Net-Zero Investing: Harnessing the Power of Unstructured Data

Climate considerations are becoming increasingly central to investment strategies as investors seek to manage risks, capitalize on opportunities, and identify engagement targets related to climate change. These strategies often hinge on the availability and quality of climate data, but existing datasets frequently fall short. They tend to focus on historical emissions rather than providing forward-looking insights, creating significant challenges for investors striving to align their portfolios with net-zero goals.

According to the authors of “Net-Zero Investing: Harnessing the Power of Unstructured Data,” achieving such alignment demands forward-looking assessments of company behaviors and outcomes, which current data often fail to adequately support. As the demand for climate-aligned investing grows, addressing these data limitations has become imperative.

The goal of the paper—coauthored by a team from Acadian Asset Management—is to explain how investors may respond to this challenge and to propose a realistic implementation that addresses it. It highlights how climate investors can leverage unstructured data through natural language processing (NLP) and other machine learning (ML) techniques to extract meaningful insights from diverse information sources, how they should incorporate new information that becomes available over time, and how they may deal with the uncertainty inherent in climate alignment estimates.

The challenges of net-zero investing lie in translating broad goals into actionable portfolio strategies, according to the paper. While the concept is simple—creating a portfolio aligned with a decarbonizing global economy—the practical execution is far more complex. It requires investors to evaluate how a company’s behavior today will evolve over decades. Most companies cannot yet claim net-zero alignment, necessitating speculative assessments of future behavior and outcomes. These assessments are complicated by such issues as data inconsistencies, measurement errors, and greenwashing. Despite these problems, emerging technologies, particularly ML, offer promising solutions to the challenges of net-zero investing.

One of the central challenges is the inherent uncertainty of predicting long-term outcomes. This paper advocates for a probabilistic approach that incorporates a range of potential scenarios. Rather than relying solely on single-point estimates, investors should consider a spectrum of possible outcomes to better manage risks and opportunities. This approach aligns with the realities of climate investing, where future developments are highly uncertain.

The paper offers a detailed case study that demonstrates how these methods can be applied to assess a company’s alignment with long-term decarbonization goals, showcasing the potential of combining multiple datasets and advanced statistical techniques to build robust climate measures.

Key Takeaways

• Data challenges: Existing climate datasets fail to provide the forward-looking insights necessary for net-zero alignment, with limited coverage across asset classes and significant inconsistencies in methodology and quality.
• Credibility issues: Corporate decarbonization commitments are often vague or exaggerated, making it difficult for investors to assess their validity. Greenwashing exacerbates this problem.
• Role of machine learning: ML, particularly NLP, is vital for analyzing unstructured data, such as corporate disclosures and external narratives, to gain deeper insights into climate risks and opportunities.
• No single solution: Net-zero strategies require blending diverse datasets and statistical techniques to create comprehensive and credible climate measures.
• Acknowledging uncertainty: Effective climate investing must incorporate uncertainty into decision making, emphasizing a range of possible outcomes rather than relying on overly precise single-point estimates.

The Problem with Current Climate Data

The paper asserts that climate data currently available to investors suffer from two primary issues: a focus on historical emissions and incomplete coverage. While historical data provide a snapshot of past performance, they fail to account for how companies might adapt to future decarbonization demands. For example, metrics such as carbon intensity measure emissions at a specific point in time but are not reliable predictors of a company’s long-term alignment with net-zero goals.

In addition, the lack of comprehensive data coverage creates challenges for asset owners managing diverse portfolios. Data providers often prioritize large-cap issuers in developed markets, leaving significant gaps for small-cap and emerging market issuers. This discrepancy complicates efforts to assess portfolio-wide climate risks and opportunities, as many asset owners aim for systemic climate alignment across all holdings.

Existing climate alignment measures often lack transparency and credibility. Proprietary metrics developed by data providers can vary widely in methodology, leading to inconsistent outcomes. For example, implied temperature scores provided by different providers often exhibit low correlation, making it difficult for investors to determine which source to trust. Moreover, many of these metrics give a false sense of precision, presenting results with decimal-point accuracy despite significant uncertainties. This inconsistency undermines confidence in the data and poses challenges for investors seeking reliable indicators of net-zero alignment.

Greenwashing—the practice of overstating environmental commitments—further complicates the reliability of climate data. Many companies make bold claims about decarbonization, but the credibility of these commitments varies. Corporate disclosures often lack specificity, making it difficult for investors to evaluate the feasibility and sincerity of stated goals. Without robust tools to parse and analyze these narratives, investors risk relying on overly optimistic or misleading data.

The Role of Machine Learning and NLP

Machine learning offers a transformative solution to the challenges of climate investing for two key reasons. First, the data required to assess net-zero alignment are often unstructured—for example, corporate narratives, external commentary, and qualitative disclosures. Unlike numerical data, these forms of information cannot be easily analyzed using traditional statistical methods. ML techniques, particularly NLP, excel at processing and interpreting such data at scale.

Second, ML provides the scalability necessary for analyzing large volumes of complex data. Human analysts can evaluate individual issuers, but this approach is not feasible for portfolios containing thousands of securities. ML enables consistent, repeatable analysis across a wide range of issuers, providing investors with actionable insights.

Unstructured Data as a Critical Resource

The most valuable insights for net-zero investing often come from unstructured data. For example, understanding a company’s decarbonization strategy requires analyzing its stated commitments, intermediate milestones, and planned actions. These details are typically presented in narrative form, making them difficult to quantify or standardize. NLP tools can extract meaningful patterns from such data, providing a more nuanced understanding of a company’s climate alignment.

Integrating ML with Other Techniques

While ML is a powerful tool, it is not a standalone solution. This paper emphasizes the importance of combining ML with other datasets and statistical techniques. For instance, Bayesian updating can incorporate new data into climate measures, providing a dynamic view of a company’s net-zero alignment. By integrating ML with these methods, investors can build more comprehensive and reliable climate measures.

Conclusion

Climate investing—in particular, net-zero strategies—presents a complex but essential challenge for investors. The limitations of existing climate data—such as a focus on historical emissions, inconsistent methodologies, and incomplete coverage—highlight the need for innovative approaches. By leveraging ML and NLP, investors can overcome these challenges, extracting actionable insights from unstructured data and building comprehensive climate measures.

“Net-Zero Investing: Harnessing the Power of Unstructured Data” emphasizes that no single solution can address all the needs of climate investing. Instead, investors must adopt a multifaceted approach that combines diverse datasets, advanced techniques, and probabilistic frameworks. While climate investing remains both an art and a science, the tools and methodologies outlined in this paper provide a solid foundation for building effective and credible net-zero portfolios.

The Paper’s Authors

Andy Moniz, CFA, Senior Vice President, Director of Responsible Investing, Acadian Asset Management, London

Devin Nial, Senior Vice President, Portfolio Manager, ESG, Acadian Asset Management, Boston

Matthew Picone, CFA, Senior Vice President, Portfolio Manager, ESG, Acadian Asset Management, Sydney

Lukasz Pomorski, Senior Vice President, Portfolio Manager, Acadian Asset Management, Boston

Jerry Yu, Assistant Vice President, Associate Portfolio Manager, ESG, Acadian Asset Management, Boston