Analytics matters: the bedrock of resilience - COP26, 8 November 2021
In recent years, we have witnessed the rapid development of global data and modelling capabilities to help quantify and monitor physical climate risk and support improved resilience decision making – much of it driven by methods and tools developed for the insurance industry. We are now entering into a new generation of models and data that will inform and guide more effective risk reduction strategies and capture the complexities of the risks of climate change.
On Adaptation, Loss and Damage Day at COP26, Willis Towers Watson gathered experts from across the global risk community to take on this topic during Analytics Matters: the bedrock of resilience.
This extended session began with the launch of the Global Resilience Index Initiative, which aims to provide the common language of risk needed to catalyse greater investment and innovation in resilience, before two panel sessions and a special showcase on risk modelling.
Natural catastrophes cost nearly $200bn globally in 2020. Since 1970, over 2 million people have died in climate-related disasters – 9 out of every 10 in developing countries. In the year running up to COP26, many countries faced a triple climate, economic and health emergency that still reverberates around the world, with prolonged impacts on people and economies.
The Global Resilience Index Initiative (GRI-I), launched at COP26, is the world’s first curated, open-source reference index which will address the data emergency that is contributing to the climate crisis. It will draw on the significant risk modelling experience of the insurance industry, constructing the basic building blocks of hazard, exposure and vulnerability to produce advanced risk analytics that translates physical risks into financial values, i.e. resilience risk modelling. Initially the focus will be on the extremes – like typhoons and floods – risks that create the biggest immediate impacts on people and economies.
The Global Resilience Index Initiative (GRI-I), launched at COP26, is the world’s first curated, open-source reference index which will address the data emergency that is contributing to the climate crisis.
GRI-I will first build a global infrastructure risk analytics platform, using advanced earth observation and artificial intelligence techniques, before building the next generation of analytics – modelling the impacts of shocks on human and economic systems and supply chains – allowing us for the first time to assess the macro-level and systemic physical climate risks that most threaten sustainable development.
By COP27, the GRI-I aims to have a live product that provides information and data on climate and resilience.
In opening the session, Rowan Douglas, chair of the Insurance Development Forum and Head of Willis Towers Watson’s Climate & Resilience Hub, pointed out that the institutions involved build on 15 years of work. GRI-I’s aim is to create “a map that can cure the world of the pandemic of climate fragility,” he said. “We are creating a human environment which is simply going to be unable to cope with and manage the risks we’re facing - we should be able to see and understand those risks but we don’t. They are invisible to the key systems that we depend on [such as] our financial system. Physics into finance is going to be the magic cure to this pandemic.”
“We are creating a human environment which is simply going to be unable to cope with and manage the risks we’re facing.”
Rowan Douglas
Head of Climate and Resilience Hub, Willis Towers Watson
In a video address, Mami Mizutori, head of the United Nations Office for Disaster Risk Reduction, and patron of the GRI expressed confidence that the GRI will be key to effective adaptation action by helping to build the next generation of risk analytics fit for the future and to help develop comprehensive disaster management plans.
“The number of people displaced by extreme weather events last year reached over 30m last year,” she said. “‘Normal’ has gone forever. The impacts of climate change are outpacing our efforts to build resilience. We desperately need climate solutions based on solid science, data and analysis. Improved risk analytics can underpin public and private sector action to support investments in resilience. There is still hope the world will finally act to rein in global warming, but we must prepare ourselves to live with and manage the consequences of the damage already done. I am confident the GRI will be a powerful tool that will guide us all through this journey.”
Ekhosuehi Iyahen, Secretary General of the Insurance Development Forum, noted that Mark Carney, UN Special Envoy on Climate Action and Finance, asked her organisation to address the challenge of a global reference index for the assessment of resilience, acknowledging how the sector’s risk management capabilities should have applications beyond insurance.
Ms Iyahen referred to the usefulness of the GRI for the world’s most vulnerable countries, the V20, in deepening risk understanding and analytics. The GRI would “set the scene for more locally country- focused work on adaptation and disaster risk financing and disaster risk programmes”.
“This effort is about democratising the availability of risk information and ultimately building a shared language of risk. It is about access to information for public good and the demand could not be higher.”
Kamal Kishore, co-chair of the executive council of the Coalition for Disaster Resilient Infrastructure (CDRI) and Secretary to Government of India, said: “While there is a quantitative side to understanding resilience, there is also a qualitative side, how regulations work, what is the level of enforcement - what is the lived experience?
“While we’re talking about global resilience, all resilience is local- and that’s the role of CDRI so that the governments use these tools. There is no glory in having a good tool that is not being used to guide investment at the national level.”
Dr Ben Caldecott, director of the UK Centre for Greening Finance & Investment (CGFI), emphasised the importance of the GRI in creating a platform that can translate the data and make it useable for investment sector as only a small proportion of data currently makes its way into the finance sector.
“The logical, natural next step of this focus on Net Zero transition plans will be resilience plans – that can happen fairly quickly. ”
Ben Caldecott | Director of the UK Centre for Greening Finance & Investment
“Our vision is for financial institutions around the world to be able to use climate and environmental data and analytics at any point on planet earth in the past, present and projected into the future for all the major sectors of the global economy and to cover – creating public goods.
“The logical, natural next step of this focus on Net Zero transition plans will be resilience plans – that can happen fairly quickly. The work that we’re doing here will support that future regulatory step and resilience finance.”
As COP negotiators gathered for adaptation, loss & damage day, risk modelling experts shared their insights on best-in-class solutions to increase resilience and help countries adapt and limit their losses to shock events.
In opening the session, Matt Foote, head of analytics in the Climate & Resilience Hub at Willis Towers Watson, introduced the panellists as those “tasked with supporting the most vulnerable in reducing the protection gap and building the resilience needed to protect us all against current and future climate risk.”
Open-source data needs to fill these gaps in knowledge for lots of difference audiences - insurance, investment, humanitarian, disaster risk reduction and utilities, agriculture, said Jim Hall, Professor of Climate and Environmental Risks, Oxford University. “There is some conceptual work to develop a framework so that they can talk to each other. There are gaps in knowledge of assets on the ground, their physical vulnerability (including design standards) and their interdependencies with infrastructure. Climate conditioning with statistical data that is highly uncertain needs more work on real gap in knowledge in assets physical vulnerability."
Flood risk modelling is particularly challenging of all the physical hazards, but Emma Raven, Technical Director and Climate Change Lead, JBA Risk Management, said that they had recently started to include climate change modelling, with Artificial Intelligence helping them make a big leap forward in data, innovation and consistency around risk.
“A few years ago, if I’d said I was going to build a global flood model, it would have been met with a lot of scepticism. We’ve come a long way.”
Jeff Neal
Co-Founder and Chief Scientific Officer, Fathom
However, traditional catastrophe models need to adapt, she said: “When we set out to build our first global flood model we very quickly became aware that the volume of data and the method and the way in we build these models need to change and the traditional ways of building natural catastrophe models needed to move forwards and we brought in innovation in software to build these models more dynamically.”
Jeff Neal, Co-Founder and Chief Scientific Officer, Fathom, said: “A few years ago, if I’d said I was going to build a global flood model, it would have been met with a lot of scepticism. We’ve come a long way. We’ve now got modelling systems set up that can run globally for present day and have evolved to run into the future. We can use those in a catastrophe risk context. There’s still a lot of work to do especially if we are going to bring these out from the insurance sector into other areas.
“Flood modelling has traditionally been siloed [but] much like the climate modellers think of the climate as a series of data sets that tell them about future climate risks, we need to move that way also with flood and other perils.
“Data sets on resilience and exposure have been poor in the past. Our models need to become more aware of the impact that humans have on hydrological systems, eg how does infrastructure such as dams influence flows. Conceptually it’s quite simple but at the global scale incredibly complex and difficult to deal with.”
Christophe Christiaen, Data, Innovation and Impact Lead, Oxford University, UK Centre for Greening Finance & Investment, said the university had set up the Spatial Finance Initiative to try to mainstream geospatial data and analysis in financial decision making.
“Many of the climate or social crises we talk about today are inherently spatial – impacts or risks manifest themselves differently depending on the local context. In the past 5-10 years we’ve seen an exponential increase in the availability of earth observation data and geospatial data. It tells us what’s happening on the ground – at the real asset level, in near real time and with global coverage.
“But to link this data to financial decision making you need to know where these critical assets in our economy are located, what they look like, who owns them, etc.
“Why should they be global and open? Global because these investors and sectors operate in global markets to be able to benchmark across companies. They should be open because we believe that this type of data can increase transparency in the financial system between not just investors and investees but also between regulators and investees or investors and civil society. By making it open you create a level playing field for countries and drive climate action and resilience.”
Marc Lehmann, Head of Business Development & Partnership, Jupiter Intelligence, said: “We talk a lot about the hazard component of the risk equation. But having really good vulnerability information helps us quantify the risk for monetary values, particularly property damage and business interruption.
“After working for 20 years in the insurance sector, the challenge is often supply chain disruption. We do a pretty good job of assessing the physical damage and direct business interruption but the broader macroeconomic impact and supply chain disruption is an area I’d like to see more research into. There are a lot of initiatives but it’s a case of bringing that together so companies, government and organisations can take a systemic approach to its management.”
The industry already has precedent for multi-model open platforms, such as the Oasis Loss Modelling Framework, and led the development of an open exposure data format, with input from across the insurance market.
James Lay, commercial director, Nasdaq, said: “The insurance industry has been using catastrophe modelling for 30 years or so – it’s tried and tested, it’s pretty robust. We don’t need to reinvent the wheel here – we have a lot of these technologies available now. Catastrophe models are based on historical data, what’s happened in the past, and climate models are projections based on what’s going to happen in the future. There is uncertainty in catastrophe modelling and there’s even more uncertainty in climate modelling. We need to be clear on that and engage with the community that’s building these models to find standard ways of approaching these challenges.
“Nasdaq, OASIS and a number of insurance companies created an exposure data format that was open and usable by anybody and they don’t have to spend a penny on making that happen. These standards can’t be created in isolation and they shouldn’t be proprietary – the openness is extremely important and ultimately powerful.”
In introducing the second panel, Matt Foote quoted Dickie Whitaker, Chief Executive, Oasis Loss Modelling Framework:
“Open Data Standards (ODS) are key to ensuring continued transparency and are fundamental to improving interoperability that will increase efficiency for the market, reduce costs, and enable greater choice in the use of catastrophe models.”
“without them you get data anarchy…”
While public and private sector collaboration on analytics can be most effective when there is a focus on the financial sector and investments, are insurance-based models enough to address resilience needs?
Nicola Ranger, Deputy Director, UK Centre For Greening Finance and Investment:
“Across finance and beyond we’re not bringing physical risk into financial decision making yet and there’s underestimate of risk so there’s a mispricing that has a financial impact. When there’s a material risk that’s not being properly captured it also has a bigger societal risk because … that risk signal isn’t going to corporates, homeowners, and the incentives are not in place for people to manage that risk.”
“The area we are very keen to push at CGFI is how we align broader financial flows with adaptation and resilience goals. We’ve heard at COP a lot about aligning financial flows with Net Zero targets but adaptation and resilience has been largely missing but we need that same process to happen.
“Globally $2.7tn a year is going into infrastructure investment – so we have a measly amount of adaptation finance going one way, and a tidal wave of infrastructure finance going the other which is not accounting for resilience needs.”
Ruwadzano Matsika, Research Analyst, Quadrature Climate Foundation:
“We know that for billions of people, not just in climate-vulnerable nations but in climate-vulnerable communities that have lower adaptive capacity climate change is not if it’s not when it’s happening now so investment into adaptation and resilience measures is equally as important and we should be looking at that right now.
“But there is a job to do to make sure the translation of these climate risk analytics into operational usable information that informs climate responsive planning and adaptation measures so that we avoid the risk of maladaptation and that’s very real if you basing decisions on irrelevant data.
“We need to strengthen capacity at country level we need to think about who are the end users in that information value chain and ensure that they have the capacity to engage and interrogate these models not just that this model is a black box. We want to promote local and country led data sharing initiatives not just at town or muni or subnational level but between nations because as we know climate change impacts don’t respect political boundaries…. It’s going to take all of us including philanthropy and donors to do their part to help catalyse and scale these initiatives.”
Antoine Bavandi, Senior Risk Finance Specialist, World Bank:
“We have been developing transfer mechanisms to build capacity in those countries, to integrate country and regional level modelling and give knowledge to those who made the decisions.
“We understand the need for a consolidated, consistent platform. We have a next generation drought index which aims to produce state of the art drought risk assessment sourced from publicly available data. We believe that open data is sufficient for drought risk financing.
“We worked with central banks to assess and monitor physical climate risks in eight different countries in West Africa, digging into the financial vulnerabilities of banks to droughts, floods, heatwaves, sea level rises, to pose the question could climate risk pose systemic risk to banks and the financial sector in west Africa.
“A global platform is essential and creates the stepping stones towards more sophisticated models.”
Nick Moody, Risk Modelling Steering Group, Insurance Development Forum:
“What we’re part of here is nothing short of system change. Many of us here are very involved and especially the practitioners are involved in particular in the complexity of catastrophe modelling.
“If we are going to turn this into a movement or system change we need to be able to multiply the user base for a technically capable GIS person in a ministry or disaster risk management agency in a country … in a very accessible way to achieve meaningful results that inform local decision making that in a way in which we are going to multiply this movement.
“The introduction of climate warming pathways and scenarios does add complexity but that’s something that we’ve also got to build in and make it accessible to the end user we ultimately seek to serve.”
