- Junta Nakai is the global industry leader for financial services at
Data bricks, a big data and AI company valued at $6.2 billion. - In this op-ed, Nakai explains why companies looking to focus on environmental, social, and governance (
ESG ) goals need to use data and AI technology.
Companies and brands that want to cement themselves as environmental, social, and governance (ESG) leaders must leverage data and artificial intelligence.
Today, the broad benefits of incorporating ESG goals are well understood by companies, investors, and regulators. That's where the consensus ends. To put it another way, the "why" of ESG is clear, but the "how" and "what" are not. As a result, there is little agreement on what data companies should collect, and how they should disclose/analyze it.
While industry bodies are working to create international standards/guidance for ESG disclosure, it is not enough. For ESG to really take off and make global impact, data and AI must play a central role in collecting, verifying, and analyzing ESG performance. This can only be done effectively today by leveraging technology.
In the absence of standards, the onus falls on individual companies and investors to ensure high-fidelity ESG disclosures as well as to verify the sustainability of vendors, suppliers, customers, and counterparties. For example, seven in 10 investors cite lack of high-quality information as the biggest challenge in adopting ESG principles.
In other words, ESG's main limitation today is data and the tools required to analyze it, not the lack of regulatory guidance.
'Greenwashed' ESG
Today, companies typically issue ESG reports once a year, which are composed of a mixture of metrics and marketing.
For example, a recent 70-page sustainability report from a leading investment bank had just three pages dedicated to hard metrics. The remaining 67 pages undoubtedly contain valuable information and signals. But how does an investor or client of this bank quantify the text to objectively quantify ESG performance? Do these disclosures reflect the reality of ESG performance or are they simply the easiest observable data points?
Without an ability to quantify and independently assess ESG data, ESG has increasingly become a marketing term without solid grounding in objective analysis.
This phenomenon is well documented in what is called "greenwashing", a term designated to institutions that mislead consumers/investors about the performance of their ESG goals. When there is little distinction between stated ESG goals and supporting facts, the ESG movement risks losing its meaning and impact.
How Data and AI fill gaps in ESG
Unlike financial data, ESG disclosure currently does not have generally-accepted principles. As a result, entrepreneurs and coalitions have stepped in to fill the gap.
Today, there are over 100 providers of ESG data that serve corporations and investors. In addition, a patchwork of consortiums aim to provide guidance on specific issues within ESG. For example, Network for Greening the Financial System focuses on the "E" while a collection of European regulators are working on comprehensive ESG disclosure standards.
While these developments are commendable, ESG standards will take years to coalesce.
However, today's companies and investors do not have the luxury to "wait and see." Consumers already expect sustainability from companies they purchase from as do influential investors.
To cement themselves as ESG leaders, companies and brands must first address foundational problems of information quality (is the data accurate?), compliance (will companies adhere?), and verification (is it true?). All of this begins with data.
Read more: Investors are clamoring for pandemic bonds. Here's how Wall Street banks are responding.
1) Data Quality
Today, finding ESG data even internally is a highly manual data-collection process. Leading companies disclose a range of ESG-related data from water consumption, carbon emissions to workforce demographics. Each datapoint is likely kept in separate databases in different formats and schemas, making it difficult to ensure the data is high-quality or accurate. Centralizing the data in a single data lake can alleviate data access and quality.
2) Compliance
Once information is centralized in a modern cloud-based storage architecture, companies can get a real-time view and understanding of their own ESG performance. This enables self-correction and benchmarking, thus improving compliance with their stated goals. In addition, having the data accessible means ESG metrics can be disclosed more frequently.
3) Verification
For most large companies today, ESG verification simply means asking partners to abide by the vendor code of conduct. But how do you verify it? AI can play a central role in the verification process by using techniques from natural language processing (programmatically extracting information from text) to graph analytics (learning how different entities influence each other's ESG).
Without solving the foundational data problem, companies will not have an accurate understanding of their own ESG metrics (garbage in, garbage out). This is why ESG could be best viewed in the prism of technology, not policy.
Example of how data + AI can be leveraged
Modern CEOs must not only be aware of their ESG own scores, but also "how" they are being perceived (by the market, consumers, investors) and analyzing "what" the sustainability metrics of counterparties, suppliers, and partners are.
For illustration, consider an IT executive who wants to assess and verify the ESG performance of a manufacturer that makes its widgets. Today, you have to rely on self-disclosure and/or third-party vendors of ESG data. Your options are limited to either taking self-disclosure at face value and/or cross-checking the disclosure with ratings from an ESG data vendor.
Tomorrow, ESG leaders will leverage all types of data (structured, unstructured, alternative) and AI.
First, you will leverage alternative data to get insights on the environmental impact of a particular widget factory to analyze air-pollution levels, water quality in nearby lakes, and health records of nearby populations.
Next, you will use AI techniques such as embedding that will help you see how strongly environmental concepts are associated with the manufacturer in the media, enabling you to see if there is a gap between what is said and what is done.
Finally, you will run Monte Carlo simulations at scale to see how different climate economic conditions might impact the manufacturers ability to adhere to ESG standards.
Every step here can be done today using existing technologies and cannot be effectively addressed by alone by providing ESG guidance or standards to companies.
The answer is technology with policy and vendors
As illustrated above, data and AI can enforce ESG factors in ways that rule-making or data vendors alone cannot. Automating the data-collection process for ESG as well as leveraging AI to analyze ESG enables:
1) More frequent reporting
2) Verification of ESG disclosures
3) Holding companies accountable
Ultimately, focusing on the "what" and "how" with technology will enable ESG to become intractably tied to business performance and integrated into core business strategy and governance processes. This is what empowers ESG to make a real difference in the world.
Junta Nakai is the global industry leader for financial services at Databricks. In his capacity, he is responsible for driving the world wide adoption of the Unified Data Analytics Platform across capital markets, banking/payments, insurers and data providers. Prior to joining Databricks, Junta spent 14 years at Goldman Sachs, where he most recently served as the head of Asia-Pacific sales for the Americas in the equities division.
Read more: