The effects of Russia’s war on climate and in delaying climate action are serious and present risks to global financial stability. Understanding the nature of climate and financial accelerators and their longer-term linkages should help policymakers address the new risks and challenges of the war.
Climate scientists fear catastrophic tipping points in the global climate (Lenton et al. 2019, IPCC 2021, 2022). The global climate accelerator describes the phenomenon whereby an accumulation of greenhouse gases, in raising global temperatures, in turn leads to the release of more carbon and even higher temperatures, ultimately making much of the planet uninhabitable. We face a climate crisis, as the world is dangerously close to the tipping points at which irreversible changes would occur. The global climate accelerator, and the financial accelerator (for instance, that operated in the Global Financial Crisis (GFC)), are both characterised by highly non-linear feedback loops. In the GFC, falling real estate prices were amplified in the financial system and by its interaction with the real economy, leading to further price collapses. Thus, the tipping points and cascades in the climate literature have parallels in financial crises like the GFC or the Great Depression, manifested in falling stock markets and bankruptcies of homebuilders, the foreclosures of many homes, and failures of banks and other financial institutions.
Also common to both is that special interests and lobbying corrupt and impair the functioning of financial regulation and climate regulation, helping spread damaging distortions of the facts. Common too is that the wealthiest countries have contributed most to causing the GFC and to global carbon emissions (Chancel and Piketty 2015).
There are also crucial differences. Even without policy intervention, there is a floor to real estate prices, as housing demand eventually responds to lower price levels. However, without policy intervention, there is no natural ceiling for global temperatures consistent with reasonable survival prospects for most of humanity. A second difference concerns different time scales – mere months for the financial accelerator, but decades and eventually millennia for the climate accelerator, given the persistence of the stock of greenhouse gases. With this time scale, myopia by policymakers on climate action is disastrous. A third difference is that, in the GFC, the financial accelerator’s effects were concentrated in affluent financialised countries. However, the climate accelerator has global implications, impacting most severely many poor countries in the medium term (Cruz and Rossi-Hansberg 2021).
It is by now well accepted that the linkages between the financial system and global economy, some highly non-linear and destabilising, were not well understood before and during the GFC. These non-linear processes still largely do not figure in current policy models, and this needs to be urgently addressed. Analogous issues apply to climate models designed by economists: they need similarly to incorporate financial and climate-related non-linearities. Even without detailed policy models, policymakers need a qualitative understanding of these dual non-linearities. This column draws out the similarities of such processes for the climate and financial accelerators.
The GFC and its aftermath deflected the focus on important mitigation policies, delaying necessary action on climate change. Pre-GFC, the IMF wrote: “Climate change is a potentially catastrophic global externality and one of the world’s greatest collective action problems…The costs of policies to address climate change can be contained by ensuring that mitigation policies are well designed. It will be crucial to aim at a framework that is sustainable and provides incentives for a broad country participation” (IMF 2008). However, fiscal capacity for ‘greening’ was severely curtailed. Bankers, bailed out at enormous cost, retained their pensions and often their bonuses. The mass of the population suffered high unemployment, loss of income, and, sometimes, their homes. The reputational loss to policymakers and politicians fuelled the rise of libertarian or nationalistic populism. Vital collective action for addressing climate problems was thereby constrained.
The global climate accelerator
The global climate system is subject to amplifying feedback loops. The rise in temperatures from the accumulation of greenhouse gases – through burning fossil fuels, and the release of methane incidental to extraction of oil and natural gas and via intensive livestock farming – itself causes further temperature rises and further carbon emissions. Table 1 elucidates the mechanisms that increase the probability of irreversible shifts in the climate when various tipping points are breached. Worse equilibrium states may result involving mass species extinction, major sea-level rises, and other disasters (Hendry 2014, Lenton et al. 2019, IPCC 2021, United Nations Environment Programme 2021). Table 1 explains there are few offsetting or stabilising processes on a relevant scale. Each IPCC report paints a more disturbing picture of how close we are to these dangers. Figure 1 illustrates the alarming scale of the man-made rises in the actual global average temperatures.
Table 1 Transmission and amplification in the global climate system
Figure 1 Rising global temperatures.
Source: IPCC (2021), Figure SPM.1.
The financial accelerator in the GFC
A recent survey on international house price cycles explains the complex system of interactions between the real economy and the financial sector, operating in both directions. Figure 2 illustrates the channels for a negative shock to US house prices during the sub-prime crisis. The shocks were greatly amplified by important non-linearities.
The left side of Figure 2 illustrates transmission of falling house prices to the real economy and back to house prices, while the right side shows transmission to the financial sector and back to house prices. The interactions between sectors are indicated. The four sideways transmission arrows from the bottom right-hand rectangle represent the impact of credit conditions on the real economy components of real estate demand, construction, and consumption, but also directly to GDP. Feedback channels from real components to house prices are shown by thin upward arrows. Amplification of house price shocks can occur both within the financial sector (through contagion) and between the financial sector and real economy.
Figure 2 The financial accelerator (example for the US sub-prime crisis)
Source: Duca et al. (2021).
The US crisis began with the unravelling of an overvalued real estate market, given fragile financial fundamentals. The system was over-leveraged with poor lending quality, partly due to two deregulation measures.2 Homebuyers extrapolated the observed strong house price gains, driven by deregulation and by positive economic news, so that house prices became increasingly overvalued just as households’ rising debt levels made them more vulnerable.
A full discussion of the channels, feedbacks and potential amplification can be found in Duca et al. (2021).3 A summary of the mechanisms in the financial accelerator in the US in the GFC is given in Table 2. This table also generalises from the US to other countries, indicating how differences in institutions and regulation between countries, and within a country over time, affect the transmission and the degree of amplification. There is considerable micro- and macro-evidence for the channels between consumption and house prices discussed in the table.4
Table 2 Transmission and amplification of a negative house price shock in the GFC
Weaknesses of economic policy models during and after the GFC
The models that guided much of macroeconomic policymaking and thinking before the GFC, namely the New Keynesian dynamic stochastic general equilibrium (NK-DSGE) models, failed utterly during the crisis. They were based on linear reasoning and the assumption of a stable long-run equilibrium. They ignored coordination failures, especially between the real economy and finance, and therefore proved inadequate for understanding financial stability. They were insufficiently ‘dynamic’ and misled on real-world lag structures. They were hardly ‘stochastic’, since they lacked, for probability distributions, both radical uncertainty (the time dimension) and heterogeneity (the cross-section dimension) (Muellbauer 2010, 2018). Yet, the potential for non-linearities and structural breaks makes radical uncertainty endemic. This undermines the relevance of key assumptions of NK-DSGE models – ‘rational’ expectations and inter-temporal optimisation which hold that all economic agents share knowledge of the same stable model of the economy (Hendry and Mizon 2014).
Post-GFC, central banks have placed more reliance on semi-structural models, such as the Federal Reserve’s FRB-US, based on less unrealistic assumptions and with greater scope for the data to ‘speak’. Nevertheless, the transmission channels for credit shifts and house prices remain poorly represented in the household sector of these models (Muellbauer 2020).7
Weaknesses in economic models which attempt to integrate climate
Oswald and Stern (2019) argue that, on climate change, “academic economists are letting down the world”,8 with most top journals ignoring or mis-representing the issue.
Franta (2021) analyses the damaging role of an influential group of economic consultants, hired by the fossil fuel industry from the 1990s to the 2010s, to estimate the costs of various proposed climate policies.9 He argues that their work “played a key role in undermining numerous major climate policy initiatives in the US over a span of decades, including carbon pricing and participation in international climate agreements”. This unfortunately influenced the conservative assumptions used in integrated assessment models (IAMs) that have attempted to integrate climate science with economic models for cost-benefit analysis and for computing the social cost of carbon (e.g. Nordhaus 1991, 1992). Two critical assumptions govern the output from IAM models: the social discount rate and the climate sensitivity of GDP (i.e. damage to GDP from rising temperature and associated planetary consequences) (Pindyck 2017). If a low discount rate of 3% is assumed and a linear relationship between temperature and GDP, then even a 5°C rise in global average temperatures will have only a moderate impact on GDP. The result is an estimate of the social cost of carbon of $11 (Nordhaus 2011). This low figure contrasts with an estimate of over $200 by Stern (2007). Stern (2013) emphasises that ignoring (or under-estimating) potentially catastrophic outcomes arising from the non-linear nature of the global climate accelerator is common to many IAMs. But inclusion is crucial to realistically estimating the social cost of carbon.
Other problems with IAMs mirror those of economic policy models. These include the omission of interaction effects10 and radical uncertainty, and the assumption of rational (model-consistent) expectations and representative agents when there is much heterogeneity in practice (Farmer et al. 2015, Hepburn and Farmer 2020, Asefi-Najafabady et al. 2021). Many IAMs neglect feedbacks between climate and human systems, except in trivial ways.11
Central banks came late to addressing the impact of climate change on financial stability. In 2017, the Network for Greening the Financial System was launched, and 108 central banks and regulators are now members. Financial stability will be impacted by both physical and transition risks to assets and property from climate change, and the income-reducing effect of disruptions to trade and production. These risks are assessed through climate stress tests, and adjustments include disclosure requirements and guidance, new risk weights and capital requirements, and the alignment of monetary policy instruments (e.g. adjusting ‘haircuts’ on collateral for climate risk) (Loyttyniemi 2021, Network for Greening the Financial System 2019, Bolton et al. 2020).12 Central banks have begun incorporating climate features into macroeconomic models (ECB 2021), and some of these model developments will prove relevant in current forecasting and policy simulation.
Implications of Russia’s war for financial stability and the climate
Central banks are now better placed to address the risks to financial stability generated by Russia’s war on Ukraine. Monitoring risks to financial stability and prudential regulation does not rely too heavily on the current generation of still inadequate policy models and has improved substantially since the GFC. Risks include a new sovereign debt crisis given the downgrade or default of Russia’s foreign-currency debt, much of it held in Europe. Spillover effects from falls in asset values of companies invested in Russian assets and from large global economic shocks from disrupted supply chains and commodity prices rises could be amplified in the financial system and be followed by a credit crunch.
The war’s effects on the Earth’s climate are potentially even more serious. The reconstruction of destroyed Ukrainian cities will have significant climate implications. Between 30% and 70% of lifetime carbon emissions of buildings come from the carbon embodied in the construction phase (e.g. OECD 2021, Arup 2022). In general, retrofitting existing buildings is a far better green strategy than demolition and reconstruction. Putin’s war machine is achieving the exact opposite, and at a staggering human cost. The dash to increase oil and gas production, with energy security likely to dominate climate concerns for years to come, could delay the greening of the global economy by years. While longer term, it will speed up green investments and the development and adoption of green technologies, the proximity of tipping points in the global climate system implies that any delay now is hazardous.
Author’s note: We are grateful to David Hendry, Cameron Hepburn, Francois Lafond, Avner Offer, Ryan Raferty and Dennis Snower for their input, but take responsibility for errors.
Adrian, T, D M Covitz, and N Liang (2015), “Financial stability monitoring”, Annual Review of Financial Economics 7: 357-395.
Andersen, A L, C Duus and T L Jensen (2016), “Household debt and spending during the financial crisis: Evidence from Danish micro data”, European Economic Review 89: 96-115.
Aron, J, J V Duca, J Muellbauer, A Murphy, and K Murata (2012), ““Credit, housing collateral and consumption: Evidence from the U.K., Japan and the U.S", Review of Income and Wealth 58 (3): 397-423.
Arup (2022), “How you can reduce your carbon emissions”, online.
Asefi-Najafabady, S, L Villegas-Ortiz, and J Morgan (2021), “The failure of Integrated Assessment Models as a response to ‘climate emergency’ and ecological breakdown: The emperor has no clothes”, Globalizations 18(7): 1178-1188.
Berger, D, V Guerrieri, G Lorenzoni, and J Vavra (2018), “House prices and consumer spending”, Review of Economic Studies 85(3): 1502–42.
Bernanke, B (2018), “The real effects of disrupted credit: Evidence from the global financial crisis”, Brookings Papers on Economic Activity, September.
Bolton, P. D Morgan, L A Pereira da Silva, F Samama, and R Svartzman (2020), The green swan: Central banking and financial stability in the age of climate change, Basel, Switzerland: Bank for International Settlements.
Browning, M, M Gortz, and S Leth-Petersen( 2013), “Housing wealth and consumption: A micro panel study”, The Economic Journal 123: 401–28.
Brunnermeier, M K (2009), "Deciphering the liquidity and credit crunch 2007-2008", Journal of Economic Perspectives 23 (1): 77-100.
Calza, A, T Monacelli, and L Stracca (2013), “Housing finance and monetary policy”, Journal of the European Economic Association 11(suppl_1, 1): 101–22.
Castle J L and D F Hendry (2020), "Climate econometrics: An overview", Foundations and Trends in Econometrics 10 (3-4): 145-322.
Chancel, L and T Piketty (2015), “Carbon and inequality: From Kyoto to Paris”, VoxEU.org, 01 December.
Chauvin, V and J Muellbauer (2018), “Consumption, household portfolios and the housing market in France”, Economie et Statistique/Economics and Statistics 500-501-502: 157-178.
Cruz, J-L and E Rossi-Hansberg (2021), “Unequal gains: Assessing the aggregate and spatial economic impact of global warming”, VoxEU.org, 2 March.
Duca, J V, J Muellbauer, and A Murphy (2021), “What drives house price cycles? International experience and policy issues”, Journal of Economic Literature 59(3): 773-864.
ECB (2021), “Climate change and monetary policy in the euro area”, ECB Strategic Review, Occasional paper series 271, September.
Farmer, J, C Hepburn, P Mealy, and A Teytelboym (2015), “A third wave in the economics of climate change”, Environmental & Resource Economics 62 (2): 329-357.
Franta, B (2021), “Weaponizing economics: Big Oil, economic consultants, and climate policy delay”, Environmental Politics, DOI: 10.1080/09644016.2021.1947636
Geiger, F, J Muellbauer, and M Rupprecht (2016), “The housing market, household portfolios and the German consumer”, European Central Bank Working Paper Series 1904.
Hendry, D F (2014), “Climate change: Lessons for our future from the distant past”, VoxEU.org, 27 October.
Hendry, D F and G Mizon (2014), “Why standard macro-models fail in crises”, VoxEU.org, 18 June.
Hepburn, C and J D Farmer (2020), “Less precision, more truth: uncertainty in climate economics and macroprudential policy”, in G Chichilnisky and A Rezai (eds), Handbook on the Economics of Climate Change, 420-438, Cheltenham: Edward Elgar Publishing.
Hurst, E and F Stafford (2004), “Home is where the equity is: Mortgage refinancing and household consumption”, Journal of Money, Credit and Banking 36(6): 985–1014.
IMF (2008), Housing and the business cycle, World Economic Outlook, April.
IPCC (2021), “Crushing climate impacts to hit sooner than feared”, AR6 Climate Change 2021: The Physical Science Basis, UN: Intergovernmental Panel on Climate Change.
IPCC (2022), “Climate change: A threat to human wellbeing and health of the planet”, United Nations: Intergovernmental Panel on Climate Change.
Lenton, T M, J Rockstrom, O Gaffney, S Rahmstorf, K Richardson, W Steffen, and H J Schellnuber (2020), “Climate tipping points too risky to bet against”, Nature 575(7784): 592–595.
Loyttyniemi, T (2021), “Integrating climate change into the financial stability framework”, VoxEU.org, 8 July.
Mian, A and A Sufi (2014), ‘”House price gains and U.S. household spending from 2002 to 2006”, NBER Working Paper No. 20152.
Mian, A and A Sufi (2018), “Finance and business cycles: The credit driven household demand channel,” Journal of Economic Perspectives 32(3): 31–58.
Mishkin, F (2007), “Housing and the monetary transmission mechanism”, in Housing, Housing Finance, and Monetary Policy, Proceedings of the Economic Policy Symposium, Federal Reserve Bank of Kansas City, Jackson Hole, Wyoming, 359-413.
Muellbauer, J (2010), “Household decisions, credit markets and the macroeconomy: implications for the design of central bank models”, Bank for International Settlements, discussion paper 306, March.
Muellbauer, J (2018), “The future of macroeconomics”, The future of central banking: Festschrift in honour of Vítor Constâncio, ECB, December, 6-46.
Muellbauer, J (2020), "Implications of household-level evidence for policy models: The case of macro-financial linkages", Oxford Review of Economic Policy 36(3): 510-555.
Network for Greening the Financial System (2019), “First comprehensive report: A call for action climate change as a source of financial risk”, NGFS: Banque de France.
Nordhaus, W (1991), “To slow or not to slow: The economics of the greenhouse effect”, Economic Journal 101(407): 920–937.
Nordhaus, W (1992), “An optimal transition path for controlling greenhouse gases”, Science 258(5086): 1315–1319.
Nordhaus, W (2011), “Estimates of the social cost of carbon: Background and results from the RICE-2011 model”, NBER Working Paper 17540.
OECD (2021), Brick by brick: Building better housing policies, Paris: OECD.
Oswald, A and N Stern (2019), “Why are economists letting down the world on climate change”, VoxEU.org, 17 September.
Pindyck, R (2017) “The use and misuse of models for climate policy”, Review of Environmental Economics and Policy 11(1): 100114.
Stavins, R (2019), “A gift that keeps on giving: The contributions of Martin Weitzman to environmental economics”, VoxEU.org, 9 September.
Stern, N (2007), The economics of climate change: The Stern review, Cambridge: Cambridge University Press.
Stern, N (2013), “The structure of economic modelling of the potential impacts of climate change: Grafting gross underestimation of risk onto already narrow science models”, Journal of Economic Literature 51(3): 838–59.
Stout, L A (2011), "Derivatives and the legal origin of the 2008 credit crisis", Harvard Business Law Review 1(1): 1–38.
United Nations Environment Programme (2021), Emissions gap report 2021: The heat is on – A world of climate promises not yet delivered, Nairobi.
Vaks, A, A J Mason, S F M Breitenbach, A M Kononov, A V Osinzev, M Rosensaft, A Borshevsky, O S Gutareva, and G M Henderson (2020), “Palaeoclimate evidence of vulnerable permafrost during times of low sea ice”, Nature 577: 221–225.
Windsor, C, J P Jääskelä, and R Finlay (2015), “Housing wealth effects: Evidence from an Australian panel”, Economica 82: 552-577.
Wunderling, N, J F Donges, J Kurths, and R Winkelmann (2021), “Interacting tipping elements increase risk of climate domino effects under global warming”, Earth System Dynamics 12: 601–619.
1 However, Wunderling et al. did not include melting permafrost (Vaks et al. 2020) among their potential interactions, so under-estimate risks.
2 The Commodity Futures Modernization Act (2000) prioritised financial derivatives ahead of other debt claims, made enforceable throughout the US. Thus, private issuers of mortgage-backed securities were encouraged to use derivatives, supposedly to mitigate risk (Stout 2011). The spread of these derivatives in deepening secondary markets created the illusion that poor quality mortgage-backed loans were of investment grade. The other deregulation measure was a Securities and Exchange Commission decision, in 2004, to ease capital requirements on investment banks, which caused a dangerous rise in gearing levels.
3 In addition to contagion within the financial system, a second important non-linearity occurred because a given house price fall results in a far larger increase in bad loans and hence credit contraction, if it follows a preceding fall. A third large nonlinearity stemmed from the amplification, by tighter credit conditions, of the decline in housing collateral for consumer spending. A fourth nonlinearity can occur through escalating risk premia in user cost (which is cost of credit less expected appreciation), with disproportionate effects on the demand for housing and hence on house prices.
4 See Muellbauer (2010), Aron et al. (2012) and Calza et al. (2013) for macro evidence. Micro-research on the so-called housing wealth effect on consumption finds that this is much more of a collateral effect than a classical wealth effect; see Hurst and Stafford (2004), Mian et al. (2018), Browning et al. (2013), Windsor et al. (2015), and Andersen et al. (2016). Berger et al. (2018) provide the theory justification for concluding that the effect of housing wealth on consumption is conditional on ease of access to credit and hence varies over time and between countries. In the absence of equity withdrawal and with cautious lending practices, higher house prices tend to reduce aggregate consumption – for example, Japan (Aron et al. 2012), Germany (Geiger et al. 2016), France (Chauvin and Muellbauer 2018).
5 The corollary, relevant in 2022, is that when policy rates rise, the impact is faster in floating rate environments.
6 Adrian et al. (2015) define systemic risk as “the potential for widespread financial externalities—whether from corrections in asset valuations, asset fire sales, or other forms of contagion—to amplify financial shocks and in extreme cases disrupt financial intermediation”.
7 For example, assuming that all assets and debt can be lumped into one aggregate as the only way asset prices, liquidity and credit shocks affect consumption, given income, ignores how shifts in credit constraints alter behaviour, and so misses the ‘credit-driven household demand channel’, see (Mian and Sufi 2018), with its nonlinearities. As a result, FRB-US failed the acid test in 2007 of simulating the consequences of a fall in US house prices (Mishkin 2007). Moreover, FRB-US does not model banks’ balance sheets, capital adequacy and other regulatory ratios, and non-performing loans (NPLs) or loan-loss provisioning, to capture the two-way connection between credit conditions and NPLs. It therefore misses key parts of financial/real economy transmission mechanisms.
8 Exceptions to this generalisation include Stern (2007), Hendry (2014), Farmer et al. (2015) and the work of Weitzman; see Stavins (2019).
9 Most prominent was Charles Rivers Associates, but Franta also mentions petroleum industry-funded reports from DRI-MacGraw Hill, Wharton Econometric Forecasting Associates and MIT’s Joint Program on the Science and Policy of Global Change.
10 Castle and Hendry (2020), in modelling 800,000 years of ice core data on ice volume, CO2 and temperature, use nonlinear interactions effects, controlling for variations in the earth’s orbit around the sun and for outliers, e.g. due to meteor strikes and volcanic eruptions.
11 Hepburn and Farmer (2020) draw parallels between lessons for policy from complex systems approaches to the climate system and the financial system, emphasising the precautionary principle rather than efficiency. Farmer et al. (2015) argue for the use the agent-based approach to model human-climate interactions.
12 Major risks arising via real estate are summarised in Duca et al. (2021), p.781-782.