Climate risk has caused significant challenges for life insurance financial risk modelers. There is a lot of noise in the market, with emphasis on either very precise or very complex modelling. However, it is important to also look at the broader, systematic impacts on financial markets exposures; such as yield curves, inflation, credit spreads, risk premia, and asset class returns.
How do you find the balance and a sufficient level of granularity in your modeling? How do you select and assess the impact of different climate models? Are there risks we are missing? We have created a series of short articles to help those responsible for climate modeling in the actuarial, risk, and strategic asset allocation functions of a life insurance company. These articles begin to address and simplify the complexities of climate change in processes such as the ORSA, stress testing, and strategic asset allocation.
In the first of the series of six articles, we explore sectoral level impacts using top-down modeling.
Traditionally, when tackling either risk management or strategic asset allocation challenges, insurers analyze their holdings at asset class level. For climate risk, and in particular transition risk, it is natural to assume that there will be both winners and losers within any asset class: firms and sectors which under or outperform the overall asset class.
One approach would be to jump directly to a bottom-up analysis and look directly at each individual equity or credit holding. While this approach has merit, it can miss some impacts. Costs are passed on to the wider economy through higher prices from firms with high tax, energy, or investment costs.
Starting from a top-down model and decomposing into sector-specific impacts, attempts to capture these broader dynamics. This approach still recognizes that some sectors will be hit harder than others, and not all costs can or will be shared equally across the economy.
How granular should models be?
Sector analysis starts by defining a set of relevant sectors to model. Too few, and you risk too little distinction in the results. But too many can introduce noise, make the results harder to interpret and overestimate how accurately we can predict specific effects.
The detailed energy-climate-land system models (Integrated Assessment Models), run by groups such as the Network for Greening the Financial System (NGFS) and used in their reports by the Intergovernmental Panel on Climate Change (IPCC), usually decompose their outputs into a broad set of categories. These categories cover buildings, energy, agriculture and land use, transportation, and industry. Within these, there are often further subdivisions, particularly across the energy supply system and transportation, as these are key for those models.
Economic modeling, for example in national accounts, uses a different set of sector definitions. Often with more granularity on services and manufacturing, but less on transport and energy supply, reflecting where value is added and people are employed. Mapping between the energy system definitions and economic sectors is an important step in producing sector outputs, and is a limitation on how granular we can meaningfully make sector-specific outputs.
Model risk in sector analysis
Each Integrated Assessment Model makes different assumptions about the nature of climate transition, about the costs of different technologies, and how these will change over time. These assumptions fundamentally drive the outputs from the models, and the economic and financial impacts we infer.
At a macroeconomic level, these differences are not always large (though they can be) if the models assume similar levels of carbon pricing and emissions trajectories. Dig down, however, and the differences can be significant. One model may assume a large contribution from biomass, with associated impacts on land use, while another uses nuclear as baseload for electricity. How much carbon capture is assumed? Are vehicles and heating powered with hydrogen or electricity? All of these choices affect the demand for electricity, mining, agriculture, refining, and so on, producing quite different sector impacts.
Therefore, before diving too deeply into a detailed analysis of one model, it is important to consider the range of possible futures and assumptions that feed into your calculations. Top-down analysis can set the context for bottom-up modeling and allow you to avoid overconfidence in both the precise way in which transition will play out in practice, and where risks and opportunities will arise.
For more insights on this topic, listen to our podcast series.
Speak to our Experts about how we can help with your climate-related modeling needs.
