The challenge of phasing-out fossil fuel finance in the banking sector

The Paris Agreement's goal of limiting global warming is jeopardized by continued high fossil fuel consumption and emissions from existing assets. Financing of new and existing fossil fuel assets must be phased out to align with decarbonization pathways. Banks, as primary financiers of fossil fuel companies, play a critical role in this process. While voluntary initiatives like the UN-convened Net Zero Banking Alliance (NZBA) exist, criticisms of greenwashing and insufficient action persist. Banks, despite the long-term risk of stranded assets, maintain short-to-medium-term investments due to ongoing exposure to the fossil fuel sector. The study focuses on syndicated loans, a significant mechanism for financing capital-intensive fossil fuel projects. Syndication spreads risk among multiple banks, amplifying the impact of individual banks' investments. It also facilitates finance substitution, where banks in regions with strong climate policies withdraw financing, replaced by lenders from countries with weaker policies. This phenomenon has been observed in Australia's coal lending market, with lenders from China and Japan filling the gaps left by Australian banks. Understanding the syndicated fossil fuel debt market's response to policy measures is crucial. The study analyzes the syndicated lending network structure, the behavior of key lenders, finance substitution, and the potential role of regulation to counteract these dynamics. Market failures and lack of international coordination highlight the need for regulatory instruments to reallocate bank finances.

The paper builds upon existing literature on financial networks and their role in understanding various phenomena relevant to financiers, governments, and regulators. It distinguishes itself from studies on financial stability, focusing specifically on the network of syndicated lending relationships rather than interbank liabilities. The study aims to contribute to the development of stress-testing tools and macroeconomic models by providing a more detailed representation of the banking sector and its transition dynamics.

The analysis uses data from Bloomberg, covering syndicated debt (loans and bonds) provided to fossil fuel companies between 2010 and 2021, totaling \$7.1 trillion from 709 banks. The data primarily focus on the dynamics of global fossil fuel debt supply, lacking information on borrower location or asset type. The study defines banks' "syndication activity" as the total value of deals where a bank is a syndicate partner, reflecting their broader influence. To analyze the lending network, a network is constructed linking banks based on their co-investments in fossil fuel debt syndicates. Centrality measures (eigenvector, betweenness, closeness, and degree centrality) are employed to assess the network's topology and banks' positions within it. The analysis then investigates finance substitution and its potential counteraction through prudential regulation. An illustrative network model simulates bank phase-out and finance substitution. Banks sequentially phase out, removing their finance from deals. Deals then seek substitute financing. The model incorporates a "finance limit percentage", representing regulatory caps on new fossil fuel finance. Scenarios include random, targeted (systemically important banks first), and regional phase-outs. The model assumes complete cessation of fossil fuel investment by phased-out banks and considers two scenarios for finance substitution: 'any substitute' (any available bank can provide substitution) and 'syndicate substitute' (substitution prioritized based on existing network relationships). Additional constraints on substitution (limiting the number of candidate banks or the number of substitution attempts per deal) are also explored. The model uses data from 2021, but qualitative results are shown to be robust across other years.

The study reveals no systematic decline in fossil fuel lending since the Paris Agreement, although regional variations exist. European banks have led the phase-out, reducing both individual lending and network influence. However, Japanese and Canadian banks have increased their activity, indicating finance substitution. US banks remain dominant but have shown some reduction in lending. The model demonstrates that without regulation, finance substitution renders phase-out inefficient. A tipping point emerges with regulation, where phase-out efficiency sharply increases as banks reach their finance limits. This tipping point occurs sooner under stricter regulations. Stringent regulation can lead to a phase-out multiplier exceeding 1, meaning that the direct phase-out of one bank indirectly leads to further phase-outs. Targeted phase-out of systemically important banks significantly enhances efficiency, requiring fewer banks to exit the sector. Network structure influences substitution: the 'syndicate substitute' scenario leads to a later tipping point due to larger banks' greater capacity for substitution. Additional restrictions on substitution (limiting the candidate banks or number of substitution attempts) further accelerate the tipping point. Regional analysis reveals regional differences in phase-out efficiency. US banks, and to a lesser extent Japanese and Canadian banks, are crucial in providing substitute finance, hindering regional efforts. Chinese banks' relative isolation makes them more susceptible to a successful phase-out.

The findings demonstrate the critical role of syndicated lending networks in shaping policy responses to fossil fuel finance. The resilience of the market to uncoordinated phase-out attempts underscores the need for robust, coordinated regulatory frameworks. The model's tipping point dynamics illustrate the potential for efficient transitions with appropriate policy interventions. Targeting systemically important banks for phase-out optimizes outcomes, while regional disparities highlight the importance of international coordination. The study's insights can inform the development of climate-focused regulatory frameworks by central banks and financial supervisors.

This research underscores the significant challenge of phasing out fossil fuel finance due to the resilience of syndicated lending networks to uncoordinated actions. Prudential regulations, particularly those limiting banks' fossil fuel lending, are crucial in creating a tipping point for efficient phase-out. Targeted interventions focusing on systemically important banks and international coordination are essential for maximizing effectiveness. Future research should explore the development of more sophisticated models incorporating additional behavioral and economic factors.

The model makes several simplifying assumptions, including the complete cessation of fossil fuel investment by phased-out banks, sequential phase-outs, and the specific rules for finance substitution. The analysis also lacks granular borrower-level data, limiting the ability to assess just transition implications at a national level. Future research could address these limitations by incorporating more detailed bank-level data and relaxing simplifying assumptions.