Chaucer Group is a key client for NTT DATA and at the end of January we were invited to take part in their Sustainability Forum, co-sponsored by Moody’s. The event brought together thought leaders from across the insurance and sustainability industries through a series of excellent panel discussions. There was an obvious interplay of the issues throughout the afternoon so in this article I’ve drawn together some of those threads and look forward to where the discussion will take us next.
Theme 1: Enough data quality to drive meaningful reporting and action
Naturally, the issues around reporting standards were debated a few times. A description of the metrics to be published is in some ways the easy part – the more difficult element is the methodology, rigour and assumptions that sit behind these metrics. Insurers were encouraged to declare emissions data with the same care and cadence as financial data (which is open and trusted).
The reasoning starts from a key objective of financial disclosures, which is to measure business risk. Climate change is a category of business risk, which will have a material financial impact and the title of this article is a quote from one panellist.
Financial accounting standards have of course taken many years to emerge with the rigour we have today. For ESG standards though, we’re pushing quickly and some (perhaps more at the sharp end of the problem) openly admitted the data quality was inadequate. Programmes to fix this take 4-5 years so we need to look ahead to what we’ll need to report some distance into the future. (That was refreshing to hear – some firms I work with struggle to accept that data quality issues often have deep roots.).
In my view the antecedents of data quality issues are due to:
- Use of secondary data – usually the information we’re using has been collected for a different purpose and the quality maintained only enough to meet that primary purpose.
- Inconsistent methodologies (both calculations and what data to include) – one problem that results here is precision masking as accuracy (just because the information has been precisely reported, does not mean it is accurate).
- Some data is simply not collected – or is too hard to access. James Wright, Chief Risk Officer at Chaucer, reported that only 7% of requested client data is returned. As a CSO, responding to these kinds of requests can be exhausting – because each firm asks different questions, or the same questions in a slightly different way. So a 7% response rate doesn’t imply 93% of data is not collected somewhere but the effect is the same.
- Poor quantification – there was some debate about which aspects of ESG are properly measurable. Human rights was given as a tricky example. However, I’m not so sure – in fact even on that subject there are great examples of measurement (e.g. the Human Rights Index). The bigger issue is not whether measures exist but whether the measure is meaningful. I recently came across this problem in the context of Diversity, Equity and Inclusion – one can certainly use metrics such training uptake but that isn’t necessarily a good proxy for culture change.
- Wide confidence intervals – we see this elsewhere in NTT DATA: some suppliers of IT equipment for example supply whole life emissions data with a +/- 80% confidence interval. This has a knock-on impact through the calculation process.
- Analysis paralysis because some uncertainty always exists. I sometimes call this a fractal problem: measuring emissions is not like measuring money flows – behind every measure there’s always a more detailed model that could be created to provide greater fidelity. To have any point, each more detailed method must be supported by more fine-grained data. Finding this data gets harder and harder, so these fractal problems quickly become futile. In other words, you make perfection the enemy of good. So, as was said at the Forum, materiality is critical.
Data quality not only affects declaration today but more importantly might drive poor planning as net zero plans are made with the wrong assumptions. Because we are planning over long time horizons, small mistakes now could become significant in the long term.
Theme 2: The disclosure tightrope
Those of us working in sustainability desire more transparency (there was acknowledgement that transparency builds trust) but at the same time if the pressure for data quality is too high, then we’ll end up with minimum compliance. Since most of the time, ESG data has to be independently audited, the responsibility for accuracy is shared between two organisations. Moreover, we heard that for all the trust gained, not many customers were currently saving money on their insurance as a result of their positive ESG stance.
We saw this tension played out through the afternoon with one panellist urging firms “rather than think: what we’re going to get beaten up for? Think: what do we want to be famous for?”. But what we often see is firms wanting to be in the ‘middle of the pack’ not famous for anything but not being called out for green washing either. This is called ‘greenhushing’ – where firms are not explicit enough about their activities so they don’t stand out. These are all aspects of the politicisation of the ESG agenda which I’ve written about elsewhere.
In NTT DATA UK, we err on the side of transparency. The data is by no means perfect, but the data quality issues we’re aware of we believe to be de minimis. Moreover, we have collected detailed data internally in order to drive our sustainability initiatives so it’s easier to take the step of publication.
Regulation is of course a big part of the answer here – and we heard that many firms are adopting a principle of complying to the highest regulation that affects all their jurisdictions, bringing many more firms into the scope of CSRD (NTT DATA UK is pulled into those regulations for the same reason). The creeping grip of regulation is one reason that the carbon price is expected to escalate in the near term – because buying carbon removals will effectively become mandatory, driving up demand.
Running ahead of regulation, I see a lot more pressure in the supply chain – or from industry associations. I had previously looked favourably on industry associations as a means to drive mutual improvement ahead of regulation (for example NTT DATA Inc is looking to convene industry players in the IT hardware sector). However, this must be done with care as it has backfired in the past when the introduction of sustainability standards amongst a subset of players was viewed as anti-competitive.
Theme 3: Context is crucial
A confounding problem is that a C02 footprint and other measures also need context – what kind of industry is the client in? Where are they based? Etc. The importance of context makes it harder to outright compare one firm with another. My last employer was naturally exposed to more business travel emissions, with many IT consultants based in Northern Ireland but a lot of work in Great Britain. However, their measurement and decarbonisation programmes are ahead of others in the market.
Lack of context is one problem with sustainability ratings. And because of the aforementioned data quality issues, ESG scores are not necessarily built on solid foundations. They also look at different lenses - Moody’s for example try to measure ESG readiness. If the ESG scores from several providers do not agree, it is not always clear why. There was a healthy debate about whether insurers should use these ESG scores assigned by external agencies (which might use gross assumptions like the industry a firm is in) – or work from their own data. The latter provides competitive advantage because customers that have been unfairly penalised by industry benchmarks can get a better deal when ESG ratings begin to materially affect premiums.
Theme 4: Offsets and removals
The final discussion - on the nature of voluntary carbon markets - underlined just how complex the ESG subject area can become. As someone who buys carbon offsets, I hadn’t really considered that they might be reversed at some point (e.g. by a natural disaster) and what liabilities that might impose. Our removals focus on peatland restoration but even then, working out who’s assurance we take for additionality, permanence, leakage, and co-benefits is not cut and dried (excuse the peat-based pun). The market was rightly described as immature, but if the carbon price predictions are to be believed, that is going to change in the near term.
Conclusion
What I appreciated about the Forum was the focus on the practical issues and pitfalls - and lack of hyperbole.
Chaucer Group and Moodys found some genuine though leaders and as a technology person the event provided vital insight into the challenges that lie ahead for those trying to make a difference – and the insurance industry is in a position to do just that.