Research on climate-related risks and financial stability: An "epistemological break"?

Based on remarks by Mr Luiz Awazu Pereira da Silva, Deputy General Manager of the BIS, at the Conference of the Central Banks and Supervisors Network for Greening the Financial System (NGFS), Paris, 17 April 2019.

BIS speech  | 
23 May 2019

A growing body of research and studies by academics, central banks, the NGFS and institutions such as the BIS focus on climate-related risks (CRRs). This work is helping to trace the links between the effects of climate change (CC), or global warming, and the stability of our financial sectors. The potential financial consequences of CRRs amount to a new form of systemic risk with implications for financial stability. Since financial stability has been explicitly or implicitly incorporated into the post-crisis mandate of many central banks, they have become more concerned with CRRs.

In addition, these studies are shifting the mindset of many private sector investors towards considering CC in their financial decisions. CC has therefore started to influence portfolio choices. Why is this? The main reason is that research reveals the lack of any proper valuation of CRR in current pricing practices. Indeed, if the uninsured losses caused by weather-related accidents are so large, and if there is a growing likelihood of stricter regulation against greenhouse emissions, then the relative value of "brown" versus "green" assets must be revisited. This is creating a new awareness that is beginning to produce a repricing of CRRs. That, in turn, is affecting resource allocation and tilting preferences towards lower-carbon projects and "greener" assets. All this awareness might therefore be acting, to some extent, as a shadow price for carbon emissions.

In a nutshell, what is happening with the repricing of climate-related risk in the financial sector is perhaps an illustration, as applied to financial risk management, of Bachelard's "epistemological break".1 Our financial risk models used to neglect the CC dimension of investments, assuming that the impact of CRR on profitability, returns and financial health would be negligible. Such views probably need to be revised in line with new evidence.

The epistemological break or paradigm shift is happening because CRR is changing the perception of the financial consequences of CC. Agents are increasingly aware that the CRR consequences they thought lay in the distant future are now much closer. CC poses, including for the financial sector, a clear and present danger. The ongoing repricing is changing the terms of the "tragedy of the horizon". The horizon has been brought forward. This is good news, but it also raises the possibility of a CC-related "Minsky moment" of financial fragility.

I will briefly examine first the growing volume of evidence that is helping to raise awareness regarding the financial effects of CC; then I will look at some ongoing research that is helping to change our minds by making it easier to price CRR more accurately. Finally, I will end these remarks by proposing a role for central banks, with implications for policy.

1.     Facts: mounting evidence of CRR as a clear and present danger

Climate-related risk can impact financial stability through three types of risk.2 First, physical risks such as meteorological events (storms and heavy rains), hydrological events (floods) and other climatological events (extreme temperature, drought, wildfire) are affecting the value of financial assets worldwide. An increasing number of reports available to investors and the general public offer improved techniques for pricing CRR financial risk.3 All this new information points to the need to reassess prices. Second, transition risks may result from the adjustment of asset prices towards a low-carbon economy. Transition risks are possible in an environment where greater disclosure of financial assets is required, and new CC regulation creates new obligations to move towards a lower-carbon economy. These changes, while desirable per se, could impact the asset price of very large sectors of the economy such as coal, oil and gas but also of companies that produce cars, ships and aircraft.4 All these assets could be subject to a change in investors' perception of profitability and business sustainability. If the changes are abrupt, an archetypal fire sale might result, potentially triggering a financial crisis. Finally, liability risks stem from the increased compensation paid to economic agents affected by climate change. There are reports from the insurance industry showing a growing level of insured and uninsured losses resulting from CRR. These losses impact the financial health of both the insurer and also the equity value of firms that are subject to these weather-related events.

These developments are contributing to increase the "awareness" of all agents in the economy. And this new mindset is indicating that climate-related risk is not necessarily a distant possibility in the future but rather a clear and present danger to financial stability since the assets at physical risk are large (for producers, for financial investors etc) and the costs of weather-related accidents are also very high and rising, for both insurance companies and their clients, and especially for uninsured parties.

2.     New research: making progress towards pricing CRR more accurately

Current macroeconomic research on CC is often characterised as having two streams. First, integrating climate change into macroeconomic models.5 As stated in the April 2019 NGFS Report, our macroeconomic models may not be able to accurately predict the economic and financial impact of climate change, but the best science today states that action to mitigate and adapt to climate change is needed now. That implies that the price of CRR from a macroeconomic perspective is most likely underestimated. Second, models to assess financial stability risks from climate change (eg stress-testing models using global warming scenarios). Both strands of the literature are well established, and central banks are actively contributing to it, in particular to the financial stability strand.

But more relevant to our discussion on the paradigm shift is the strand of research that is more finance-related, focusing on the risks and returns of financial instruments, including green finance instruments (eg green bonds), as well as the measurement of financial risks associated with climate change for particular industries and companies (as opposed to systemic risks). The risks and returns of financial assets can become new key determinants of how effectively climate change can be combated. If investors assess and price financial risks properly, then polluting assets will become more costly. In turn, more investments will flow into green assets, driving the transition to a low-carbon economy. There are some indications that these risks are not yet fully priced in, but research is starting to change that state of benign neglect.

Several new vectors of academic research show that investors are not pricing in the risk of droughts for agricultural companies.6 Another prime example of mispricing is stranded assets - typically defined as those covering fossil fuel extraction, which are very likely to lose value if carbon emissions are restricted. Researchers at the Smith School (Oxford University) have done pioneering work on stranded assets, showing that the likely losses on those asset values are substantial.7

Arguably, the prices of financial assets of fossil fuel-producing firms do not fully reflect their CRRs. Investors, at this stage, face a difficult task in assessing these risks - there is, for instance, no equivalent of credit ratings for climate-related financial risks, although this is changing rapidly.8 But the greater availability of data and increased awareness may be changing investors' attitudes towards climate-related financial risks. Future research has to go further and develop models and measures of CRRs that can be applied to individual assets. That would not only help investors to make better decisions, but also help society as a whole, as much needed investment would be shifted towards greener assets.

Research conducted at the BIS is also looking into the potential mispricing of climate change-related risks.9 The research examines if climate change risks, especially those related to climate policy risks, are priced in in the bank syndicated loan market by combining syndicated loan data with environmental exposure data. The key findings are: (i) the premia for CRRs, as measured by firm-specific CO2 emissions, have significantly risen since the Paris accord; and (ii) the rise in risk premia is especially due to increased awareness of transition risks. We also look at the environmental risk exposures of institutional investors.10 This research examines a global data set of individual fixed income holdings at institutional investor funds. Given their long investment horizon, such funds are naturally exposed to financial risks related to climate change. Various classifications and techniques are used to determine the risk exposures of individual fixed income assets and overall exposures of institutional investors in different countries, and the international spillover of revaluation effects is then analysed, using scenario analysis combined with network analysis, as these effects might arise due to common asset holdings and potential amplification through fire sales.

This new research is contributing to a change in investors' mindset - what I referred to above as an "epistemological break". Indeed, some of these changes are already tangible: insurance companies are reassessing their cost of insuring physical risk; rating agencies are repricing CRRs and reassessing the quality of credits; asset managers are becoming increasingly selective and inclined to start picking "green assets" for their portfolios; and pension funds are beginning to reassess their exposure to CRRs and "brown assets". In general, since COP21, CRR reporting requirements have been strengthened, especially for the financial sector, eg Article 173 of the French Energy Transition Law, and the Task Force on Climate-related Financial Disclosures (TCFD).11 Institutional investors are encouraged to explain whether and how their policies and targets align with national strategies for energy and ecological transition. For industry-led TCFD, recommendations require companies to disclose how their governance, strategy and risk management take CRR into account, and to conduct scenario analysis referring at least to the two-degree scenario. All these developments are reinforcing awareness, ie the perception is growing that CRRs are not yet properly priced and that some adjustment is taking place.

3.     The role of central banks and policy implications

There is a promising and noticeable change in mindsets and progress in repricing CRRs. But despite that, it might be necessary to accelerate the implementation of other public policies to combat CC because the effects of the ongoing changes that will control greenhouse gas (GHG) emissions and eventually reduce them might not be available in time to avoid irreversible damage. As we discussed above, the present and clear danger of a climate-related financial crisis is rising.

There is no need for central banks, regulators and supervisors to become the "only game in town" in such a complex topic as CRRs. But, equally, there is no need to dismiss their participation entirely, because this would not require their mission to be modified. Instead, they would be sticking to their existing mandates of price stability and financial stability. CBs need to be concerned about CRRs because of their financial stability implications. CRRs impinge on global financial stability as soon as they are seen as a problem of the commons. And CBs can be inspired by Elinor Ostrom's12 principles for governance of common pool resources (CPRs). For example: (i) clearly identify and quantify the risks to the common pool resource, ie financial stability; (ii) find actions that reduce CRRs at the global level and also at a decentralised (local) level; (iii) monitor these arrangements; and (iv) design and enforce rules for system stability, which implies coordination, local participation, and a sense of fairness in burden-sharing, incentives and penalties etc. The NGFS is contributing precisely in these three dimensions, having conducted an inventory of existing best practices and issued practical recommendations.13 The G20, international financial institutions (IFIs), central banks and financial sector regulators/supervisors around the world are already acting on CRRs in a variety of ways.

However, support and guidance from central banks, regulators and supervisors are also necessary, especially to help explore the many ways of financing the adaptation to a lower-carbon economy. Even if markets are already financing some of these costs, there is also a need for more public policy intervention to mitigate CRRs, through new research and also practical policy.

Developing "greener" financial investment instruments. The market for these instruments has boomed, but more progress in the taxonomy of what is "green" is necessary to avoid "green-washing" and excessive free-riding on the green label. Moreover, these instruments need to serve to finance technological innovation14 and not only investment projects that reduce the carbon footprint of firms and households. The market for "green bonds" is growing fast, including though the incorporation of environmental-social-governance (ESG) criteria into the management of CBs' own reserves and pension funds, and also with new tools that are capable of handling extreme climate-related events (eg catastrophe or CAT bonds).

Assessing the cost-benefit of more regulatory, direct interventions. In addition to progress on disclosure of CRRs, some regulators and supervisors are also exploring the pros and cons of more interventionist approaches: "green" relending facilities using adequate collateral; a subsidised administrative credit policy favouring "green" projects; and even ad hoc macroprudential measures, eg a "green" supporting risk factor or a "brown" penalising factor for corporate capital structures. The obvious issue here is to assess whether this proactive approach could create other distortions that could hamper the greening of the financial system and delay some of the initiatives described above.

Coordinating with other policies. Finally, CBs and supervisors cannot take on these new CRR-related challenges alone. They will need support from other policies conducted by other actors. However, we might currently be at a special juncture: (i) combating the effects of CC could be the best way, over the next few decades, to create the necessary new science-technology-engineering-maths or STEM jobs in new green industries, services and infrastructure that will compensate for the jobs that will most likely be significantly reduced by technological progress in the new digital economy; (ii) where fiscal space is available, financing the adaptation investment costs to CC with public debt could be politically less controversial than through carbon taxation. It could also be economically sustainable in the post-crisis low interest rate environment, as somehow suggested by Olivier Blanchard, Larry Summers and Brad DeLong.15

It is well known that the textbook solution to mitigate the effects of GHGs is a globally coordinated Pigovian tax, based on the adequate pricing of carbon through a carbon tax. However, progress in the appropriate repricing of CRRs could produce complementary effects by reallocating financing towards a low-carbon economy. Working on both the real and the financial side of our economies is complementary, and the epistemological changes that new research on CRRs is producing might just help to prove that.

1 Gaston Bachelard (1884-1962) was a French philosopher of science who proposed the concepts of epistemological obstacle and epistemological break (obstacle épistémologique and rupture épistémologique). For Bachelard, scientific knowledge is not a process of accumulation through continuous progress but rather a series of "epistemological breaks" - discontinuous jumps, with abrupt shifts in the meaning of concepts. Thomas Kuhn (1922-96), an American physicist, held a similar view, stating that the advancement of scientific knowledge occurred through periodic "paradigm shifts", in the manner of Copernicus's revolution, rather than through a continuous linear progress.

2 The taxonomy comes from M Carney, "Breaking the tragedy of the horizon - climate change and financial stability", speech given at Lloyd's of London, September 2015, . See also Bank of England, "Climate change: what are the risks to financial stability?", KnowledgeBank, November 2018, .

3 A recent example is BlackRock Investment Institute, "Getting physical: scenario analysis for assessing climate-related risks", Global Insights, April 2019, . The report combines BlackRock's asset-level expertise with the latest climate science and big-data capabilities, as well as recent advances in climate and data science, to generate some 160 terabytes of data - a granular picture of investment-relevant physical climate risks. It assesses direct physical risks to assets on a local level - today and under different future climate scenarios - and helps to assess whether the risks are adequately priced by markets. The report suggests that investors must rethink their assessment of vulnerabilities. Weather events such as hurricanes and wildfires are underpriced in financial assets.

4 If government policies were to change in order to abide by the Paris Agreement, "the vast majority of reserves (could be) 'stranded' - oil, gas and coal [-] will be literally unburnable without expensive carbon capture technology, which itself alters fossil fuel economics". See also Carney (2015), op cit, p 11.

5 Many different approaches exist to linking CC with its macroeconomic impact. Various discount rates have been adopted in comparing these future impacts, eg in the Stern Review and other macroeconomic modelling work (see N Stern, The Economics of Climate Change: The Stern Review, Cambridge University Press, 2007). Starting with his pioneering work in 1974, William Nordhaus, who won a Nobel Prize for integrating climate change into long-run macroeconomic analysis, has developed Integrated Assessment Models (IAMs) and Dynamic Integrated Climate Economy (DICE) models. (See W Nordhaus, Managing the Global Commons: The Economics of Climate Change, MIT Press, 1994, and "Integrated Economic and Climate Modelling", in Handbook of CGE Modelling, vol 1, Chapter 16, Elsevier, 2013.)

6 H Hong, F Li and J Xu, "Climate risks and market efficiency", Journal of Econometrics, forthcoming, .

7 B Caldecott, Stranded assets and the environment, risk, resilience and opportunity, Oxford Sustainable Finance Programme, Routledge, May 2018.

8 S&P Global Ratings, "COP24 Special Edition: Shining a light on climate finance", December 2018.

9 T Ehlers, K de Greiff and F Packer "Pricing of environmental risk in syndicated loans", presentation at the NGFS-Bundesbank conference in Berlin, October 2018, mimeo.

10 T Ehlers and V Stolbova, "Environmental risk exposure of institutional investors and international spillover effects" mimeo.

11 See Task Force on Climate-related Financial Disclosures (TCFD),

12 E Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action, Cambridge University Press, 1990.

13 In NGFS, "A call for action: Climate change as a source of financial risk", April 2019, the NGFS makes six recommendations: (i) integrating climate-related risks into financial stability monitoring and micro-supervision; (ii) integrating sustainability factors into own-portfolio management; (iii) bridging the data gaps; (iv) building awareness and intellectual capacity and encouraging technical assistance and knowledge sharing; (v) achieving robust and internationally consistent climate and environment-related disclosure; and (vi) supporting the development of a taxonomy of economic activities.

14 Technological innovation through R&D subsidies is also an indispensable tool in practice because of their greater political acceptance. In addition, see eg D Acemoğlu, P Aghion, L Bursztyn and D Hemous, "The environment and directed technical change", American Economic Review, vol 102, no 1, 2012, pp 131-66; these authors conclude that combining carbon taxation and research subsidies is superior to a policy based solely on a carbon tax.

15 O Blanchard, "Public debt and low interest rates", AEA Presidential Address, January 2019; and J Bradford DeLong and L Summers, "Fiscal policy in a depressed economy", Brookings Papers on Economic Activity, Spring 2012.