How useful are climate scenario models?
Pardon the Interruption
This article is just an example of the content available to mallowstreet members.
On average over 150 pieces of new content are published from across the industry per month on mallowstreet. Members get access to the latest developments, industry views and a range of in-depth research.
All the content on mallowstreet is accredited for CPD by the PMI and is available to trustees for free.
Several local authority pension funds have published climate scenarios which project a failed transition would lower annual investment returns by between 0.07% and, at the higher end, 1.21%. How useful are scenarios to understand the possible impact of global warming?
Local Authority Pension Scheme funds are not yet subject to reporting in line with the Task Force on Climate-related Financial Disclosures, but many of them already produce either a TCFD or climate risk report. A consultation is currently underway proposing that the first reporting year for LGPS would be 2023-24, with first reports expected by December 2024.
LGPS investors seem confident that their assets will not perform significantly worse even when the transition to a greener economy has failed and the world is set to warm by 4 degrees, despite the extreme physical risks that would crystallise.
For example, the West Midlands Pension Fund states that under a 4-degree scenario, after 40 years, annualised returns would be just 1.1% lower with its current and 1% lower with its target asset allocation. At the Worcestershire Pension Fund, it’s 1.21% in both cases, and Leicestershire sees returns down by 1% with either its current or an alternative asset allocation after 40 years.
So does the Derbyshire scheme, which updates its scenarios every two to three years. A spokesperson for Derbyshire County Council said the fund is clear there is great uncertainty for investors around the risks associated with climate change and with changing climate policies.
“Climate scenario analysis forecasts different possible eventualities across a range of scenarios. As a developing field, which by necessity uses assumptions about inherently unpredictable matters over long time horizons, it is prudent to view the outputs from the analysis as directional information on the sensitivity of the fund’s portfolio to different climate scenarios, to be considered in tandem with all the other factors which have the potential to impact on investment returns,” the spokesperson said.
Leicestershire said the ‘failed’ scenario presents a significant reduction in investment returns.
“This scenario presents one of many plausible scenarios. However, the probability of this, or any given scenario, is hard to determine given it is based on factors that are subject to considerable uncertainty,” a spokesperson for Leicestershire County Council said.
“The analysis should not be used as a tool to predict the future but as one way we can try to better understand the impact climate-related risks would have on the fund,” the spokesperson added.
Perhaps surprisingly, the Environment Agency Pension Fund also modelled scenarios, in a 2019 report by Mercer, claiming its annual portfolio returns would shrink just 0.19% by 2100 at 4 degrees warming with no sustainability allocation, and 0.16% using its strategic asset allocation. The analysis, undertaken every three years alongside a review of the strategic asset allocation, is currently being completed again by Mercer. This year, the EAPF will consider the risk from nature loss, as well as transition and physical risks from climate change.
LGPS Central, the investment pool with which the aforementioned funds – apart from the EAPF – are associated, says it uses climate scenario analysis to manage climate-related risks in its investments, in line with the TCFD requirement.
A spokesperson said: “The findings from our climate scenario analysis affirm the materiality of climate risks to our investment portfolio. These risks, encompassing both transition and physical risks, have the potential to significantly impact the financial performance and long-term sustainability of our investments. Our approach to climate risks will continue to evolve, in line with industry best practices.”
With the agreement of its partner funds, LGPS Central procured Mercer in 2019, with a first analysis in 2020 being followed by a second in 2022. The latter saw Mercer partner with Ortec Finance and Cambridge Econometrics.
The usefulness of climate scenarios that produce exact numbers does not appear to be seen as overly problematic, although when asked, funds generally say that they need to be taken with a pinch of salt.
The LGPS Advisory Board, in its response to the consultation on TCFD reporting, said that “while challenging, consideration of different scenarios is valuable in terms of bringing these reports to life. In terms of communicating with members, scenarios can have a valuable educational role as well.”
It suggested two or more scenarios showing the effects of above 2 degree warming. “We believe that these scenarios should not be limited but that funds should choose from a small, defined set of scenarios to ensure comparability between funds,” it added, suggesting the inclusion of “a small set of plausible scenarios” in statutory guidance.
Board secretary Jo Donnelly told mallowstreet that the board plans to consider scenarios further: “We will soon be discussing with pools what climate scenario modelling will be done at pool level for funds, and whether/how that can be standardised, whilst acknowledging that this is an evolving area and science.”
Regarding the use of advisers for climate reports, she noted only that “this is still a relatively new market and demand for services is high”.
Is there a lack of diverse views in LGPS funds’ climate scenarios?
Most of these LGPS funds’ climate reports are based on a single approach by consulting firm Mercer, ‘Investing in a time of climate change – the sequel’ from 2019, which offers to model three scenarios – the world is on a path to 2 degrees, 3 degrees and 4 degrees warming. The latter involves sea level rise of approximately 70cm on average, 50% less water availability and the strongest Northern Atlantic cyclones increasing by 80%, according to Mercer.
Despite this, the consulting firm’s own corporate TCFD report shows it sees a model growth portfolio annualised return impact of just -0.7% after 40 years if the world is failing to transition.
The consultancy highlights that the annualised return impacts lead to larger falls in asset value, with a –0.7% p.a. impact for the model growth portfolio leading to a 24% fall in asset value after 40 years relative to the baseline projection.
“These cumulative return impacts form a key part of the justification behind Mercer's finding that an orderly transition is an imperative for long-term investors,” a spokesperson for the firm said.
The impact on a globally diversified equity portfolio (MSCI ACWI) is a reduction of value between 30% and 40% under a failed transition by 2062 compared with the projected portfolio value under a baseline scenario, the firm said.
“Nonetheless, Mercer does recognise that there are limitations to its approach to climate scenario modelling, as with all models,” the spokesperson said, noting that these are set out in the report.
This includes rather big-ticket items – "for example, that the model does not fully capture climate tipping points, the impacts of war or mass migration. Therefore, physical risks could be understated.”
TPR: ‘We have reservations about some scenarios’
Large private pension schemes are now obliged to produce a TCFD report, and the Pensions Regulator has recently reviewed the first round of these, finding some good examples but room for improvement.
A spokesperson for the regulator said: “Our review of a selection of occupational pension schemes’ annual climate reports showed a wide variation in climate scenarios adopted by trustees.”
Most scenarios were broadly aligned with the three described in the Department of Work and Pensions’ statutory guidance when it was issued in June 2021: an orderly transition, a disorderly transition and a ‘hot house’ world.
“We expect the market to evolve and a degree of consensus around scenario analysis to emerge, but we have reservations about some scenarios more generally,” the spokesperson said.
“We accept climate change impacts are difficult to effectively model. It is important [that] trustees, and their advisers, consider the potential impacts on scheme performance in each of the scenarios chosen and explain the reasons for their conclusions in their climate risk reports."
TPR will examine scheme reports on scenario analysis in more detail by carrying out a thematic review on scheme resilience to climate-related scenarios, it said in its 2021 climate change strategy.
Official BoE scenarios come under fire
The question of whether climate scenarios are useful tools has been raised in the Lords Grand Committee, where Baroness Bryony Worthington, an independent peer, criticised the Bank of England’s Climate Biennial Exploratory Scenario, which uses an early action, late action and no action scenario.
“The [Prudential Regulation Authority] offers up sample temperature rise scenarios and underlying assumptions of the implications for different assets, and firms plug in their portfolios to get the impairment data out as a result. This all feels safe and precise, but the climate is something that cannot be predicted specifically in those ways with any degree of accuracy,” she said during a debate in March.
“Critical methodological problems have led to perverse outcomes, such as the suggestion that a 3-degree temperature rise, global average, offers the optimal balance of benefits and costs,” she noted, saying there is global consensus that such temperature rises would lead to steep drops in food production, water shortages, urban heatwaves, forced migration and mass extinction.
She cited a former chief economist of ING Group, who said central bank models ignore or downplay the rising frequency of extreme weather events and critical triggers, tipping points and interdependencies between the climate, the economy, politics, finance and technology.
“That is true for the CBES model,” she argued, which “highlight minimal economic impact from inaction on climate change over the next 25 years and a reduction in GDP growth of only 0.12% in 2050 — another ludicrously precise number, given all the future uncertainties that lie ahead of us. That is very poorly aligned with the scientific consensus.”
Baroness Joanna Penn, a parliamentary secretary for the Treasury, responded that while all models have their limits, the Bank of England had “considered the views of experts in the field; they therefore do not need to be directed to do so”.
The CBES were produced in partnership with leading climate scientists, applying climate-economy models that have been “widely used to inform policymakers — not to mention being used by and continuing to be used by the Intergovernmental Panel on Climate Change”, she said.
But some climate risk advisers are also discontent with the CBES. Mike Clark, founder of specialist advisory Ario, says the official scenarios play out over too long a timescale and are therefore not useful when making investment decisions.
“Boards, including pension fund boards, are often attuned to a three to five-year time horizon for strategy decisions, not a 30-year one, so urgency is lost,” he warns.
Modelling which "skates over innovation and stranding” is a long way from the real world, he argues, and advocated the use of narrative scenarios over those that produce numbers.
“We need narrative scenarios to ask the right questions – geopolitics, tipping points, climate policy developments – before we even start to model”, he says.
The official scenarios matter for pension funds because they are adopted by some investment consultants. For Clark, consultants’ thinking has not evolved fast enough.
“If some LGPS schemes are publishing very modest reductions in expected returns at significantly higher temperatures, as is the case, [the Department for Levelling Up, Housing and Communities] should be listening to the climate scientists who don’t know whether to laugh or cry at what is stated in some of the public reports,” he says.
Do the models currently used for climate scenarios produce plausible outcomes?