Politicians have to take action globally and put in place the framework conditions to mitigate anthropogenic climate change. The switch from fossil to climate-neutral technologies that this requires will give rise to additional costs for households and firms. Since both have ties with the financial market, the costs of transitioning to a net zero economy also have implications for the risk situation in the German financial system.

The 2021 and 2023 Financial Stability Reviews looked at the long-term impact of full decarbonisation on financial stability. 1 This report augments the analysis by examining the short-term impact of an unexpected, immediate carbon price increase on the economy and financial markets. This expanded analysis thus makes a contribution to the discussion of short-term risk scenarios that is currently ongoing in the international arena. It also shows how disclosure of climate-related information by firms, especially of greenhouse gas emissions, could mitigate the impact of this kind of shock.

The results demonstrate that unexpected, immediate carbon price increases can drive up risk in the German financial system. In the scenario under analysis, the impact looks to be manageable for Germany’s financial system. A consistent disclosure requirement helps to mitigate the impact both generally and in the face of unanticipated events.

The stress test simulations confirm that a credible and predictable stance on climate policy is advantageous. Disclosure requirements for firms and macroprudential instruments can also play a material role in the targeted identification and management of climate risk in the financial sector.

The carbon price increase considered in the analysis is unexpected and immediate compared with a gradual carbon price increase in a stable framework for an orderly transition to net zero, and leads to greater risks to the financial system. An unexpected and immediate increase in the carbon price gives firms and households little time to gradually adjust their emissions. This drives up the costs of transitioning to net zero. 2 Stress tests are used to examine how this would impact on the financial system. One key finding is that requiring firms to disclose emission intensities can enhance financial market efficiency and reduce the negative impact of an unexpected increase in the carbon price.

The analysis exploits a dynamic macro model to examine the short-term impact of an unexpected and immediate carbon price increase on the economy and financial system. In a first step, a dynamic macro model which models macroeconomic relationships is used to determine how an unexpected and immediate carbon price increase in Europe impacts on the economy and financial system. 3 This is benchmarked against a scenario in which the current carbon price is maintained. The effects in Germany and Europe are assumed to be comparable. In the second step, these results are used to analyse the impact on the German financial system and the risks to German financial intermediaries. The analysis builds on earlier work concerning carbon price sensitivities. 4 While those analyses focused on medium to long-term impacts, this one is particularly concerned with short-term effects occurring over the course of one year. 5 It thus examines short-term interactions in the overall economy and financial system that previously were not visible (for an overview of current analytical methods, see the supplementary information entitled “The Bundesbank’s tools for sustainability-related financial stability analysis”).

In addition, the macroeconomic impact of a disclosure requirement for firms is examined and evaluated in the context of the stress tests. If firms are required to disclose their carbon emissions, investors can factor in the impact of future climate policy and the aggregate reduction in carbon emissions on the firms they are considering financing. One outcome of full transparency is better functioning financial markets. The analysis focuses on how strong the positive effects of disclosure, as opposed to imperfect information, are for the overall economy.

The main scenario shows the risks of an unexpected carbon price increase. Putting a price tag on emissions of greenhouse gases such as CO₂ is considered an efficient instrument for reducing emissions. 6 This is why the analysis also looks at a carbon price increase, specifically of €100/tCO₂ within one quarter for the main scenario. The increase starts from the current carbon price of around €75/tCO₂ (see Chart 5.2.1). 7 A baseline with no further carbon price increase is used as a benchmark. The effects described below should therefore be understood as deviations in the main scenario from this baseline pathway. 8 The assumption of the carbon price increase lines up with international studies and is based on historical market developments. 9 Throughout 2021, the price of emission allowances in the EU Emissions Trading System (ETS) rose by around €70/tCO₂. 10 Similar and larger carbon price rises are also possible in the future. 11 This analysis assumes that the carbon price increase will be higher than observed in the past, in keeping with stress test logic (see Chart 5.2.1). Seeing as the EU ETS will be expanded to include buildings, road transport and additional sectors (industry not yet covered by the existing EU ETS) in future, the carbon price increase in the model covers the entire economy. 12 In the model calculations discussed below, this leads to a 25 % reduction in net emissions in Europe by 2030. Unlike other studies, this analysis does not look at the further trajectory of the carbon price, but leaves it unchanged after the increase. The change in the carbon price is thus a one-off, sudden increase.

Compared with an alternative “status quo” scenario, an unexpected and immediate carbon price increase gives rise to additional costs due to risks materialising in the short term. Very short-term adjustments in production and consumption in the wake of a carbon price increase result in additional costs especially for energy-intensive firms and consumers. This moment of surprise and the shorter horizon differentiate the main scenario from stress scenarios that examine the risks of long-term, continuous green structural change.

The costs and risks of the carbon price increase are independent of the assumed long-term transformation in the scenario. To assess risks to financial stability, a risk scenario (scenario of an unexpected, significant carbon price increase) is compared with a baseline pathway. The latter describes developments that are currently expected. The baseline pathway therefore simply extrapolates the status quo over the modelling horizon. It is also possible, however, to assume other future economic and climate policy trajectories, as the additional costs entailed by switching from the baseline to the carbon price increase scenario are not contingent on the selection of the baseline pathway itself. 13

Two alternative scenarios show how a gradual and partly anticipated increase in the carbon price as well as a disclosure requirement can reduce the risks posed by an unexpected carbon price increase. In the first alternative scenario, the price increase is spread across two quarters, at €50/tCO₂ each time, with the gradual raising of prices being announced at the time of the first increase, meaning that the second one no longer comes as a surprise. This phased increase gives market players more time to adjust. In the second alternative scenario, the carbon price is increased as in the main scenario; the difference is that a disclosure requirement allows market participants to assign the risks stemming from carbon emissions more accurately to the emitters. This results in better financing decisions and more efficient allocation of capital.

The macroeconomic impact of the increase in the carbon price is examined using a dynamic equilibrium model. To analyse the effects on the financial system, the carbon price increase is first fed into a macroeconomic model. This simulates how the increase affects macroeconomic variables that are important for the financial market. The study exploits an environmental dynamic stochastic general equilibrium (E-DSGE) model. 14 An innovation in climate risk analysis is the endogenous inclusion of shocks to financing conditions. In addition, this allows the financing of energy-intensive and non-energy-intensive sectors to be mapped. It is assumed that participants in the economic and financial system do not expect the carbon price increase assumed here. This approach allows us to simulate the largest possible losses under the given model assumptions. 15 Only once the price increase occurs does the future carbon price pathway become known to all model agents. Transparency about carbon emitters is assumed to be limited in the main scenario of a carbon price increase. In the baseline, the degree of disclosure – that is, the percentage of emissions which investors can accurately assign to emitters – is set at 80 %. 16

Market players are assumed to have very limited options for short-term adjustments. Given a carbon price increase at short notice, this can lead to significant additional costs for the economy as a whole. The exclusion of technological change or reinvestment of recycled carbon tax revenue plus the rising financing costs caused by interactions with the financial system increase the short-term risks.

Different sectors of the economy are affected to varying degrees by the carbon price increase. The results of the E-DSGE model are therefore fitted to German economic sectors. 17 It is assumed that the effects in Germany broadly match the results of the European E-DSGE model. 18 First, a sectoral equilibrium model is used to determine how much the individual sectors’ value added declines due to the carbon price increase. This makes it possible to allocate the overall effects on value added calculated in the E-DSGE model to the individual economic sectors. 19

In subsequent steps, financial market and credit risk models are used to calculate write-downs and losses on shares and corporate as well as government bonds in order to analyse the impact on the financial system. Based on historical data, relationships between credit default rates and firm metrics in the respective scenario are estimated. Projected firm metrics are then used to calculate firm-specific credit default rates in each scenario. In the same way, sectoral losses on shares are linked to the respective climate scenario variables. Bond prices change depending on the interest rate developments and yield spreads assumed in each scenario. Yield spreads of financial and non-financial bonds are estimated based on developments in sectoral variables, interest rates, and bond-specific metrics such as ratings and terms. 20 Given that government bonds are assumed to be risk-free, only movements in interest rates affect their prices here.

The increased carbon price leads to lower economic output and lower enterprise values relative to the baseline. The focus here is on the impact on Germany’s real economy and financial sector. The impact of the carbon price increase in Europe is examined to take into account strong trade links. 21 GDP falls by around 1.4 % within the first four quarters. Enterprise values drop by about 28 % in the first quarter (see Chart 5.3.1). These strong decreases are attributable, in large part, to interactions with the financial system and their impact on the real economy. 22 Since there are no further increases in the carbon price and new investment takes into account the adjusted carbon price pathway, enterprise values recover very swiftly, however. Next, the sector model is used to allocate GDP and enterprise values to the different economic sectors. Sector-specific developments are thus made ready for further use in the financial market and credit risk models (see section 5.5 “Anticipated, credible climate policy course and market transparency yield benefits”). 23 The initial sharp rise in inflation by around 0.5 percentage point compared with the baseline rapidly subsides again (see Chart 5.3.1). Seeing as there are no further carbon price increases and associated spikes in inflation, the risk-free interest rate only rises temporarily. 24 In the model, however, the long-term risk-free interest rate drops immediately due to the economic downturn.

A first alternative scenario assumes a gradual and thus partly anticipated carbon price increase. Now, instead of a one-off unexpected carbon price increase, the rise is partially anticipated. The macroeconomic effects of the immediate carbon price increase of €100/tCO₂ are compared against the alternative scenario in which it is known that the carbon price is going to be rising. Both scenarios assume imperfect disclosure. The first alternative scenario supposes that the carbon price goes up unexpectedly by just €50 in the first quarter. The price does not rise again until the second quarter, when it goes up by a further €50 to €175, the final carbon price level envisaged in the risk scenario outlined above. This second increase is already expected in the first quarter.

The fact that the carbon price increase is partly foreseeable significantly reduces the losses incurred by the economy as a whole. Chart 5.3.2 shows that the costs to the real economy are considerably lower: GDP and enterprise values decline less sharply, and inflation and interest rate levels do not climb by as much. The lower costs also benefit households and firms.

Standards for the disclosure of sustainability-related information are increasingly being put in place, and policymakers are working on setting out reporting requirements. Mandatory disclosure is currently the subject of public discussion. Various policy initiatives have already made for improved disclosure by firms. Most notably, last year saw the International Sustainability Standards Board (ISSB) publish IFRS Sustainability Disclosure Standards. 25 In Europe, more sustainability information, particularly with regard to large firms, is set to become available in the coming years due to the Corporate Sustainability Reporting Directive (CSRD) and the EU Taxonomy Regulation. The Basel Committee on Banking Supervision (BCBS) is currently discussing a Pillar 3 disclosure framework for climate-related financial risks. In Europe, disclosure obligations around sustainability risks also form part of the banking package (Capital Requirements Directive VI and Capital Requirements Regulation III (CRR III)). 26

The disclosure of information allows market participants to better adapt. It enables them to accurately attribute the carbon emissions generated by production and consumption. This helps households, who operate as investors in the model, gain a better understanding of the actual costs of the carbon price increase. That, in turn, reduces the macroeconomic costs and lessens the risks to financial stability.

Better attribution of emissions to their source can mitigate an economic downturn. In the main scenario described above, disclosure is assumed to be imperfect, with households wrongly attributing 20 % of emissions to zero-emission sectors. 27 The scenario presented in the following assumes full mandatory disclosure, which applies from the outset and allows emissions to be attributed with precision. In addition to the €100/tCO₂ carbon price increase, the degree of disclosure is also upped from 80 % to 100 %. Enhanced disclosure has two effects:

Enhanced disclosure significantly reduces the costs of the carbon price increase (see Chart 5.3.3). 28 Enterprise values drop by roughly 7 %, around three-quarters less than in the main scenario. GDP shrinks by around one-quarter less than in the main scenario, falling by slightly more than 1.1 %. The increases in inflation and the risk-free interest rate are each cut by around one-fifth. These effects place the economy on a better trajectory and indirectly bolster financial stability. 29 Mandatory disclosure does, however, come at a cost which may lessen the estimated benefits and even exceed them. Besides direct administrative costs, these also include competitive disadvantages created by information sharing. 30 Information relevant to the assessment of risk does not necessarily have to be disclosed to the public at large. Even just giving banks and insurers access to this information in the lending process is enough to improve risk assessment in financial markets and thus financial stability (see the supplementary information entitled “Prudential plans, transition pathways and scenario calculations”).

The German financial system remains resilient when exposed to the risks of a sharp increase in the price of carbon. The national impact of the carbon price increase as derived from the E-DSGE model – the declines in GDP and enterprise values, for instance – are first transposed to sectors by means of the sector model. These sector-specific macro-financial variables are then fed through market risk models to derive financial market prices and changes in the value of financial market portfolios. 31 The unexpected and immediate carbon price increase triggers significant but temporary losses where shares are concerned, while bonds are less heavily impacted. The immediate decrease in the long-term risk-free interest rate tempers the declines in corporate bond prices and even puts government bond prices on an upward trajectory. The analysis covers a horizon of just one year; carbon prices remain at the same level after the initial jump, and the road ahead for climate policy is uncertain. By the end of the analysis horizon, the carbon price increase brings down emissions by approximately 22 %. Effects on the present value of securities stemming from the evolution of emission prices following the initial carbon price increase are not examined here.

Losses experienced by banks are moderate. In the first quarter, banking book securities post losses of 1.9 %, compared with recognisable losses of just 1.1 % (see Chart 5.4.1). 32 The limited magnitude of the effect is down to two factors. First, German banks hold comparatively small amounts of shares and non-financial bonds. Second, with the increase in the risk-free interest rate being only temporary, lower-risk bonds – which make up a greater share of the portfolios – only see a small drop in value. Potential credit losses in the stressed corporate lending portfolio after a year are minimal, coming in at 0.03 %. Default risks amongst enterprises disadvantaged by the transformation show a slower response than market prices because they adapt sluggishly to the economic changes.

In the insurance and fund sector, the unexpected and sharp increase in the carbon price results in significant losses. Overall, German insurers’ securities portfolios lose 4.0 % of their value, though they do recover quickly. It is a similar story for German funds, whose securities portfolios shed 4.2 % of their value in the space of one quarter. Their net asset value tumbles by 3.9 %.

Overall, the losses incurred in the stress test scenario as a result of an unexpected carbon price increase are of a manageable magnitude for the financial system. The banking system sustains losses amounting to a maximum of 2.2 % of common equity tier 1 (CET1) capital. In the fund sector, the sizeable mark-to-market losses could prompt fund share redemptions, potentially triggering knock-on effects resulting from fire sales in financial markets. Insurers, meanwhile, experience mark-to-market losses amounting to 7.5 % of regulatory own funds in the second quarter of 2024. 33

The risks associated with full decarbonisation are not examined in this analysis. The stress test simulations focus on short-term responses by economic variables such as inflation, the risk-free interest rate and frictions in the financial system. These variables and the interplay between them are often neglected in long-term climate models in the interests of casting a detailed depiction of the transformation in full. The extent to which the unexpected carbon price increase lessens long-term risks up to 2050 is also left unaddressed. The carbon price increase brings the reduction in carbon emissions forwards in time, which reduces risks and vulnerabilities previously estimated on the basis of long-term scenarios. A full analysis of all risks is not possible at this time and requires a comprehensive analytical framework, which is currently under development within the NGFS. 34

Risks from the scenario positing a sudden carbon price increase are likely to be manageable for the German financial system. Losses in Germany’s financial system are moderate and bearable. Banks’ excess capital and insurers’ solvency are both high enough that they can keep meeting the regulatory capital requirements.

The stress test analyses show the importance of a credible and predictable climate policy pathway, which revolved around the carbon price in this particular instance. Unexpected shifts in the climate policy trajectory drive up the aggregate costs and could delay the transition to climate neutrality. The short-term impacts of an unexpected and immediate carbon price increase on the aggregate economy and financial system are broadly independent of the long-term risks associated with the transition to climate neutrality which were explored in previous reports. 35 Immediate and unanticipated adjustments in the expectations of financial market participants ought to be avoided.

Mandatory full disclosure is another important instrument that policymakers can use to lessen the costs of transition. Imposing the fullest possible disclosure requirement on non-financial corporations and – within an appropriate framework – financial intermediaries and including, for example, the costs of red tape can reduce potential burdens for financial market players.

Financial intermediaries should keep an eye on the risks of higher greenhouse gas emission prices for firms and, where necessary, work to strengthen their resilience. Emission price rises that come as a surprise can act as a further drag on the real economy and financial sector, adding to the burdens already stemming from the planned, long-term structural change.

Macroprudential supervisors in Europe are looking into which instruments could be used to specifically counter potential climate-related risks in the financial sector. 36 Macroprudential instruments do already exist to curb climate-induced systemic risks, but the general resilience of the financial system could be bolstered by systemic risk buffers. However, this necessarily involves assessing climate risk to the financial system from a systemic viewpoint. Sector-specific capital buffers, by contrast, could capture climate-induced systemic risks with even greater precision. Operationalising this kind of systemic assessment could come at a greater cost to financial intermediaries and supervisory authorities, however.

Agora Energiewende and Agora Verkehrswende (2023), Der CO₂-Preis für Gebäude und Verkehr: Ein Konzept für den Übergang vom nationalen zum EU-Emissionshandel.

Allen, T., M. Boullot, S. Dées, A. de Gaye, N. Lisack, C. Thubin and O. Wegner (2023), Using Short-Term Scenarios to Assess the Macroeconomic Impacts of Climate Transition, Banque de France Working Paper No 922.

Anderson, A. and D. T. Robinson (2022), Financial Literacy in the Age of Green Investment, Review of Finance, Vol. 26, No 6, pp. 1551‑1584.

Bachmann, R. and C. Bayer (2023), Respekt vor unterschiedlichen Ausgangsbedingungen: Horizontale Fairness in die CO₂-Bepreisung bringen, ECONtribute Policy Briefs, No 54.

Berger, P. G., J. H. Choi and S. Tomar (2024), Breaking it Down: Economic Consequences of Disaggregated Cost Disclosures, Management Science, Vol. 70, No 3, pp. 1374‑1393.

Carbone, S., M. Giuzio, S. Kapadia, J. S. Krämer, K. Nyholm and K. Vozian (2022), The Low-Carbon Transition, Climate Commitments and Firm Credit Risk, Sveriges Riksbank Working Paper Series, No 409.

Cucinelli, D. and M. G. Soana (2023), Investor preferences, financial literacy and intermediary choice towards sustainability, Research in International Business and Finance, Vol. 66.

Degryse, H., A. Di Giuli, N. Sekerci and F. Stradi (2023), Sustainable Investments: One for the Money, Two for the Show.

Deutsche Bundesbank (2024), Project Gaia enables climate risk analysis using artificial intelligence, press release of 19 March 2024.

Deutsche Bundesbank (2023), Financial Stability Review.

Deutsche Bundesbank (2022), Monthly Report, January 2022.

Deutsche Bundesbank (2021), Financial Stability Review.

Diluiso, F., B. Annicchiarico, M. Kalkuhl and J. C. Minx (2021), Climate Actions and Macro-financial Stability: The Role of Central Banks, Journal of Environmental Economics and Management, Vol. 110.

Ernst, A., N. Hinterlang, A. Mahle and N. Stähler (2023), Carbon Pricing, Border Adjustment and Climate Clubs: Options for International Cooperation, Journal of International Economics, Vol. 144, p. 103772.

European Central Bank and European Systemic Risk Board (2023), Towards macroprudential frameworks for managing climate risk, European Central Bank; European Systemic Risk Board.

European Commission (2024), Prudential requirements.

European Commission (2023), Sustainable Finance: Commission takes further steps to boost investment for a sustainable future, press release of 13 June 2023.

European Parliament (2024), Amendments to the Solvency II Directive, April 2024.

European Systemic Risk Board (2022), The Macroprudential Challenge of Climate Change: ECB/ESRB Project Team on Climate Risk Monitoring.

Falter, A., M. Kleemann, L. Strobel and H. Wilke (2021), Stress Testing Market Risk for German Financial Intermediaries, Deutsche Bundesbank Technical Paper No 11/2021.

Farvaque, E., C. Refait-Alexandre and D. Saïdane (2011), Corporate disclosure: a review of its (direct and indirect) benefits and costs, International Economics, Vol. 128, pp. 5‑31.

Filippini, M., M. Leippold and T. Wekhof (2024), Sustainable finance literacy and the determinants of sustainable investing, Journal of Banking & Finance, Vol. 163.

Financial Stability Board (2023), FSB Roadmap for Addressing Financial Risks from Climate Change: 2023 Progress Report, Financial Stability Board.

Frankovic, I. (2024), How do carbon prices spill over along global supply chains? The impact on Europe and Germany, Journal of Policy Modeling, Vol. 46, No 5, pp. 887‑907.

Frankovic, I. and B. Kolb (2024), The role of emission disclosure for the low-carbon transition, European Economic Review, Vol. 167.

Frankovic, I., T. Etzel, A. Falter, C. Gross, J. Ohls, L. Strobel and H. Wilke (2023), Climate transition risk stress test for the German financial system, Deutsche Bundesbank Technical Paper No 04/2023.

Gutsche, G. and B. Zwergel (2020), Investment Barriers and Labeling Schemes for Socially Responsible Investments, Schmalenbach Business Review, Vol. 72, pp. 111‑157.

Hinterlang, N., A. Martin, O. Röhe, N. Stähler and J. Strobel (2023), The Environmental Multi-Sector DSGE model EMuSe: A technical documentation, Deutsche Bundesbank Technical Paper No 03/2023.

Kaldorf, M. and M. Rottner (2024), Climate Minsky moments and endogenous financial crises, Deutsche Bundesbank Discussion Paper No 26/2024.

Kalkuhl, M., M. Kellner, T. Bergmann and K. Rütten (2023), CO₂-Bepreisung zur Erreichung der Klimaneutralität im Verkehrs- und Gebäudesektor: Investitionsanreize und Verteilungswirkungen, Mercator Research Institute on Global Commons and Climate Change.

Network for Greening the Financial System (2023), Conceptual Note on Short-term Climate Scenarios: Technical Document, October 2023.

Schrems, I., S. Fiedler, F. Zerzawy and J. Hecker (2023), Einführung eines Emissionshandelssystems für Gebäude, Straßenverkehr und zusätzliche Sektoren in der EU, Für Mensch und Umwelt, Federal Environment Agency.

Tol, R. S. J. (2023), Social cost of carbon estimates have increased over time, Nature climate change, Vol. 13, No 6, pp. 532‑536.