Embedding good Environmental, Social and Governance (ESG) practices is a top priority for companies across business sectors. Many investors have long been using ESG data scores to make informed decisions, and these scores can be used to help understand insurable risk too. Sam Day, Head of Pricing Analytics ESG, and Katie Lennon, Head of ESG UK & Lloyd’s, discuss how ESG data can be used to model, price and underwrite risk.
ESG considerations are now integrated into the way many companies across all industries operate and interact with their stakeholders. It’s increasingly important for companies to be able to demonstrate good ESG practices to their shareholders, their customers and their employees. To that end, many are now using ESG ratings or scores to better understand their ESG performance, and to differentiate themselves from their peers. And, increasingly, ESG data forms part of the risk submissions we receive from clients.
As it is for our clients, developing ESG practices is a fundamental priority across the AXA group, and where permissible under local jurisdictional law, at 色多多视频we are exploring ways in which we can make this a core part of not only how we operate, but how we model, price and underwrite risk. We want to find sophisticated ways to use available ESG data to understand the links between good ESG performance and good risk management. And be able to communicate that to all parties in the insurance chain – clients, risk professionals, brokers and underwriters.
We recently hosted a market-wide ESG event to share and gather knowledge from across the insurance industry and beyond about the ways good ESG practice can improve risk profiles. One of the discussion topics was how ESG data can be used to understand, assess, model and transfer risk.
There are three main use cases of ESG data in underwriting. The first is responsible underwriting, whereby ESG information is used to restrict or limit underwriting of certain risks – like restricting coverage for companies engaged in thermal coal, for example. The second, we refer to as ‘underwriting plus’; this is where ESG information forms additional data points to help underwriters better understand and price risks.
And thirdly, we are aiming towards a position whereby we can practise ‘impact underwriting’; a scenario where there is a focus on social objectives to our overall portfolio as well as the financial objective in the way a risk is underwritten.
For example, AXA’s net-zero underwriting targets will begin to layer a company’s decarbonisation journey into underwriting, in combination with traditional risk assessment and appetite considerations.
Causation and correlation
There is evidence that there are both causation and correlation links between ESG performance, risk information and loss ratios, and this is an area where we can use ESG data in underwriting to achieve our goal of ‘underwriting plus’.
For example, for a given company there is a direct relationship between the quality of their employee health and safety standards and the respective employer’s liability (EL) underwriting risk. This is known as an ESG causal factor; enhanced safety standards lower the EL risk.
Correlation between ESG performance and loss ratios may be less immediately obvious. But our research has shown that there is often a distinct correlation between loss ratio performance and how well a company is able to manage its most financially material ESG risks.
Recently, for example, 色多多视频found a relationship between ESG risk and the loss ratio in one of our professional liability classes; this supported us in remediation work carried out on this book of business. Following this work, we undertook a study to determine the extent to which ESG data could help to predict losses in other lines of business.
The availability – or lack thereof – of ESG data continues to be one of the biggest challenges in assessing causation and correlation links, so we and others in the market, are putting in place strategies to improve this availability and better understand these links.
Finding the link
Our recent study examined the extent to which a section of clients’ loss ratios correlated with their ESG scores since 2015. Those ESG scores were obtained from a range of ESG ratings providers that all use slightly different ways to measure ESG performance but broadly assess a company’s resilience to long-term, financially-relevant, ESG-related risks.
We aimed to examine the link – if any – between the ESG score at the start of an underwriting period and loss performance over the course of that period; if we were able to establish a link then we’d be able to use that as a starting point to incorporate ESG scores into underwriting risk assessment and pricing matrices. This analysis was further bolstered by running it through actuarial models to view the ‘pure’ effect from ESG data.
Overall, the ESG study showed statistically significant results for some professional and casualty lines of business, and mixed results for property. We are undertaking further work to understand the results from property by, for example, removing the more volatile catastrophe events from the loss data.
How ESG data could shape underwriting
The ESG scores used by investors typically use weightings based on company and industry-specific materiality – essentially, what could impact the balance sheet. These ratings are relative to an industry benchmark. But underwriters also must consider so-called ‘double-materiality’. This takes into account not only the financial impact of ESG-related factors on the company – but also how that company is impacting the planet and society.
If we can harness ESG data and integrate it into pricing models, we’ll be able to remove friction from placements by obtaining data from authoritative and trusted third parties in addition to the qualitative and strategic input we hear from clients. This will reduce the need for additional information requests, making the underwriting process more efficient for all involved.
There’s huge potential to use ESG data in positive ways; not only to better model and price risks to help our clients be more resilient to ESG threats but also to help them become more net-positive to the world – and this is where we hope to get to with our aim of enabling impact underwriting.
It is clear achieving this goal will require collaboration. For the use of ESG data to be meaningful, we need consistency across the insurance market. We need a certain degree of commonality about the questions we ask of our clients, to ensure that data points are comparable and that we can derive useful insights from them. Our recent London market ESG forum discussion was heartening in this respect; there was a real sense in the room that others share our aims and goals, and we will continue to drive this collaboration in the months and years to come.
