Financial Condition Testing for Climate Change: A Possible Approach

Imagine you are the chief risk officer of XYZ life insurance company. One afternoon, you receive an email from your chief actuary. Following the CIA’s email on FCT updates (August 28, 2025), which includes references to climate-change stress testing – they are trying to convince both you and the CFO to support including these considerations in the next FCT exercise. The chief actuary is also questioning if you could provide assistance in this regard.

You panic while reading the email as your mind races to formulate a response. Climate-change risk is something you have been aware of on the horizon of your work, but you have not given it the attention it deserves.

That is about to change. You respond to the chief actuary by letting them know that you will be looking into the matter and will report back shortly. Your stress level has increased significantly. But have no fear: the pathway below will shine a light on one possible, practical approach to modelling climate change for FCT purposes.

Relevant practice standards and educational guidance

Your first step is to pull together some of the reporting standards and educational notes that you can find on climate-change disclosures. You reference:

• CIA, Climate Scenario Analysis
• CIA, Practice Resource Document: Climate Change Scenario
• CIA, Financial Condition Testing
• Office of the Superintendent of Financial Institutions, Climate Risk Management
• Autorité des marchés financiers, Climate Risk Management Guideline
• International Financial Reporting Standards, IFRS S1 General Requirements for Disclosure of Sustainability-related Financial Information and IFRS S2 Climate-related disclosures
• International Actuarial Association, ISAP 8 – IFRS S2 Climate-Related Disclosures (adopted by the CIA) and IAN 200 – Application of IFRS S2 Climate Related Disclosures
• CIA, Integrating Nature to Climate Scenario Analysis for Enhanced Resilience

Much of this guidance draws on similar conceptual frameworks, typically divided into sections such as governance, strategy, risk management, and metrics and targets. Many also refer to physical and transition risk.

Physical risk can be defined as the risk from climate events such as floods, wildfires, storms, heatwaves, cold snaps, etc. Transition risks refer to the risks, and in some cases opportunities, as we move to a lower greenhouse gas world.

Scenario selection as a starting point

But how do you incorporate these standards and educational notes into FCT? The brilliant genie that lives rent-free in your head suggests that you first start with identifying scenarios to model.

You recall a webinar discussion on Network for Greening the Financial System (NGFS) scenarios, seven long-term climate scenarios used by central banks and supervisors to “provide a window into plausible futures”:

• Net Zero 2050
• Below 2 °C
• Delayed transition
• Nationally determined contributions
• Fragmented world
• Current policies
• Low demand

Recently, the NGFS further developed four short-term scenarios:

• Highway to Paris
• Sudden wake-up call
• Disasters and policy stagnation
• Diverging realities

These short-term scenarios focus on a five-year horizon and place significant emphasis on physical and transition risks. However, you decide to use the long-term scenarios for your stress tests, given their alignment with balance-sheet and capital considerations.

Linking climate scenarios to mortality and morbidity

You begin by looking for research articles on relationships between climate change and mortality and morbidity. The data is sparse. One possible approach is to project out energy usage and the transition of CO₂ emissions under the long-term NGFS scenarios. Then you could look at causal relationships between the CO₂ pathways and mortality and morbidity outcomes.

• At least one study has linked climate change to lung cancer.
• The Canadian Institute for Health Information has published an excellent report in 2023 on the effects of excessive heat on mortality and morbidity.
• Other research from Sun Life Financial focuses on the health effects of climate change on health disorders.
• Other studies have looked at climate change effects on diabetes and a host of other diseases.
• One study found that there is a 60% spike in mental health illness in the six months following a wildfire for those impacted.

The challenge for you is that many of these research articles don’t provide an adequate roadmap of how to incorporate climate change into FCT work. This limits their direct applicability for FCT modelling.

Liability stress testing using climate-adjusted mortality tables

Given the thin and disjointed level of data on climate mortality and morbidity, you decide to take a different approach. You incorporate a climate mortality table into your scenario testing for your life business to get a better handle on the effects of climate change on your capital and balance sheet.

You will use the same table for your annuity and pension de-risking business. For your employee benefits/worksite marketing business, you will use a climate morbidity table for stress testing the NGFS scenarios.

Let’s assume that XYZ has 10,000 policyholders in its life insurance business. All are age 65 and female. Each has a life insurance policy with a death benefit of $100,000. An excerpt from the climate mortality tables for the Net Zero 2050 scenario looks like this:

Scenario: Net Zero 2050

Tables reprinted with permission from the CliScen model.

Claims payout at age 65 = 10,000 * $100,000 * 0.00003

                                                     = $30,000

For simplicity at this point, discount rates, premiums, lapse rates and expenses will not be discussed. The model used shows an uptick in reserve for all future claims of $350,000 under the Net Zero 2050 scenario. Under the seven long-term NGFS scenarios, the range of results is $225,000-$1.2 million.

This becomes a liability stress that must be funded by free capital. Resiliency can be demonstrated by showing there is enough free capital to cover this uptick in mortality from climate change.

Asset modelling under climate stress

On the asset side, one possible approach is to apply a market value climate adjustment algorithm to depress the market value of energy assets on the balance sheet. Correspondingly, the sustainable assets can have their market values increased with the algorithm.

Let’s delve into this concept a little further. XYZ owns an energy asset that’s currently trading at 250 basis points over the Bank of Canada’s 10-year bond yield. However, the climate asset algorithm suggests that the asset should really be trading at 400 basis points above the Bank of Canada’s 10-year bond yield given the current climate (no pun intended) around energy assets with many investors. The climate asset market value algorithm would be based on established parameters. It reduces the market value of the asset based on what it perceives to be the “correct” spread in the market for an energy asset. On the flip side, renewable assets may have their market values increased.

Let’s look at some of the assets in XYZ’s portfolio (in $):

After running the assets through the climate risk algorithm, scenario market values come back as:

Scenario: Net Zero 2050

The B Energy and R Renew assets have had their market values adjusted under the NetZero 2050 scenario by the climate risk market value algorithm. Under that scenario, the combined market value of these two assets were depressed by:

$22,500,000 – $21,700,000 = $800,000

Combined balance-sheet impact

The total capital reduction under this scenario from both the liability and asset side of the balance sheet becomes:

Capital reduction = $350,000 + $800,000

                                         = $1,150,000

XYZ holds $30 million of surplus allowance. This climate change scenario would represent a:

Surplus change = ($1,150,000)/$30,000,000

                                          = -3.83% of the surplus

This does not represent a significant impairment to XYZ’s capital; however, the results reflect the NetZero 2050 scenario. Other scenarios may be more or less punitive, and additional blocks of XYZ business remain to be modelled.    

Conclusion: One practical roadmap

We’ve looked at how climate change stress testing could be incorporated into insurers’ Own Risk and Solvency Assessment and financial condition reporting. While it’s certainly not the only approach, it provides a roadmap for practitioners challenged with how to incorporate climate risk modelling into their overall risk management framework.

It is important to recognize that climate risk modelling remains an evolving area of practice. Data limitations, modelling uncertainty and jurisdictional differences all pose challenges, and results should be interpreted as indicative rather than predictive.

As more knowledge and best practices emerge under climate change risk modelling, it’s expected that the approaches and actuarial standards of practice will become more streamlined.

Terry Narine, FCIA, FSA, is a climate actuary, peer reviewer and President of ACTUWIT Consulting, an actuarial firm specializing in climate risk, regulatory reporting and sustainability solutions for insurers. He is the CIA’s Delegate to the Climate and Sustainability Committee of the IAA and has held senior volunteer leadership roles with both the CIA and the IAA, including in the IAA Health Forum. He can be reached at [email protected].

Any views and ideas expressed in the article are the author’s alone and may not reflect the views and ideas of the Canadian Institute of Actuaries, its’ members, or the author’s employer. No AI was harmed in the production of this article.