How Should Banks Manage Climate Transition Risk

What is climate transition risk?^[1] How could it translate into material financial risk to banks? And how should that transition risk be measured and managed? In a recent blog post^[2], Frank Elderson, Member of the Executive Board and Vice-Chair of the Supervisory Board of the ECB, emphasized the importance of banks managing their transition risks, writing: "The misalignment with the EU climate transition pathway can lead to material financial, legal and reputational risks for banks. It is therefore crucial for banks to identify, measure and − most importantly − manage transition risks, just as they do for any other material risk."

To illustrate how banks could quantify transition risk, the ECB issued a report^[3] that measured the degree to which banks are lending to companies that are aligned (or misaligned) with the EU Commission's net zero goals.^[4]

The ECB report concluded that the euro area banking sector is substantially misaligned with EU climate goals (which are based on the goals of the Paris Agreement), and that lack of alignment equates to "transition risk", which translates into elevated default risks in banks' credit portfolios as well as increased legal risks.

The ECB's conclusion is based on the following points:

1. The EU Climate Law requires carbon neutrality by 2050.
2. The EU will impose policies that will require companies to decarbonize to keep pace with the EU Climate Law's net zero by 2050 target.
3. Companies that are not decarbonizing quickly enough to align with the EU Climate Law are at risk of defaulting on their bank loans.
4. Therefore, banks that are lending to companies that are not aligned with EU climate goals are exposed to material credit risk. Worse, the more that these companies are misaligned with EU climate goals, the greater the credit risk becomes.
5. These banks are also exposed to litigation risk because they have publicly committed to aligning their financing with net zero by 2050; a lack of alignment creates litigation exposure.
6. Banks should manage this material credit risk and litigation risk by constraining lending to companies that are not decarbonizing in line with the EU's net zero goals.

The challenge with this approach is that it assumes that transition risk to a company's business-and by extension, credit risk to the bank that lends to that company-stems from the lack of alignment with long-term government policy commitments, rather than lack of alignment with the pace of transition in the actual economy.

The assumption that misalignment of a company with government climate goals always leads to increased transition risks is false. In fact, a company's transition risk can increase if a company is transitioning in line with government goals but the underlying economy and policy environment in which the company operates are not keeping pace. Transitioning too fast may create risks for a company's business.

This is why measuring emissions misalignment with government climate goals does not actually quantify transition risk as a driver of credit risk. To incorporate transition risk usefully into credit risk management processes, we need to quantify how a particular counterparty's reduction in emissions could translate into a decline in the risk that that counterparty will default on its loan. Similarly, to incorporate transition risk in operational risk management, we need to understand how exactly reducing a counterparty's emission could reduce litigation risk.

Ultimately, if a bank employed the methodology the ECB proposed, it would not be measuring or managing its transition risk. In this note, we suggest an alternative methodology to measure transition-driven credit risk in which we estimate the transition-related default risk premium. This estimation allows us to quantify by how much a counterparty's default risk might decline if its emissions decline - assuming that the company's business model is unaffected by cutting its emissions; this is a critical assumption since reducing emissions can be resource-intensive and can negatively affect a company's business, increasing the company's default risk.

In contrast to the ECB methodology, we conclude that a bank should manage transition-driven credit risk by separating its efforts to align its financed emissions with climate goals, which is a strategic business decision, from how it manages credit risk impacts from transition risk. To manage transition-related legal risk, a bank should take into account the current legal uncertainty in managing its transition-related legal risk strategy rather than assume that legal risk always increases if misalignment of a bank's credit portfolio rises.

Estimating the Transition-driven Credit Risk Premium

In a recent article, Bolton and Kacperczyk^[5] estimated the global carbon equity risk premium. Our strategy to estimate the carbon premium in the credit markets will be to run a similar panel regression^[6] in which the dependent variable that we predict is the one-year ahead default probability of a counterparty. In the panel regression, we include variables that normally predict default probability, such as firm leverage, return on equity and stock volatility. We also include measurements of scope 1, scope 2 and scope 3^[7] emissions in the regression. If emissions have an independent ability to predict default probability, then we have found evidence that transition risk is priced in the credit markets. More importantly, the coefficient estimates would provide a way for us to estimate the magnitude of the effect.

We collect data on 1162 firms globally over the years 2004-2023 from S&P IQ Pro. Estimates of emissions by counterparty are provided by S&P TruCost data. We estimate real world default probabilities of counterparties on a monthly basis using the one-year and five-year default swap spreads and the Black and Cox^[8] extension of the Merton^[9] model.^[10] Coefficient estimates on scope 1 + scope 2 emissions as well as scope 1 + scope 2 + scope 3 emissions are strongly statistically significant.^[11] Results are depicted in Table 1 and Table 2.

Table 1

Table 1 shows the effect of reducing scope 1 + scope 2 emissions by 50 percent for typical annual default probabilities by rating category. Although the transition-risk default premium that may be attributed to scope 1 and scope 2 emissions is highly statistically significant, it is very small economically.

Table 2 shows the effect on default probability of reducing all three emission scopes by 50 percent. In this table, we have also included the average annual S&P default probability of an upward notch in the rating. For example, S&P PD+ in the B column denotes the default probability of a B+ counterparty. The rating agencies and many banks will notch ratings, providing an objective standard for defining a material change in the default probability.

Table 2

In this case, the effect of a 50 percent drop in all emissions implies approximately a one-notch upgrade in default probability.

What Do the Results Mean For Management of Transition-related Credit Risk?

The regression evidence suggests that transition risk can be incorporated into business-as-usual credit risk management under some circumstances. If a counterparty is able to make a fairly substantial reduction in all scopes of emissions, by at least 50 percent for example, then, other things equal, its credit risk will improve by a notch on average. The significant caveat is that this result depends on "other things equal"-that is, assuming that the company's business model is unaffected by cutting its emissions. If a counterparty is reducing its dependency on emissions too quickly in a way that is inconsistent with the pace of the underlying economic and policy transition, it can actually increase its credit risk. Since other things are almost never equal, transition-driven credit risk cannot be mechanically managed by reducing emissions. Rather, emissions can be one of the many factors considered in assessing a counterparty's credit risk, but only under the right circumstances.

To see how transition risks can increase if a company decarbonizes too quickly, consider the following hypothetical case. The bank is considering financing two car rental counterparties that are identical in their credit risk except that one is aggressively replacing its rental fleet with electric vehicles while the other is not. When the bank measures each rental company's scope 1 emissions^[12], it would see that emissions are being reduced much more quickly by the car rental company that is converting to electric vehicles. The bank decides to provide more financing to the car rental company that is rapidly transforming its rental fleet to electric cars.

Did the bank necessarily reduce its transition-driven credit risk by financing the company that is more closely aligned with net zero by 2050 goals? No, the bank's credit risk may have easily increased. A real-world example was Hertz's recent experience converting to an electric rental car fleet too quickly.^[13] The company lost money because it didn't properly anticipate the lack of consumer demand to rent an electric car. Hertz also had underestimated how quickly the electric vehicles would depreciate and how high the maintenance costs would be, producing losses on the resale market.^[14] By increasing financing to the rental company that had lower emissions in our hypothetical example, the bank may have increased its counterparty credit risk.

The extent to which decarbonization increases transition risk depends on the discrepancy between a company's plans and the changes that are actually happening in the underlying market. Decarbonization that is too fast may increase risks in companies that are headquartered in the European Union, which has enshrined the goals of the Paris Agreement into EU law. Both Mercedes and Volkswagen have reduced their plans to produce electric cars, finding that the current demand is not sufficient.^[15] Since both companies sell automobiles in a global market, a decarbonization strategy that may be sensible in the EU market may increase transition risks in other areas of the world that do not share the EU's climate policy framework.

In interpreting the results of the regression analysis, it is also important to note that an upgrade in default probability is only associated with an increase in all three emission scopes, which includes scope 3 emissions over which the company does not have control. The majority of companies that have emissions reductions plans do not include scope 3 emissions. As chart 1 shows, of the companies that have emission reduction plans, only about 33 percent of them include scope 3 and about 15 percent partially include scope 3.

Chart 1

Emission Targets Should Be Separated From Transition-related Credit Risk Management

Because transition risks may go up or down as a company decarbonizes and are dependent on factors that are not necessarily related to alignment with government climate goals, emission measurements and targets that some banks employ do not measure transition-driven credit risk and cannot be straightforwardly used to manage credit risk. Emission targets are rather strategic business decisions of banks. Banks as a matter of course decide where to strategically target business. They may decide to increase exposure in a particular developing country or region or to focus on a segment of the market, such as the home improvement lending market. They may also decide to allocate capital to meet the demands of the market as it decarbonizes. As circumstances change, they may rethink and change their previous business decisions. If banks change their emissions targeting strategies, that is a business decision unrelated to transition risk. A bank is not necessarily increasing or decreasing its transition risk if it modifies or even scuttles an emission targeting policy.

Because emission targets are strategic business decisions that cannot be used to measure transition-driven credit risk, target regimes should be separated from the credit risk management process. Credit risk managers should also monitor counterparty emissions, but for a different purpose: to use as an input into the credit assessment process as appropriate.

How Large Are Transition Risks in Credit Markets?

When compared to other factors that affect credit risk, such as the risk of an economic downturn, the estimates suggest that transition-related credit risks are not very large. Chart 2 shows realized default rates over time for different rating categories. Realized default rates can jump substantially during recessions, increasing to as much as three times the average default rate. In comparison, the regression estimates suggest that a 50 percent reduction in all emissions of a company, which may take a decade to accomplish in practice, can reduce the default rate only about 35 percent. Default risk can also be significantly affected by disruptions in an industry, sovereign risk, changes in the legal environment, the imposition of tariffs and numerous other factors. Transition risk can be material for credit risk in some cases, but it is by no means the most important consideration in the credit decision.

Chart 2

Managing Transition-Related Legal Risk

Transition risk as a driver of legal risk is a very new area in operational risk management, with new cases having been filed over only the last few years. As more cases are filed and decisions are rendered, the risk landscape will become more clear. A review of the recent evidence suggests that transition-related legal risk is not automatically equivalent to the degree of misalignment of a bank's credit portfolio with Paris emission goals.

Venue is Crucially Important

Since climate-related laws, regulations and precedents vary across jurisdictions, transition-related legal risk will vary as well. In the landmark Netherlands case Milieudefensie v. Royal Dutch Shell, Shell was ordered to reduce its own emissions and end-use emissions by 45% by 2030, relative to 2019. In the UK, on the other hand, ClientEarth v. Shell Board of Directors, was dismissed by the UK High Court in 2023. In that case, ClientEarth alleged that Shell's directors had failed to set sufficiently effective emissions targets and failed to establish an emission reduction strategy that would have reasonably led to the Paris Agreement goal of limiting global temperature increase to 1.5°C. The court however remained unconvinced that there is some universally accepted methodology that the directors should have reasonably followed to achieve the net zero goal. Moreover, the court recognized that the administration of emission targets is not a mechanical process; a company of the size and complexity of Shell would have to balance competing goals and interests. In 2023 in Metamorphose v. TotalEnergies, the shareholders sued the French company, alleging that assets had been overvalued because the company underestimated the future price of carbon and scope 3 emissions.

In the US, depending on the state, legal risk may increase when companies or organizations consider climate and other environmental and social-related factors in retirement plan investments. In 2023 in Spence v. American Airlines, Inc., the plaintiff alleged that American Airlines breached its fiduciary duty under the Employee Retirement Income Security Act (ERISA) by offering investment options in its employee retirement plan that employ strategies that consider climate and other environmental and social factors. A similar fiduciary duty case, Wong v. New York City Employees' Retirement System, was filed in New York in late 2023.

The evidence we have so far suggests that transition-related legal risk does not always decrease when a bank's portfolio is aligned with a net zero scenario and increase when the portfolio is misaligned; the relationship depends on the jurisdiction. It is still very difficult to predict damages, making it impossible at this point to quantify the materiality of transition-related legal risk.

Some Complicating Factors

There is a countervailing tendency for transition-related legal risk in the EU to increase with misalignment with government climate goals. Since lawsuits that allege that misalignment with government climate goals constitute inadequate management of transition risk are typically brought by independent environmental organizations or NGOs, some targets of these lawsuits are retaliating with Strategic Lawsuits Against Public Participation (SLAPP) suits that may impose civil or criminal liability on those bringing climate-related lawsuits. By one count^[16], 209 SLAPP suits have been filed around the world. The possibility of SLAPP lawsuits complicates making predictions about the future legal risk environment in Europe, and in particular any damages that may need to be paid. This year, the Committee of Ministers issued recommendations to EU member states to combat SLAPP lawsuits^[17] but it is unclear at this point to what extent SLAPPs will deter climate litigation in the EU or other areas of the world.

A very new area of environmental law could cause legal risk in Europe to increase as companies navigate alignment or misalignment with EU climate goals. In "just transition" legal theory,^[18] cases may arise if the costs of transition are being unfairly distributed. Just transition legal theory does not dispute the need for a transition to net zero but insists that any transition must respect human rights and not worsen inequality. Only a handful of cases have been filed to date, making it impossible at this point to forecast how this area of legal risk will play out. But just transition cases represent a potential legal risk for banks that are financing transition-related economic activity.

Conclusion and Recommendations

Since transition risk cannot be measured by misalignment of a bank's credit portfolio with Paris Agreement goals, a bank's emission targeting regime is not a transition risk management tool. Emission targets are rather strategic business policies about where to allocate capital. Changes in a bank's emission targets do not imply that a bank is taking more or less transition-related credit risk. Regulators should not encourage or require banks to quantify transition risk by measuring the consistency of its credit portfolio with an emission reduction scenario.

The empirical results reported on the transition-related credit risk premium suggest that the emissions of banks' counterparties could serve as a potential additional input into the credit rating process. However, the estimates also show that transition-related credit risk is not very large in comparison to the other risks a bank faces. Regulatory authorities should be careful not to encourage or require banks to divert risk resources from other, larger risks by overemphasizing transition-related credit risk.

Transition-related legal risk is very uncertain currently and depends heavily on the venue. Legal risk cannot be managed solely by reducing financing to high-emitting counterparties. As a result of the current legal uncertainty, in its external communications about its climate transition policies, a bank-especially a global bank that operates in multiple jurisdictions-may consider carefully defining and differentiating the terms "commitment," "binding commitment," "aspiration," and "ambition". In one jurisdiction, a lack of alignment with government policy goals could be perceived as a failure to manage transition risk properly and could lead to increased legal risk while, in another jurisdiction, efforts to align with government policy goals could also produce elevated legal risk.

[1] The ECB has defined transition risk as "an institution's financial loss that can result, directly or indirectly, from the process of adjustment towards a lower-carbon and more environmentally sustainable economy. This could be triggered, for example, by a relatively abrupt adoption of climate and environmental policies, technological progress or changes in market sentiment and preferences." ECB Guide on climate-related and environmental risks (Nov. 2020), at p. 10, available at https://www.bankingsupervision.europa.eu/ecb/pub/pdf/ssm.202011finalguideonclimate-relatedandenvironmentalrisks~58213f6564.en.pdf.

[2] Elderson, F, "Failing to plan is planning to fail-why transition planning is essential for banks," January 2024, available at https://www.bankingsupervision.europa.eu/press/blog/2024/html/ssm.blog240123~5471c5f63e.en.html

[3] European Central Bank Banking Supervision, "Risks from misalignment of banks' financing with the EU climate objectives: assessment of the alignment of the European banking sector," January 2024, available at https://www.bankingsupervision.europa.eu/ecb/pub/pdf/ssm.bankingsectoralignmentreport202401~49c6513e71.en.pdf

[4] To measure alignment of banks' loan portfolios with the European Climate Law, which requires carbon neutrality by 2050, the ECB used the Paris Agreement Capital Transition Assessment (PACTA)[1], an open-source software package that assesses how well financial portfolios align with different net zero scenarios.

[5] Bolton, P and Kacperczyk, M, "Global Pricing of Carbon Transition Risk," Journal of Finance, August 2023, available at https://onlinelibrary.wiley.com/doi/10.1111/jofi.13272

[6] We run the regression ln(P_it/1-P_it) = ln(emissions)_it + controls_it + fixed effects + ϵ_it, where P_it is the real world annual default probability of counterparty i at time t. Counterparty as well as month-year fixed effects are included in the specification.

[7] Scope 1 emissions are a company's direct emissions from owned or controlled sources. Scope 2 emissions are a company's indirect emissions from the generation of purchased energy. Scope 3 emissions are all indirect emissions (not included in Scope 2) that occur in the value chain of the company, including both upstream and downstream. See Greenhouse Gas Protocol FAQ, available at https://ghgprotocol.org/sites/default/files/2022-12/FAQ.pdf. emissions.

[8] Black, F, and Cox, J, "Valuing Corporate Securities: Some Effects of Bond Indenture Provisions," Journal of Finance, 1976

[9] Merton, R, "On the Pricing of Corporate Debt: The Risk Structure of Interest Rates," Journal of Finance, 1974

[10] Assuming that asset levels of the firm follow geometric Brownian motion, we use the estimated equity risk premium to move from risk-neutral default probabilities derived from CDS spreads to real world default probabilities that serve as dependent variables in the panel regressions.

[11] Standard errors are double clustered by company and year.

[12] Scope 1 emissions are emissions that are emitted from sources that the company owns directly.

[13] CNN Business, "Hertz CEO out following electric car 'horror show'," March 2024, available at https://www.cnn.com/2024/03/18/business/hertz-ceo-departure-ev/index.html

[14] Autoweek, "Hertz is selling off more EVs after major losses," April 2024, available at https://www.autoweek.com/news/a60635964/hertz-rental-ev-losses/

[15] Cleantechnica, "Mercedes latest German automaker to pull back on its electric car plans," May 2024, available at https://cleantechnica.com/2024/05/18/mercedes-latest-german-automaker-to-pull-back-on-its-electric-car-plans/

[16] Business & Human Rights Resource Centre, available at https://www.business-humanrights.org/en/from-us/slapps-database/

[17] Committee of Ministers, Council of Europe, "Recommendation CM/Rec(2024)2 of the Committee of Ministers to member States on countering the use of strategic lawsuits against public participation (SLAPPs)," April 2024, available at https://rm.coe.int/0900001680af2805

[18] Savaresi, A and Wewinke-Singh, M, "A just transition? investigating the role of human rights in the transition towards net zero societies," Select Proceedings of the European Society of International Law, 2024, available at https://dare.uva.nl/personal/search?identifier=aa5ab502-e8d4-417e-b97f-57907f1368ed

Public link

Please use the above public link if you want to share this noodl on another website.