Accounting for biodiversity in investment portfolios

Channels: Regulation

France has introduced new requirements for companies to consider their impact on biodiversity, as well as climate change. Michele Lacroix, head of group investment risk and sustainability at SCOR, discusses the reinsurer's exploratory work in that area. Interview by Vincent Huck

Why was article 173 replaced by article 29 mandating biodiversity disclosures?

Article 173, a French law, has been instrumental in driving the European Commission's work, and there is a clear willingness on the part of France to stay ahead of the curve.
So, we now have a new Law for Energy and Climate, released in November 2019. Article 29 of this law was enacted with a decree in May this year. The big difference with article 173 is that it requires more detailed disclosures, even though the 'comply or explain' approach remains, and also that it puts climate change and biodiversity on the same level.

And I believe SCOR has already done some work on biodiversity?

Yes, we're completely aligned with the spirit of the new law, but there is still a lot to do. We signed the Finance for Biodiversity Pledge in December, and we joined the Finance for Biodiversity Foundation this year. SCOR also hosted a webinar to celebrate a report published by the French Natural History Museum, financed by the SCOR Foundation for

Science, which analysed the risks that biodiversity loss poses to insurance and reinsurance.Michele Lacroix

If you look at our 2020 Sustainable Investment Report relating to the 2020 financial year, on the asset side we have already performed an analysis on deforestation risk. We have partnered with the French consulting firm Iceberg Data Lab to produce a preliminary biodiversity footprint.

All of this is highly experimental.

It's far too early to be able to take decisions based on this, but we wanted to make sure we were as far ahead of the curve as possible, pushing the envelope. We are a small investor with €20bn ($24bn) of assets, but we are trying to show that it's possible to do something, even if it's complex and there are a lot of uncertainties, even if the methodologies involved are not aligned and data is not available, and so on. It's time to start trying to figure out what data we can get, and how we can use it.

It doesn't mean that the work on climate change is done and dusted though?

No, far from it. We are trying to gain a greater understanding of climate scenarios and how we can better use them in our investment strategy. A particular area of focus is how we can deal with uncertainties over the long term. Dealing with transition risk over a five-year time horizon is not easy, but it is doable. It's much more difficult to project our portfolios over 20 years, for instance, which is increasingly what regulators are asking us to do.

When we are asked to assess the climate risk of our portfolio in 20 to 30 years' time, the first question will be: what will the portfolio look like then?

It's already difficult to know which climate scenarios to use today, how to translate those results into macroeconomic variables and how to apply them to the portfolio. But if you have to deal with uncertainties in terms of time horizons, scenarios, macroeconomic impact, and portfolio exposures, it can start to get very tough – not to run scenarios, you can always put in numbers and get outputs, but tough to determine the value of the output.

At the end of the day, we need to take decisions on the asset allocation because it must be more than just a theoretical exercise showing that we are willing to do something. We need to take action now to fight climate change. From a qualitative stance, SCOR has fine-tuned its sustainable investment policy and we have increased our engagement with investees on climate and biodiversity.

Going back to biodiversity, you mentioned experimenting with a first biodiversity footprint, which is a really hard concept to picture. By comparison, climate change seems straightforward to conceptualise – even if not easy in practice - there are carbon emissions that can be measured. What is a 'biodiversity footprint'?

You are right, a big difference between the climate change journey and the biodiversity journey is that the UN Global Compact existed before climate change became a hot topic, so we had the carbon footprint metric to start with, even if it wasn't 100% satisfactory in terms of reliability, scope or comparability.

But it's worth noting that the story has shifted slightly from 'pure carbon footprint' to temperature rise – i.e. how your portfolio aligns with a global warming scenario.

For biodiversity loss, we are starting from nothing: biodiversity is not yet measured, valued or priced. We don't have a common understanding of what it means, we don't have any data on what it should be. However, some methodologies are progressively being developed through various competing initiatives.

I mentioned our partner, Iceberg Data Lab, which works on the Corporate Biodiversity Footprint based on the mean species abundance (MSA) indicator. This calculates local terrestrial biodiversity intactness as a function of six human pressures: land use, road disturbance, fragmentation, hunting, atmospheric nitrogen deposition and climate change.

In layman's terms, it essentially looks at how many square kilometres have been destroyed by your investment portfolio.

I understand the logic, but in practice it seems impossible... How does one account for all biodiversity within a square kilometre and then assess what will be lost?

In a nutshell, you start with a mean species abundance of 100%, which is the starting point of the assessment for a given surface. Then, if the area you are looking at is fully artificialized, your MSA drops to 0%. So, using the Iceberg Data Lab tool, by combining all six pressure points and using the life cycle assessment you can measure how the assessable part of your portfolio contributes to biodiversity loss - how many square kilometres you have (theoretically) fully artificialized.

That spurs two questions: one is that there is one step missing in the context of double materiality – included in article 29 – as you have now assessed your impact but you also need to loop back on how this biodiversity loss affects your portfolio's resilience. You are left with two challenges: how to assess the risks of biodiversity loss on the value of your assets? And once this is done, how to steer the portfolio to minimize the risks and reduce adverse impacts?

Two very good points, you're right - the Corporate Biodiversity Footprint addresses the impact materiality assessment, and not the resilience of the portfolio to biodiversity loss. Biodiversity is much broader than climate change, which was already a challenge. Biodiversity, and nature as a whole, are really much more challenging. We need to accept that we will have to take action before we get the full set of metrics and the robust methodologies that we need to be able to measure and manage.

We need to learn to work with qualitative information in some cases, and for biodiversity this is currently at the exploratory stage. Although we don't have universal metrics, some emerging methodologies look promising. It is too early to know which one will take the lead and become the standard, but the Corporate Biodiversity Footprint is one possible example of something we could use to try to measure impact.

As for resilience, it's a bit like what we did for climate change and transition risk. If you remember, at the very beginning of our work on climate change, the idea was to identify which sectors, and which sub-sectors, would be most impacted if policymakers were to decide to quickly shift to a low-carbon economy. Through a heatmap of our portfolio, we measured the portfolio's exposure to what could be the most impacted sectors, and assessed whether that impact would be significant.

We're doing the same with biodiversity, but once again the topic is so multi-dimensional that we need to narrow it down and select the areas we are most concerned about. That's where the ENCORE tool (Exploring Natural Capital Opportunities, Risks and Exposure) can provide valuable information.

We began looking at nature-related topics through deforestation on our investment portfolio. This is obviously strongly connected to climate change, as carbon sinks are instrumental in the journey to net-zero. Deforestation is one of the main focus points for many countries, and we try to assess the part of our portfolio which may be at stake if the environment comes under additional pressure from deforestation. And from there, we see which sectors are or will be most impacted, like agriculture for example.

But I just want to highlight the fact that impact - the effect the business has on the outside world - and resilience - the consequences of outside pressures on the business - are intertwined. The impact that you have as a company today will loop back and affect your resilience tomorrow. Thinking about climate change, the speed at which we finance the transition to a low-carbon economy will have an impact on global warming, and indirectly on the future of underwriting.

You said that you looked at 'the assessable' part of the portfolio, what did you mean by that?

Today, one of the main challenges to portfolio assessment is data collection, which links back to company disclosures. For instance, Iceberg Data Lab has been able to cover only 12% of SCOR's corporate bond portfolio, starting of course with companies operating in the sectors most likely to have adverse impacts. When we say we have covered 12% of our corporate bond portfolio, it's actually likely that we have covered much more of the real footprint of this portfolio. Materiality is as important as coverage ratio when it comes to providing relevant information.

We need to bear in mind that nature-related services are neither assessed, valued nor priced. This is a major hurdle for the quantitative assessment of both dependencies and impacts.

What was the outcome of the assessment on the 12% of the corporate bond portfolio?

As mentioned in the 2020 Sustainable Investment Report, the outcome of the Iceberg Data Lab computation was 1300km² MSA (mean species abundance) per year. This means that the impact of the corporate bonds analysed amounts to the equivalent of 1,300 km² of pristine nature being converted into a completely artificialized zone. Again, this metric was experimental and has provided no information that could be used to influence the investment strategy. It's early days yet, but initiatives gain momentum. The Taskforce on Nature-related Financial Disclosures is likely to start its work in September, and should build on the success of the TCFD. And I'm quite confident that, within a couple of years, we should have more mature frameworks, and more robust data and methodologies. However, metrics are not the solution to nature-related challenges. The solution is action, which starts now, through stronger engagement.

This interview was conducted as part of a series looking at insurers' approach to sustainable investing. The interviews will be published on a weekly basis over the summer and will form part of a report on insurers' sustainable investment practices sponsored by JP Morgan Asset Management. All published interviews are available here.