Dealing with the ‘unknown unknowns’: a precautionary approach to financial regulation in the face of climate change
By Hugues Chenet, Josh Ryan-Collins and Frank van Lerven
Twitter: @jryancollins and @frank_vanlerven
In one of his last interviews before departing, Bank of England Governor Mark Carney made clear that the financial sector was not doing enough to face up the threat of climate change. To try and speed up action, the Bank announced before Christmas that the UK’s major banks and insurers would be ‘stress-tested’ on their ability to deal with the financial risks posed by different climate change scenarios. The French and Australian financial regulators have made similar announcements.
These stress tests may well reveal some useful information to regulators about how ready financial institutions are for climate-related shocks. But the question will then be what action regulators will take, if any. The logic behind stress tests is questionable, as we argue in a new IIPP paper. They are based on the idea that the risks arising from climate change can be calculated and then acted upon accordingly — that is, that they are knowable and quantifiable.
That the physical risks from climate change are unpredictable is becoming all too obvious, and heartbreakingly so. Take the Australian bushfires which have engulfed an area the size of South Korea along Australia’s East Coast, destroying 6000 buildings, killing 29 people and an estimated 1 billion animals so far.
With hindsight, it is clear that the long drought the country has experienced and very high temperatures — both associated with climate change — had created unusually dangerous conditions. But in New South Wales, one of the worst affected areas, fire detections this year are currently four times higher than the previous highest (2002–2003) and almost eight times the seasonal average (see figure below). The country was understandably unprepared. It is questionable whether regulators would have been taken seriously had they presented such a scenario to banks as one that should be ‘stress-tested’ for.
As well as physical risks, climate change also poses transition risks as policy makers attempt to decarbonise the economy, resulting in ‘stranded assets’ as the market valuation of certain sectors (e.g. coal) suffers unanticipated losses.
Stress-testing, along with disclosure initiatives such as the Financial Stability Board’s Taskforce on Climate-related Financial Disclosure, lies very much within a ‘market-fixing’ approach, which assumes that the financial risks from both the ‘transition’ and ‘physical’ elements of climate change arise from a lack of information and transparency.
Once this information shortfall (or asymmetry) is corrected, the logic goes, financial institutions and markets will use the newly-provided information to efficiently ‘price in’ climate-related factors, and thus avoid a build-up of unseen risks.
We argue instead that climate risks are subject to radical uncertainty. Whereas the concept of ‘risk’ in financial forecasting implies the ability to assign a probability to an event happening, under conditions of fundamental uncertainty, it becomes impossible to do so. As Donald Rumsfeld famously put it, there are some things we do not know we do not know.
This radical uncertainty arises out of the complex interaction between the natural environment — a multi-dimensional non-linear system composed of the atmosphere, ocean, cryosphere, biosphere, etc — and policy and regulatory changes, technological innovation, changing consumer preferences and the highly interconnected global financial system. The latter has the capacity to propagate and amplify uncertainty (and the problems that come with it) rather than containing it within particular institutional, sectoral or spatial domains.
Instead, we propose applying a ‘precautionary approach’ to climate-related financial policy. The ‘precautionary principle’ justifies the use of discretionary preventative policies that protect human health and the environment in the face of scientific uncertainty. It is well established in the environmental protection sphere, but was less well accepted in the sphere of financial regulation up until the global financial crisis (GFC) of 2007–08.
However, the GFC made clear the limitations of conventional financial risk-modelling approaches that attempt to forecast future risks based upon previous data. In its aftermath, regulatory innovations — in particular macroprudential policy, resolution planning and stress testing — can be seen as a shift in the direction of a precautionary approach. The starting point of macroprudential policy is to take preventative action to increase the resilience of the financial system to hard-to-predict-shocks, including rare ‘tail’ events such as financial crises.
Regulators are reluctant to take more concrete policy action on sustainable finance because they lack scientific certainty about the nature of the risks to the financial system posed by climate change. But they took the same stance prior to the financial crisis of 2007–08 in regard to the housing market — and regretted it afterwards.
Applying the precautionary policy approach to climate ‘transition’ is straightforward: transitioning away from high emissions activities and products creates significant uncertainty over future financial stability and raises the risk of a rare and catastrophic event — such as a financial crisis. Under such conditions, the precautionary policy maker has a strong incentive to act to insure the economy against such events, even if there are no available models that can predict the probability of such an event happening.
In terms of implementation, a precautionary financial policy approach should make it easier to justify the comprehensive integration of climate ‘risks’ — however poorly understood — into the central bank and supervisory toolkit.
For example, a penalising factor could be introduced to force banks to hold more capital if they have large exposures to high-carbon sectors. Climate-related financial risks could also be integrated into monetary policy operations via adjustments to asset purchases and collateral criteria which are currently ‘market neutral’ but in fact create portfolios that are weighted towards supporting incumbent greenhouse-gas emitting sectors. Quantitative credit controls and credit guidance could also be considered.
These policies would entail following a more interventionist approach than central banks have become accustomed to, but it is one that is potentially justified in the light of the catastrophic threats the world now faces, as highlighted in the Intergovernmental Panel on Climate Change’s recent 1.5 degree warming report.
A precautionary policy approach also means rethinking how regulatory interventions are appraised and evaluated. Conventional cost-benefit analysis is not appropriate because the well-established irreversible nature of climate change threat gives rise to potentially infinite costs. The entire concept of discount rates, necessary for optimising against a future point in time, is deeply problematic in this context.
Instead of optimising (short-term) market efficiency and focusing on prices, the focus should be on avoiding tipping points and thresholds, and building system resilience.
Taking this approach, financial supervisors can steer market actors in a clear direction — towards a managed transition — to ensure that a scenario that minimises harm to the financial system and wider economy in the future is the scenario that actually occurs.
Josh Ryan-Collins is the Head of Research at the UCL Institute for Innovation and Public Purpose (IIPP).
