2025 banks’ energy financing ratios: the good, the bad and how to move forward

BNP Paribas was the only global bank willing to publish an Energy Supply Financing Ratio (ESFR) in 2024. By the end of 2025, four more banks had followed suit. The ESFR measures how much a bank finances sustainable power supply, such as renewable electricity from wind and solar (1), relative to fossil fuels. This provides insight into the bank’s climate action. Today, five banks either publish this indicator, its methodology, or have committed to achieving one in the medium term. Despite their different approaches, some common features are emerging, as are loopholes that could affect their credibility. A regional divide is also emerging, with European and North American banks adopting notably different approaches. This analysis examines these differences and highlights where the methodologies succeed, fall short, or distort the reality.

To avoid the worst effects of climate change, two major shifts in global annual energy investment are required: a rapid decrease in fossil fuel investment and a rapid increase in the supply of sustainable energy, primarily in the form of power. If the world is to achieve carbon neutrality by 2050 and limit global warming to 1.5°C, as described in the International Energy Agency’s Net Zero Emissions (NZE) scenario, banks should allocate six dollars to sustainable energy for every dollar they allocate to fossil fuels by 2030 (a 6:1 ESFR) (2). Yet our research shows that, between 2021 and 2024, the 65 largest banks worldwide allocated only 42 cents to sustainable power supply for every dollar directed to fossil fuels (an ESFR of 0.42:1). Banks therefore have a long way to go.

The ESFR has gained momentum in recent years. Institutions such as Bloomberg New Energy Finance (BNEF), the European Banking Authority (EBA), the Science Based Targets Initiative (SBTi), the World Resources Institute (WRI) and the Institut Louis Bachelier (ILB), have increasingly studied the metric as an indicator of banks’ climate action (3). Shareholder pressure has significantly accelerated this trend. In 2024 and 2025, the New York City Comptroller (NYCC), which oversees the city’s public pension funds, and SHARE filed resolutions urging major North American banks to publish a “clean energy supply ratio”, without prescribing a methodology. These negotiations led banks including JPMorgan Chase, Citi, Scotiabank and RBC to disclose or commit to disclose their ratios. As the NYCC has noted, the metric is poised to become a standard disclosure and a key tool for investors to judge whether banks are shifting capital at a pace consistent with global climate goals (4).

French banks BNP Paribas and Crédit Agricole, are the only major global banks to publish their current ratio and commit to 9:1 targets by 2030 and 2028, respectively (5). But their approach relies on narrow definitions of fossil fuel activity, incomplete coverage of financing instruments and opaque methodologies (see Table 1).

Both banks restrict the fossil fuel component of the ratio to certain segments of the oil and gas value chain, meaning that some fossil fuel activities are excluded from the calculation. Crédit Agricole covers only extraction, while BNP Paribas includes extraction, production, and refining. Major sectors such as liquefied natural gas (LNG) and gas-fired power generation are excluded altogether. In addition, both banks calculate their ratio solely on credit exposure, omitting financing provided through bond and equity underwriting. Between 2021 and 2024, these activities represented 36% of BNP Paribas’s and 27% of Crédit Agricole’s fossil fuel financing (6), meaning their ratios significantly understate their true support for fossil fuels. Furthermore, neither bank publishes its detailed ratio methodology, making their calculations opaque, particularly with regard to which technologies qualify as “green” in the case of Crédit Agricole.

Across the Atlantic, the picture looks different. US banks Citi and JPMorgan Chase do not set medium-term ratio targets, but they do disclose their current ratios along with detailed methodologies (7). Canada’s RBC publishes its methodology but not its current ratio (8). The strengths of these methodologies include covering larger portions of the fossil fuel value chain (although Citi does not explicitly state whether this includes LNG) and incorporating both lending and bond underwriting, which provides a more comprehensive overview of fossil fuel financing (see Table 1).

However, the integrity of these ratios is weakened by the inclusion of fossil-related technologies on the “low-carbon” side, or the inclusion of sectors unrelated to energy supply. Citi and JPMorgan Chase both allow fossil-based hydrogen and certain carbon capture and storage projects on the “low-carbon” side (9). JPMorgan Chase even includes fossil-fired power with carbon capture. RBC’s inclusion of “decarbonisation finance” is also problematic because it encompasses activities such as cement manufacturing and other industrial processes that are not related to energy supply. JPMorgan Chase also includes public electric vehicle (EV) chargers on the “low-carbon” side, while fossil fuel counterparts, such as filling stations, are not covered on “high-carbon” (10). These methodological choices inflate the sustainable portion of the ratio and risk overstating progress toward the energy transition.

While some progress was made in 2025 regarding the adoption of ESFR by banks, much more is needed. A rapid, global and fair transition to sustainable energy sources is vital, and banks should support this urgent shift instead of perpetuating fossil-based energy sources. To do so, banks must disclose their current ESFR and set a dedicated target for 2030. This requires reducing financing for fossil fuels and ending all support for their expansion, while accelerating financing for sustainable power supply. The ESFR methodology should be robust and transparent, covering all fossil fuel financing products and services across the entire value chain, and being based on sustainable technologies.