Any conversation around how infrastructure investors approach ESG almost inevitably comes back to one word: data. The preoccupation with data stems from the widely held mantra that what gets measured, gets managed.
Investors that put their money into infrastructure funds that pledge to produce sustainability benefits rightly expect measurable results. Regulators, meanwhile, are intent on directing capital towards sustainable assets and insist on data to measure progress towards a range of policy goals.
The result is a plethora of regulations that require data to be disclosed. Many asset owners also require managers to disclose data against voluntary frameworks. And managers are likely to have to supply data in a multitude of formats to LPs that have their own templates for collecting sustainability data.
As a result, managers have their work cut out to cope with the demands to supply performance data. Can they find time to even think about using data – including solutions driven by artificial intelligence – to start actually improving sustainability performance?
It is nearly impossible to find anyone who is happy with the current range of bespoke approaches around compiling ESG data.
The data conundrum
“Infrastructure fund managers face significant challenges meeting demands for ESG data from investors, regulators and stakeholders,” says Eleanor Fraser-Smith, head of sustainability at energy transition-focused manager Victory Hill Capital Partners. “Standardising this data is challenging due to varying metrics, reporting frameworks and industry-specific nuances.”
“The volume of requests we receive increases every day; it is very time consuming,” adds Lucinda Callow, head of ESG for infrastructure at German asset manager DWS. She says that the introduction of the EU’s SFDR has resulted in her firm receiving much more detailed ESG due diligence questionnaires from LPs.
“You get these requests from investors, with tight deadlines in general, and they have to take priority,” Callow says. “I spend most of my time on requests like that at the moment and it would be much better if I could spend my time actually working with our portfolio companies to improve their ESG to make ESG improvements directly.”
A particular problem, Callow laments, is that LPs often ask for data on similar KPIs that are measured in slightly different ways. “We try and avoid going repeatedly back to our portfolio companies to ask them to measure something slightly differently because it is not a good use of their time.”
Efforts are underway to improve consistency in the sustainability metrics that infrastructure managers and their portfolio companies are required to report. But views are mixed on whether the SFDR is a help or a hindrance in this regard.
Lars Meisinger, head of international client advisory at Hamburg-based Aquila Capital, points out that a fund classified as Article 9 under the SFDR that holds 10 clean energy assets needs to report on 350 ESG data points each quarter. This means, he says, that “all those governed by this regulation need to standardise their ESG data structures and processes”.
However, Meisinger adds that “the degree to which standardisation is possible across the industry is still uncertain”, given that the SFDR gives flexibility in some areas, for example by allowing industry players to implement bespoke definitions of sustainable investments.
Meisinger also notes that the SFDR is set to undergo revisions next year, as the European Commission responds to feedback from stakeholders. “It is still too early to see whether industry-wide standardisation is plausible,” he says.
Meanwhile, other jurisdictions, including the UK, are also in the process of introducing their own version of the SFDR. “The map of expectation seems to be constantly changing,” says Fraser-Smith.
Callow agrees that a revised SFDR could “complicate things a bit more”. But she is more positive about the potential of the SFDR to promote standardisation, including through its requirements for financial institutions to disclose the negative effects – known as principal adverse impacts (PAIs) – that an investment has on sustainability factors.
“Everyone is starting to ask for the principal adverse impacts,” Callow says, noting that these PAIs “cover most of the key data points”. Requirements to disclose PAIs were tightened under rules that came into force in January 2023.
Beyond the SFDR, Callow notes that GRESB’s assessments for infrastructure funds are a helpful way of being able to share sustainability performance data with investors in a standardised way. The ESG Data Convergence Initiative, through which a group of GPs and LPs across private markets have been attempting to standardise ESG reporting, is also “definitely helping”, she says. “Quite a lot of our LPs send us requests based on that template.”
Can AI help?
The past year has seen a frenzy of excitement, mingled with trepidation, around the potential of AI to transform many elements of the global economy. Infrastructure managers have increasingly embraced the technology to help achieve sustainability benefits, notably around predictive maintenance on assets such as wind turbines to help extend their lifespan. And there is certainly some scope for AI-based solutions to help fund managers process sustainability data more efficiently.
Yet many observers are cautious about assuming that AI can ride to the rescue. “There is some talk about AI, but it feels premature,” says Rory Sullivan, CEO of advisory firm Chronos Sustainability.
He cautions that AI tools are only as good as the underlying data. “AI solutions grounded on poor or inadequate data will simply reinforce the ‘garbage in, garbage out’ problem in performance tracking and reporting.”
Meisinger believes that AI can play a useful role, however, but notes that AI tools themselves will only work well with standardised metrics for sustainability.
“There are still some areas where AI can make a major difference in the speed and consistency of how we analyse and evaluate ESG data,” he says. “However, this is only possible if there is a certain degree of standardisation – a level playing field as to what, where, and how ESG data is reported.”
If you do not receive this within five minutes, please try to sign in again. If the problem persists, please email: