Impact Management | May 15, 2023

Leveraging ESG data for impact investing

Panagiota Balfousia
Guest Author

Panagiota Balfousia

Management guru Peter Drucker is credited with the popular maxim “what gets measured gets managed.” This holds true also in impact investing where measurement of environmental and social outcomes drives portfolio positioning and active stewardship, with the aim of achieving targeted impacts. Measurement presupposes the availability of environmental, social and governance data. 

The good news is that there has been an explosion of non-financial data in recent years, driven by digitization, regulatory requirements and stakeholder engagement. Yet there is room for improvement, in terms of coverage, relevance, quality and assurance. The effort to ensure a degree of interoperability across jurisdictions and reduce duplication of efforts, is welcome, if not highly overdue. 

Input-oriented data

From a practitioner perspective it is helpful to differentiate between types of ESG data. Input-oriented data have to do with the firm’s ongoing operation and form the basis for assessing the sensitivity of a corporates’ financial performance in the face of environmental, social and governance risks and exigencies. 

Data providers collect a growing number of such input-oriented variables, sometimes many thousands in the case of listed equities and an increasing number in the case of private companies. To mention some examples: gigajoules energy used, cubic meters of water withdrawn, tons of hazardous waste produced, % of board committee independence, and the presence of a raft of policies such as regarding labor conditions and diversity. 

Outcome-oriented data

Outcome-oriented ESG data indicate the effect of a firm’s operations and of the outcomes of those operations on the environment and society. Examples include accumulated GHG emissions in the atmosphere, depletion of freshwater resources, environmental pollution, checks and balances in governance structures, the occurrence of employee injuries and fatalities, and the prevalence of bias in the workplace. Non-financial outcome-oriented data also relate to the products and services, such as number and features of healthcare products, related product safety attributes and characteristics and location of customers served.

The shift of focus from firm operations to the social and environmental outcomes of those operations, adds complexity. Outcome-oriented data are not as widely available or standardized. 

What gets measured gets managed… or only reported?

Non-financial input-oriented data (referring to firm operations) and outcome-oriented data (referring to effects on the environment and society) are very important for putting together ESG reports. However, for “what gets measured” to also “get managed” and not just “get reported,” ESG data must be brought in an appropriate context. 

Goals and thresholds that provide a framework for using ESG data to assess the impact of investment opportunities, include the achievement of climate neutrality, water resource sufficiency, a level of biodiversity supportive of well-functioning ecosystems, minimization of corruption, employee health and safety, and social spaces free of bias, such as based on gender or ethnicity. 

To give a concrete example, the effect of the cubic meters of water withdrawn by a corporation with respect to the availability of freshwater resources, may be assessed on the basis of data related to water sufficiency in local communities. Another example is whether a healthcare product addresses a medical need that was previously unmet, or with a treatment that is substantially superior from those previously available, or at a substantially more accessible price.

Data and analytics providers

Alongside the increase in ESG data, there has been a proliferation of data and analytics providers. As well as collecting data, these service providers clean and standardize the data, use estimation methodologies to extend coverage where data is not reported – increasingly with the use of artificial intelligence – and analyze the raw data by constructing ratings, scores or signals, with the aim of supporting users in the assessment and evaluation of the data. 

The proliferation of ratings with an emphasis on outcomes has been a notable trend. These outcomes are often proxied through revenue streams, mapped onto sustainability themes, typically drawing from the United Nations Sustainable Development Goals. 

There is a large discrepancy in the granularity of the revenue disaggregation, ranging from broad industry classifications to granular activity and product-based classifications. Some providers have or are in the process of developing taxonomies based on variables that are more forward-looking than historical revenues (such as capital or operational expenditure). 

Others map both financial and non-financial data to impact themes, by scoring or assigning monetary value to imputed social and environmental benefits.

From sustainability data to sustainable investments

Increasingly, providers of ESG data are taking a stance in evaluating whether an issuer can qualify as a ‘ESG investment.’ This can take several forms with notable variation across service providers. Typically, a threshold is set on the basis of the performance against a sustainability theme based on a proprietary and/or regulatory taxonomy. The performance may be measured in terms of revenue alignment or percentile performance versus peers or on the basis of meeting a condition, such as having an approved Science-Based Target. 

Perhaps the most important difference is with respect to the transparency and traceability that data and analytics providers offer users into their datasets and calculated ratings and scores. This transparency – perhaps above all else – is important for investors to benefit from the increased data availability and analytical offering in the space.


Panagiota Balfousia is head of sustainable business strategy at Kieger.