A step-change: how a systemic risk buffer could benefit transition finance
The financial consequences of climate change and the necessary transition to global Net Zero by 2050 have made it difficult for financial institutions to change the way they make decisions quickly enough. A working paper published by researchers at the European Central Bank provides evidence for how the financial sector could be insulated from any losses by creating a systemic risk for the entire sector.
One of the challenges in addressing the financial losses associated with climate change and the climate transition is the difficulty of quantifying future losses, because the realized losses won’t follow the historical patterns that have typically informed banks’ approaches to risk. One major feature of the climate transition is a need to invest in green technology and to rapidly transition assets that are currently misaligned to Net Zero so that they do align.
Until now, most of the regulatory responses to the risks associated with climate change have been incentives for non-financial companies to make green investments, greater disclosure by corporations and financial institutions about their emissions, and climate stress-testing exercises by central banks and supervisors.
One of the strongest (or bluntest, depending on the perspective) tools for regulators to address climate risk is to adjust capital requirements for banks depending on whether they are financing green or unsustainable assets. These have often been proposed in the form of either a ‘green supporting factor’ or ‘dirty penalizing factor’, which adjust the risk weighting of individual counterparties depending on whether they qualify as either green or unsustainable.
These supporting or penalizing factors have been commonly proposed but rarely implemented because the implications would be significant. Consequently, they need to be clearly evidenced as effective and efficient in achieving their stated objective of either encouraging more green finance or disfavoring unsustainable activities. There is a general lack of evidence to support the proposition that individual green investments have in the past demonstrated a significantly better credit risk profile compared to unsustainable activities notwithstanding the significant likelihood of prospects for green and unsustainable activities.
The ECB has been among those central banks conducting climate stress tests, and the results were used in this new research to provide insight into the transition risk exposure of about 100 European financial institutions. The research was designed to demonstrate how a supervisor might go about calibrating a systemic risk buffer (SyRB) to account for the transition risk of the financial institutions involved.
The SyRB was developed to reflect the overall level of near-term transition risk exposure of the financial institution – within the coming three years — and not be linked to individual green or dirty assets. Instead of adjusting the risk weighting of individual exposures, as a green supporting factor or dirty penalizing factor would do, it groups financial institutions into buckets based on the potential transition risk relative to their risk-weighted assets. The method for estimating the transition risk builds on the loan-level estimates of the climate stress test while filling in data gaps in loan-level data with ‘country-sector’ probability of default levels for both loans and bond holdings while using country-level default probabilities for household exposures.
The different scenarios used in the stress tests for corporate exposures affect default probabilities due to partial spillover of cross-sector gross value-added shocks caused by energy price rises, a fall in corporate profitability, and increased levels of indebtedness for renewable energy and energy efficiency investments. Household default probabilities are similarly affected by reduced disposable income due to energy price rises and higher indebtedness for energy efficiency improvements, along with real estate prices and long-term interest rates.
European regulations require that risk buffers increase in 0.50% increments based on the output of a calibration that models the increase of transition risk in an accelerated transition, compared to current policies which introduce the risk of a metric being ‘gamed’ by banks. However, with a multi-tier system (in the calibration they cover five buffer levels between 0% and 2%), this allows for a significantly graduated increase in the buffer as risks rise.
The researchers also leave open that the SyRB could be used to cover system-wide climate risks on less than a 1-for-1 basis to reduce the impact on banks. However, the limits of this from their study is that if the capital buffer is scaled down by a factor of 4 so that transition risk exposure of 2% of risk-weighted assets leads to an increased capital of 0.5%, it nearly ceases to provide a buffer compared to the expected (and conservatively estimated) climate losses.
It would be one thing if these risks were spread across the financial system as a whole, where even the full impact of the losses is distributed across many institutions. One of the notable findings in the assessment of the transition risk is that it is heterogeneously spread, with concentrations of transition risk more with banks that have less of an excess capital buffer (excess CET1 ratio). This means that within the eurozone at least, transition risk losses would be more likely to affect already weaker banks, which increases the systemic risk impact of climate change.
This analysis produces a few notable conclusions that are relevant across markets, particularly those where transition finance is most needed. For starters, it provides a clear use for the model applied during climate stress tests without making those test outputs determinative of capital levels. Climate stress tests have often been intense exercises for both banks and regulators to undertake to provide evidence about the vulnerability of financial institutions, as well as their preparations to be able to weather climate risks.
Using the collected data for calibrating a systemic risk buffer provides a tangible use for the stress tests and a data-guided way to balance the risk of financing climate-exposed sectors with the short-term gain that banks have by continuing to provide financing. Interestingly, one of the concerns with loan-level, risk-weighting adjustments up or down for ‘green’ or ‘dirty’ investments is that it provides more incentive for banks to finance ‘green’ and avoid ‘dirty’ investments, whereas with transition risk buckets, there would be substantial leeway for banks to finance companies transitioning activities from unsustainable to sustainable activities without facing increases in their capital requirement.
