Building a reporting layer around licensed ESG data for pension funds
How pension funds and pools can build a governed reporting layer around licensed ESG and climate data, with coverage, methodology and stewardship made clear.
Onarvo Team
Local Government Pension Scheme funds and the pools they invest through increasingly rely on environmental, social and governance data. Most of it comes from specialist providers under licence: climate metrics, carbon footprints, controversy flags, ESG scores and alignment assessments.
Licensing the data is the easy part. Turning it into consistent, explainable reporting for committees, members and regulators is harder. This article sets out how we think about building a reporting layer around licensed ESG data.
One point first. A reporting layer does not produce ESG ratings or climate assessments. Those come from the providers a fund chooses to licence. The reporting layer's job is to bring that data together with holdings, apply it consistently and make clear what it does and does not show.
The problem: figures that do not agree
Ask several people in a fund for its portfolio carbon footprint and you may get several answers. That is rarely because anyone has made a mistake. More often:
- Different holdings snapshots were used
- Different providers, or different datasets from the same provider, were applied
- Securities without data were treated differently, excluded in one case and estimated in another
- The provider changed its methodology between periods
- Pooled fund holdings were looked through in one report and not in another
Each of these choices may be reasonable. The problem is that they are made silently, in spreadsheets, and are not visible to the people reading the report.
Start with a governed data model
The foundation is a data model that holds holdings, securities, issuers, provider data and reporting outputs in a structured, documented way. Governance means each part has an owner, a definition and a change process.
Key design choices include:
- Holdings captured as dated snapshots, so any report can be reproduced for its period
- Provider data stored with its as-of date and the licence it came under
- Calculations defined once and reused, rather than rebuilt in each report
- A clear record of which data and rules produced each published figure
This is not a large platform for its own sake. It is the minimum structure needed for two people to get the same answer from the same question.
Match identifiers carefully
Holdings data from custodians and managers uses one set of identifiers. Provider data may use another. Securities map to issuers, and issuers sometimes map to parent entities in different ways depending on the provider.
Identifier matching is often where coverage is quietly lost. We build matching as an explicit, logged step: which identifiers were tried, which matched, which needed a manual decision and which remain unmatched. Manual mappings are reviewed and kept, so they do not have to be redone every quarter.
Show coverage with every metric
A metric without its coverage can mislead. A carbon intensity figure based on part of a portfolio says something different from one based on almost all of it.
We recommend that every reported metric carries its coverage alongside it, ideally both by number of holdings and by value. Where estimated data is used, the share of estimated versus reported values should also be visible. Readers can then judge how much weight to place on each figure.
Version methodologies
Providers update their methodologies, and funds change their own reporting approaches. Both can move figures without any change in the underlying portfolio.
A reporting layer should record which methodology version applied to each dataset and each output. When a methodology changes, the fund can then explain movements, restate prior periods where appropriate and avoid comparing figures that are not comparable.
Keep stewardship records alongside the data
Engagement and voting are central to how many funds approach responsible investment. Stewardship records, including engagements, objectives, escalations and outcomes, are often held separately from the quantitative data, sometimes by managers and pools rather than the fund itself.
Bringing summaries of these records into the same reporting layer, linked to the relevant issuers, lets a fund show not only what its exposure is but what it is doing about it.
Design to be provider-neutral
Funds and pools change providers, add new datasets or use different providers for different purposes. A reporting layer tied tightly to one provider's formats makes every change expensive.
Provider-neutral design means a common internal model, with each provider's data mapped into it through its own adapter. Licence terms still need to be respected, including any restrictions on redistribution or on how data can be shown in public reports, so the model should record those terms against each dataset.
Reporting aligned to TCFD
Many funds structure their climate reporting around the recommendations of the Task Force on Climate-related Financial Disclosures, covering governance, strategy, risk management, and metrics and targets. Regulatory expectations for LGPS climate reporting continue to develop, and funds should take their own advice on what applies to them.
From a reporting layer perspective, the important thing is that metrics and targets sections draw from the same governed data as everything else, with coverage and methodology notes attached. That makes year-on-year reporting more consistent and easier to assure.
How we can help
Our data and AI team builds data models, matching processes and reporting around the data funds already licence. We work alongside fund officers, pools and providers, not in place of them.
Talk to us
To find out more, see our ESG and sustainability data reporting solution or contact us for an initial conversation.