact measurement. It develops models to assess how environmental and social performance influence market valuation, alongside tools that enhance transparency, comparability and data-driven decision-making across firms.
SGFIN research on technology and data for sustainability examines how advanced analytics and digital tools can accelerate sustainable finance and environmental decision-making. It highlights predictive modelling, risk assessment frameworks, integrated data platforms and scalable solutions that enhance transparency, strengthen disclosure, improve capital allocation and support measurable climate and sustainability outcomes across sectors.
Harnessing Data-Driven ESG Partnerships to Scale SME Sustainability in Asia-Pacific Region
Loi, T.S.A., and Zhang, W. (2025). Harnessing Data-Driven ESG Partnerships to Scale SME Sustainability in Asia-Pacific Region. In Best Practice Guide – For How to Modernize Sustainable Small Business Training, Advisory, & Advisory, & Capacity Building Offerings that Work (pp.24-25) [Flipbook]. IFC SME finance Forum. https://online.fliphtml5.com/yywio/gxkk/#p=25
AI & Corporate Sustainability
Ong, K., Mao, R., Satapathy, R., Filho, R. S., Cambria, E., Sulaeman, J., & Mengaldo, G. (2025). Explainable natural language processing for corporate sustainability analysis, Information Fusion, 115(102726). https://doi.org/10.1016/j.inffus.2024.102726
This paper examines how the inherent complexity of sustainability leads to subjectivity in corporate sustainability assessments. It highlights that sustainability disclosures are often incomplete, ambiguous, unreliable, and highly sophisticated, making consistent evaluation challenging. Moreover, interpreting such disclosures is a resource-intensive process prone to human bias. The authors propose that Natural Language Processing (NLP) can automate parts of the sustainability analysis, improving efficiency and reducing some aspects of subjectivity. By further integrating linguistic analysis techniques with Explainable Artificial Intelligence (XAI) capabilities, the study suggests a pathway to enhance transparency, interpretability, and reliability in sustainability assessments.
Harmonized Framework for Corporate Sustainability Evaluation
Asda Pandiangan | Saranraj Rajindran | Zhang Feimo | Johan Sulaeman
This whitepaper outlines SGFIN’s Sustainability Evaluation Framework (SEF) that integrates corporate operations and value chain, strategic planning, and external validation through independent audits and adherence to global reporting standards.
Improving the Integrity of Sustainability Data: Reviewing Environmental Coverage of Sustainability Data Providers
Saranraj Rajindran | Tifanny Hendratama | Johan Sulaeman
This whitepaper aims to highlight the inconsistencies in underlying sustainability data that informs sustainability ratings, quantify the data gaps in Southeast Asia, and provide insights to enhance the usefulness of sustainability data for commercial, financial and research purposes.
ESG Data Primer: Current Usage & Future Applications
Tifanny Hendratama | David C. Broadstock | Johan Sulaeman
This whitepaper aims to clarify the fundamentals, sources, and limitations of ESG data, offering a practical introduction to help investors and researchers better understand its use, variability, and implications for responsible investing.
Institutional Investors & Corporate Environmental Performance
Nofsinger, J. R., Sulaeman, J., & Varma, A. (2019). Institutional investors and corporate social responsibility. Journal of Corporate Finance, 58, 700-725. https://doi.org/10.1016/j.jcorpfin.2019.07.012
Despite the growing interest in sustainable investments, our understanding of how various aspects of CSR affect institutional investors’ portfolios remains limited. This study provides empirical evidence supporting the notion that economic incentives play a key role in shaping institutional investors’ preferences for a company’s Environmental (E) and Social (S) performance. Utilizing institutional investors’ stock holdings data, we observed a trend where institutions tend to reduce their holdings in firms with weaker ES performance, yet they appear largely indifferent to firms with strong ES performance. This suggests that firms with significant ES weaknesses face higher downside risks, including the potential for bankruptcy or delisting from stock exchanges due to poor performance. In the case of firms with strong ES performance, we conclude the investors’ ambivalence arises from a lack of clear economic benefits associated with such performance.
At SGFIN, our research on impact pricing revolves around estimating the effects of environmental features on corporate valuation. When examining the financial implications of a firm’s environmental features, the focus of previous studies is often on the dynamics of specialized trading markets of each specific feature, especially carbon emissions trading markets (carbon market).
Our primary objective is to analyze the stock market pricing of environmental features of (public) companies, including their Greenhouse Gas (GHG) emissions and energy consumptions. Our research also aims to identify which environmental variables have material effects on firm values and how these effects evolve over time across regions and industries. With numerous corporate fundamental features (e.g., profitability, efficiency, and leverage) likely related to market valuation, their impacts on valuation are complex and non-linear. This complexity underscores the inadequacy of conventional approaches, such as linear regression analysis, in linking corporate environmental features to stock market valuation.
We therefore employ a machine-learning-based approach to achieve our objectives. With our machine-learning framework covering 90+ financial variables for firms listed on 100+ financial markets globally, our research contributes to the valuation analysis by allowing non-linearity in linking environmental features with firm values. The valuation models we are developing would be useful for industry practitioners in estimating the impact of environment-related features on corporate valuation and generating more accurate valuation of both public and private firms, which would reflect the sensitivity of their future cash flows to environmental events, such as natural disasters and environmental regulatory and policy changes.
Against the context of needing to achieve a rapid and comprehensive decarbonisation of our economic systems, Environmental, Social, and Governance (ESG) data has emerged as the predominant mechanism for consistently recording, tracking and benchmarking corporate sustainability performance. ESG data is rapidly growing in availability and has become a key tool for assessing sustainability practices of companies, and facilitate socially responsible investors' capital allocation decisions.
Despite the increasingly mainstream use of ESG data, there remain many misconceptions and incomplete understandings, hence the need for the primer contained in this whitepaper. Moreover, under scrutiny we observe (potentially material) ESG data inconsistencies which could result in unintended investment allocations. We probe this issue closely to highlight that not only are there inconsistencies within and between ESG data providers, but also that there are inconsistencies in underlying data used by ESG data providers versus companies self-reported sustainability data.
Informed by our observations of the ESG data construction methodology, and the research use cases it has been used for, we outline an ESG data users’ action plan, distilling the most central points of focus for new and experienced ESG data users to keep in view when working with these data.