Model-based Financial Regulations Impair the Transition to Net-zero Carbon Emissions

April 15, 2024Published by Nature Climate Change

Model-based financial regulations could inadvertently contribute to the build-up of systemic climate risks by discouraging banks from divesting from high-carbon assets, say researchers from the University of Oxford’s Institute for New Economic Thinking.

The use of backward-looking data in financial accounting frameworks underestimates the risk associated with high carbon assets, creating a bias in loan provision requirements that penalises clean energy investments, according to the paper’s analysis.

While this research focuses on accounting, the authors – Matteo Gasparini, Matthew Ives, Ben Carr, Sophie Fry and Eric Beinhocker – state that future work should investigate the potential bias in other model-based financial regulations, such as capital requirements regulations.

In the study, the authors analysed data from the European Banking Authority’s transparency exercise to compare the provision coverage ratio (PCR) – the ratio of loan loss reserves to outstanding loans – of high and low-carbon sectors.

PCR is a proxy for the bank’s estimates of expected credit losses. A higher PCR indicates that the bank has to set aside more reserves to cover potential future losses due to higher perceived risks.

The results show that in 2021 the average PCR of EU banks was substantially lower for high-carbon (1.8%) than low-carbon sectors (3.4%). This result is consistent for banks of different portfolio sizes and across different countries.

The authors then considered the impact of a divestment strategy away from high-carbon sectors, and found that biases in backward-looking analyses can result in substantial and adverse impacts on bank profitability.

Banks looking to divest from polluting activities would have to increase their loan loss provisions to cover the higher estimated risk of low-carbon ones, eating into their profits.

If the 59 largest EU banks were to divest from high-carbon sectors and reinvest in low-carbon activities, they would, on average, record losses equivalent to 15% of profits from the previous five years. The results also show that banks with a larger difference in PCR between high and low-carbon firms see the most substantial effects.

The core problem with using backward-looking data to assess climate-related components of risk is that it cannot capture the “structural breaks” that occur during major transitions, like the shift to a low-carbon economy, say the authors.

While oil and gas have historically had lower financial risk metrics (such as interest coverage and debt ratios), emerging and unprecedented climate-induced risks will substantially alter this when compared to renewable energy firms.

For instance, a US$100 per ton carbon tax or a $20 per barrel write-down of oil reserves would significantly worsen the financial ratios of fossil fuel sectors.

The authors argue that a more forward-looking assessment framework would be better suited to capturing these and other emerging climate-related risks. Such forward-looking models are likely to incorporate climate scenarios and, while developing them, analysts should assess whether they include similar biases, state the authors.

This page was last updated April 15, 2024

Share this article