Banks' exposure to climate risk is much broader than they are disclosing — and than is commonly understood by regulators and investors. In an October report, "Financing a Net-Zero Economy: Measuring and Addressing Climate Risk for Banks," Ceres found that more than half of bank lending is potentially exposed to climate risks due to the failure of many bank clients to plan for the transition to a net-zero economy. A worst-case scenario could result in hundreds of billions of dollars in losses for U.S. banks and potentially trigger a new financial crisis.
In assessing climate risks, many banks reason that transition risk may be less of a threat because most of their loans are short term, and they simply can decide not to renew the loans that might face increased risk due to climate change. New analysis from Ceres finds that that reasoning is flawed, and that banks must look beyond the individual loan to the long-term and multi-faceted relationship with the client. When that lens is applied, the need for banks to address climate risk in their portfolios becomes much more urgent.
The urgency comes from the fact that, due to the complexity of these long-term client relationships, it might take a decade to implement a proactive engagement strategy that puts clients on the path to decarbonization. Right now, most banks aren’t even thinking about time horizons longer than five years. The solution is for banks to do three things:
- Think longer term by adopting 10-year time horizons for risk management
- Plan longer term by creating 10-year financing plans for each sector
- Act longer term by engaging clients starting now and incorporating phases of evaluation, collaboration and execution over the next 10 years.
Those solutions are supported by quantitative analysis from our partners at CLIMAFIN and Ceres’ in-depth interviews with industry experts and investors on the topic of relationship banking. The long-term focus is needed because banks can’t quickly move away from long-term client relationships, which are built on a valuable combination of customer loyalty and the mutual sharing of information, without significant costs. And for decarbonization to happen within banks’ portfolios, as one former banker said, "it has to be a profit center rather than a cost center for the bank."
By examining the unique aspects of relationship banking, we outline recommendations below for how banks can think, plan and act in ways that will minimize the risk of climate change for themselves, their clients and society.
Part 1: The value of relationship banking to banks
If all loans and other banking services were independent of each other, banks theoretically could move quickly to react to escalating climate risks within a five-year planning window.
However, even casual observers can see that interdependencies exist. Just like most consumers establish a relationship with their bank that starts by opening a checking account but then grows to include savings and loans, banks’ commercial customers rely on them for many related products and services.
More than half of bank lending is potentially exposed to climate risks due to the failure of many bank clients to plan for the transition to a net-zero economy.
We’ve found through multiple interviews that banks work to move clients up a "food chain" of financial services. Typically, this means offering clients favorable terms on lower-margin products such as revolving credit facilities to try to secure more profitable services (typically investment banking- and advisory-related). Losing out on a loan doesn’t just forgo the revenue from that loan, it also jeopardizes these other lucrative opportunities. As one former bank executive put it, "Banks will typically fight over opportunities to participate in revolving credit facilities because that’s seen as an entrée into conversations about other, more profitable services."
Relationship banking is also valuable to banks because it helps them determine the creditworthiness of their customers. Public information on this is often imperfect, creating market inefficiency that banks can exploit. A sustained relationship gives banks private information about their clients (such as through credit agreements) and reinforces loyalty because it isn't easy for clients to prove their creditworthiness to other banks. This relationship banking can help banks maintain profitability despite competitive pressure. Research shows that it increases in more competitive lending markets, such as the U.S.
Given these dynamics, it doesn’t make sense to assess the risk of individual corporate loans in isolation. That would misleadingly suggest that because loans only last 25 months on average, it’s possible for a bank to just walk away when that time is up. In reality, because client relationships typically generate revenue from many products and services over decades, a different unit of analysis and a longer time horizon for risk management and strategic planning are required.
Recommendation: When evaluating climate risk, the unit of analysis banks should use is the client relationship, including all the revenue derived from that relationship over time.
Recommendation: To effectively assess the financial impact of climate change on client relationships, banks should extend their planning and risk management time horizons to include a 10-year view.
This longer term view is uncommon in the U.S., where the regulatory framework is built around time horizons of nine and 20 quarters. Confidential Ceres research from 2019 shows that only one of nine banks studied is looking at a risk management time horizon longer than five years. Two others are looking longer term with respect to climate risk, but not for more conventional risks such as credit risk or market risk.
Part 2: Relationship banking and climate policy relevant sectors
If they do have to confront a disorderly carbon transition, banks that have deep relationships with high-carbon firms will face a difficult trade-off. They either will have to refinance customers whose credit quality suddenly has deteriorated or stop lending to them. In the first case, they might avoid losses in the short term but end up increasing their long-term transition risk. Alternatively, banks could abruptly terminate lending relationships with certain borrowers (by declining to roll over credit facilities). This would destroy value on two levels. First, for some banks, it would lead to a substantial loss of customers and drop in market share. As one former JPMorgan banker put it, "It would be very damaging to the relationship with a client if a bank said it wouldn’t roll over a revolver, particularly if other banks continued to extend credit." Such abrupt action would destroy the informational value of the lending relationship.
The way to avoid this value destruction — and mitigate the worst impacts of climate change — is for banks to help clients gradually and strategically reduce their risk through transition plans aligned with the objectives of the Paris Agreement. This means that banks and their clients need to set net-zero targets that aim to decarbonize most sectors by 2040, and the whole economy by 2050.
A sustained relationship gives banks private information about their clients and reinforces loyalty because it isn't easy for clients to prove their creditworthiness to other banks.
Banks can help clients make the business case for change (and reduce their own risk) by adjusting the pricing of products and services to match the level of climate risk their clients face. Banks also can help clients by providing transition finance — many firms’ decarbonization plans will require significant injections of capital.
While target setting, pricing adjustments and transition finance are approaches that can help clients in all sectors lower their climate risk, other engagement considerations (material issues, due diligence procedures, data availability, level of sophistication, technology pathways) might vary depending on the sector.
As a result, banks need to prioritize certain sectors, at least in the near term. While the most important criteria that banks should use to decide which sectors to prioritize are their level of financial exposure to a given sector and the level of climate risk that sector faces, the extent of relationship banking in a sector is also important.
In sectors with high levels of relationship banking, additional value is at risk beyond what’s on the balance sheet. It may take longer to move away from certain clients in these sectors if engagement isn’t successful. Conversely, high levels of relationship banking in a sector may allow engagement to move faster and have a higher chance of success, not only because a bank might have more leverage in these situations but also because of pre-existing trust and connectivity.
Recommendation: Banks should adjust their sectoral engagement strategies based on the level of relationship banking in each sector. Sectors with many valuable relationships should be prioritized, engagement should move faster and more time should be allocated for these relationships to potentially unwind or be restructured.
Banks know who their most valuable clients are, but this may be difficult to see from the outside. Until banks provide greater transparency, investors and other stakeholders can use two main indicators common in academic literature to measure the extent of relationship banking in a sector: duration of loans and exclusivity of the lending relationship (number of lenders). Long-term loans concentrated among a small number of lenders may be indicative of a high level of relationship banking.
To analyze the interplay between climate risk and relationship banking, Figure 1 maps economic sectors (using CLIMAFIN’s CPRS classification system) along these two dimensions.
This indicates that relationship banking is more prevalent in the most climate-relevant sectors, especially electricity, oil and transportation. Substantial bank resources should be directed toward client engagement in these sectors as soon as possible. It is also apparent that the average maturity of loans in the coal sector is lower than for other fossil fuels. This gap has emerged recently, and hints at an increase in the perceived riskiness of coal and the associated decline of banking relationships in many cases.
Figure 1 is based on the Refinitiv DealScan dataset focused on syndicated loans. These loans have a longer maturity than the average commercial and industrial loans, 60 months compared to 25 months. Our inferences about relationship banking in climate-relevant sectors ideally would rely on internal bank data, but we have no reason to expect that would change them significantly; if anything, the nature of syndicated loans (involving many banks) could mean that the extent of relationship lending is underestimated.
Part 3: The need for a 10-year plan to mitigate risks and capitalize on opportunities
Banks have built their reputations on "keeping clients happy … and ensuring they are well positioned." This is why it has been easier for banks to capture opportunities associated with climate change, such as sustainable finance, than it has been to mitigate climate risk, which inevitably involves having to decline certain transactions.
Yet, while relationship banking can add risk, it paradoxically provides an opportunity to develop a positive approach to climate risk mitigation that banks can sell to their clients. Typically, environmental financing restrictions apply to individual transactions (for example, "We won’t lend you money to build a new coal mine"), which can feel like a lost opportunity. However, if both bank and client think at the relationship level, this perception can be reversed. Chase Savage, ESG Research Analyst at Fidelity Investments, suggests that banks could present their clients with a 10-year Paris-aligned financing plan for their sector and ask them how they’d like to fit in, rather than presenting them with a list of restrictions and asking them to work around them. "The next step is to have a much more in-depth client experience," said Savage.
Losing clients will be a concern as banks move through these phases. If banks move too fast, there could be a short-term cost as affected clients look elsewhere for financing. This could be counterproductive, especially if those clients move to less regulated parts of the financial system where they won’t face climate-related scrutiny. This is another reason why banks need to allow a longer lead time for execution, and why the largest U.S. banks need to begin now.
The progression through these stages will vary by sector and firm. Banks are already in the execution phase in the most exposed sectors, such as coal. In other key sectors, such as oil, gas and electricity, banks are quickly moving into the collaboration phase as their risk becomes increasingly clear. The remaining climate-relevant sectors are in the evaluation phase, as banks grapple with the risk these sectors could present and the technological paths their transitions could take.
Getting all clients on the path to success means that banks will have to be well into the execution phase by the end of the decade. It will take time to exit relationships that don’t fit with a bank’s climate strategy. More specifically, our research shows that, given the share of banks’ portfolios in the fossil fuel and electricity sectors and the average rate of change of overall portfolio composition, it could take several years for most banks to make important structural changes to this part of their lending portfolio.
Figure 2 shows that, for most U.S. banks, the sectoral composition of their loan portfolios evolve slowly, with a yearly rate of change between 5 and 10 percent, despite market shocks such as the 2007-2008 financial crisis. This rate has been decreasing progressively over the past 20 years.
Figure 3 shows the evolution of the exposure that banks face to the most climate-relevant sectors. CLIMAFIN research finds that while close to two-thirds of bank lending is climate-relevant, a smaller proportion (about 15 percent) is in "core" sectors related to fossil fuels and electricity. These core sectors, shown in the graph, have the highest levels of relationship banking and, in many assessments, they also face the most climate risk.
Comparing the two charts at a high level, we see that the share of banks’ portfolios in the fossil fuel and electricity sectors is substantially larger than the average rate of change of overall portfolio composition. This means decisions about the success or failure of client engagement in these sectors will have to be made in the next few years, to allow sufficient time for the relationships to be unwound or restructured, if necessary.
The coal industry (see Figure 4) provides a useful illustration of what happens when a climate-relevant sector comes under gradually increasing but ultimately severe economic pressure. The fall in loan value of about 70 percent from the peak reflects both transition (as companies move away from coal and their loans are reallocated to a different subsector) and disengagement (as coal companies that did not transition are deemed a poor investment.)
In transition scenarios where a sudden crisis is avoided, other fossil fuel sectors might follow a similar pattern. The fall in coal lending, spread over a decade, represents a reasonable upper bound for the possible speed of transition and should inform the development of banks’ 10-year plans. A rapid exit from any of these sectors is unlikely, as one former banker noted: "There will be tremendous pressure not to fire profitable clients … the fear of not being top of the league table is the cultural driver." A more feasible transition will take time and should start as early as possible. "They will need to make the case clear to the client with a lot of runway to get them to change," said the same former banker. Working backward from 2030 means that much decision-making within banks will have to be done in the next three to five years. This adds to the urgency created by the possibility of a near-term systemic shock and should prompt banks to take immediate action.
Recommendation: Banks should immediately prioritize client engagement in climate-relevant sectors, with the aim of determining, before 2025, the ability and willingness of clients in the riskiest sectors to implement robust transition plans.
Some banks are already acting. JPMorgan Chase recently established a Center for Carbon Transition focused on client engagement. Other banks must devote even greater resources to this challenge to address the risk to capital markets and society more broadly. Banks that are proactive will have major upside over the next few years, with winners and losers being determined quickly. As one former Goldman Sachs banker told us: "Fossil fuel valuations have collapsed already … the [leading] banks see what’s going on … they are not looking back, they’re looking forward."
Technical analysis by CLIMAFIN consultants Stefano Battiston, Antoine Mandel and Irene Monasterolo.