The current results of climate stress-tests conducted by financial supervisors and the private sector suggest that climate-related risks are less pronounced than ‘traditional’ risks assessed as part of ‘normal’ stress-test exercises.
Current estimates around potential financial losses from physical climate risk oscillate around 5-10%, although certain institutions may have higher results. By being more ‘conservative’ in the models underpinning these approaches, they cater to the sensitivities of the central banks developing them. However, as a result, they also may hide more extreme outcomes (University of Exeter 2023).
A key challenge in interpreting these outcomes is the extent to which these exercises are limited in their scope of application.
Climate stress-tests (or scenario analyses as they are sometimes called) typically build on the NGFS “Hot House” scenario. There a number of challenges to this approach however:
- The scenarios represent a ‘central’ estimate under a high-carbon future and not some of the more pessimistic outlooks about the potential economic dislocation that higher temperature outcomes may bring.
- They typically do not consider the additional effects of climate tipping points (physical or social).
- They similarly understate climate risks at lower temperature outcomes as being effectively negligble.
- Finally, these scenarios focus exclusively on direct climate impacts and do not take into account the potential social or ecosystem shocks that may arise as a result of climate change.
This paper seeks to identify the potential levels of financial losses under a more proper ‘stress-test’ scenario for equity markets, focusing on the integration of climate, as well as social and ecosystem tipping points.
The paper combines academic research on the potential GDP effects of climate change with and without climate tipping points, ecosystem service loss, and social tipping points. It aggregates these risks to develop alternative GDP pathways until 2050 where risks from these factors materialize. These alternative growth pathways are then plugged into a multi-period discount dividend model to simulate implied valuation losses or lower return implications of these scenarios.
The models and calculations used here are by design rudimentary.
We describe the approach in this paper as “quantiative heuristics”. We do not provide a fully developed, integrated economic model, but rather build a simple GDP and discounted cash flow model that is shocked based on the estimates of economic impacts of the these different “tipping points” on GDP, sourced from third party literature. The effects are aggregated and then integrated into a “global discounted cash flow” that seeks to estimate the loss in future asset value in terms of today’s net present value. The simple approach allows for a clean isolation of different potential tipping points individual effects as well as a clear communication on the findings. Crucially, the analysis is not an attempt to quantify mispricing, but simply the delta in net present value of equity markets under a ‘no impact’ and ‘impact climate estimate.
While this report seeks to provide a broader basis for discussion about climate risks, it recognizes that it too may understate the total impacts. The ultimate economic impacts of climate tipping points is contested (Carbon Tracker Initiative 2023). What is more, dramatic economic shocks will obviously not just impact the valuation through lower cash flows but also higher long-term risk premia (the subject of an upcoming report). Our “quantitative heuristics” isolates the pure “economic” impacts and how they pass through to cash flows and thus provide a clear indication of the potential effects of these tipping points.
They are developed based on third-party academic and grey literature, and historical shocks involving similar events where they exist. Despite the uncertainty of these estimates, including them likely represents a more complete picture of what a “climate stress-test scenario” should or could look like.
Our findings suggest that climate tipping points, ecosystem (including biodiversity) decline, and social risks have the potential to amplify the financial losses in equity markets from climate change by a factor of 2.5-3.5x.
A high baseline climate risk (i.e. using a climate stress-test model with meaningful baseline GDP losses over the next 30 years) stress-test scenario can create a 10% shock to global equity markets. A combination of climate tipping points, ecosystem decline, and social risks can increase that number as a cumulative risk to 27%, almost 3x the baseline losses. A low baseline scenario of a 4% shock in turn turns into a 14% shock when considering these other factors. These losses are dramatic as they are secular and not cyclical. It is worth flagging that this event would be unprecedented in modern financial market history.
