Climate risks requirements for banks: protecting the earth or the financial system?

Reading time: 7 minutes

Save the planet!

I have become addicted to nature documentaries. Watching 40 colorful birds doing a funky mating dance in Australia or the Amazon rainforest cheers me up. However, most of these films – the best ones – ultimately leave me with a minor depression. After zooming-out, the camera shows the bigger picture – the dancing birds only have a very small patch of forest left. The rest is burned.

Climate change and its effects are everywhere. Banking supervisors are also trying to find tools to save the earth. What almost seemed unthinkable a few years ago, is now starting to become a reality. Prudential supervisors are considering to incorporate climate risks in their supervision on banks. The Dutch Central Bank (De Nederlandsche Bank, DNB) has been one of the frontrunners in this area. Now they take their initiatives one step further.

What is the policy status?

In October 2017 DNB published its ‘Waterproof’ Report, in which it researched the climate risks on the Dutch financial sector. Shortly thereafter, it declared ‘fostering a forward-looking and sustainable financial sector’ one of its three core supervisory priorities. End of 2018, DNB also calculated the exposure of Dutch banks on climate risks. One of DNB conclusions was that in the most severe climate risk scenario for Dutch banks, the average capital ratio will drop from 15.6% to 11.3%; more than a 20% decline.

At a global level, other banking supervisors also have sprung into action. The Central Banks and Supervisors Network for Greening the Financial System (the NGFS) has been incorporated. In the April 2019, the NGFS published its first Report with a number of non-binding policy recommendation. Also, very recently, the International Monetary Fund (December 2019) and the Bank of International Settlements (January 2020) published their analyses on climate change and financial risks.

The ECB, the primary banking supervisor of the European Banking Union (the SSM), and the direct supervisor of all significant banks in the Eurozone, also sees climate risk as a financial risk to banks. However, the ECB does see limits to its supervisory mandate. Andrea Enria, Chair of the SSM Supervisory Board of the ECB at a speech in November 2019:

“The role of bank regulators and supervisors is limited; our job is to ensure safe and sound banks. So, if climate change leads to particular risks for banks, we have a duty to take this into account. And if there should be a risk differential between green and brown assets, we will take this into account too. All else is beyond our mandate.”

DNB is one of first supervisors demanding concrete policy action from banks. In November 2019 DNB published its Good Practice ‘Integration of climate-related risk considerations into banks’ risk management’ for consultation. In this document, DNB expects that Dutch banks take climate risks into account in their governance, risk management and disclosure frameworks. They for instance require banks to include climate stress tests in their Internal Capital Adequacy Assessment Process (ICAAP), but also to implement a set of concrete quantitative measures to reduce or avoid climate-related risks.

But can DNB currently demand banks to incorporate their capital risks in their risk management frameworks? And as their ultimate tool, can they require those banks to hold additional capital for their climate risks?


What is climate risk?

But first, what is ‘climate risk’ and why is it relevant to banks? ‘Climate risk’ is a financial risk consisting of:

  • physical risks: these are risks resulting from increased direct climate change-related losses and damage. Think of the effects of hurricanes, draughts, forest fires, floods, etc.
  • transition risks: these are risks resulting from a disruptive transition to a carbon-neutral/low-carbon economy. Think of the risk of sustainable innovation or legislation leaving unsustainable companies or assets behind, decreasing the profit models and value of those companies or assets.


Another relevant distinction is between climate risks materializing towards the bank itself, and those that have a financial impact on the bank’s customers that are particularly exposed to such risks. The banks’ climate risk exposed client group includes, for instance, energy plants, factories, car manufacturers and farmers, but also commercial real estate investors and home-owners.

These climate risks are real. As a simple example, currently, in the Netherlands farmers and builders are all significantly economically hit by the nitrogen (stikstof) rules. This is clearly a client transition risk for banks that finance these farmers and builders. This may result in a credit risk for those banks.

Risk management and capital requirements

A bank inherently faces many risks that continuously jeopardize its financial solidity. A banking supervisor’s statutory core task is to ensure that they adequately mitigate those risks. Thus, banks are subject to a large number of very detailed risk and capital requirements. In the EU, these rules are laid down in the EU Capital Requirements Directive (CRD IV, as implemented in the Dutch Financial Supervision Act (Wet op het financieel toezicht, Wft)) and the Capital Requirements Regulation (CRR).

Currently, no specific banking requirement explicitly takes climate risks into account. However, under the current CRD IV/CRR framework, a banking supervisor can require banks to assess their climate risk impact on other specific risks to capital. Climate risks, both physical as well as transitional, may have an impact on a bank’s credit risk, operational risk, market risk and concentration risk. CRD IV/CRR requires these risks to be considered in a bank’s capital requirements.

Also, the 2017 Crédit Mutuel Arkéa/ECB judgment of the European Court of Justice, confirms that under CRD IV, in their annual Supervisory Review and Evaluation Process (SREP) banking supervisors may require banks to include all risks to which they are or might be exposed in their ICAAP. This may even result in the bank having to hold additional capital if that would mitigate a concrete, yet future, climate risks. In taking its SREP decisions, a banking supervisor has very broad discretionary powers. A court can only marginally assess those SREP decisions.

Still, according to the same judgement, those SREP decisions should meet general principles of EU (or, where relevant, national) administrative law. So a decision to take into account climate risks, should e.g. meet:

  • the principle of proportionality (i.e. the requirement should not apply to smaller and less complex banks),
  • the duty to provide reasons (i.e. climate risk must be sufficiently probable to that bank and duly motivated by the authority), and
  • the principle of equal treatment (i.e. similar banks within the SSM should be subject to similar requirements).


DNB climate risk requirements for banks?

Now how about DNB’s Good Practices? In my view, DNB cannot apply ‘climate risk’ as a generic risk to banks. They must specifically determine which elements of climate risk have an actual potential adverse impact on that bank’s capital risks; e.g. credit, market, operational and concentration risks. DNB’s Good Practices consultation document seem to apply this approach. There are however a number of critical remarks to be made.

Firstly, the legal character of the expectations. DNB bases its expectations in the Good Practice consultation document on an open norm, and not on actual concrete CRD IV/CRR rules in this connection. Apparently, DNB considers the generic requirement that banks must have in place robust governance arrangements – including effective processes to identify, manage, monitor and report the risks to which they are or might be exposed – to be sufficient. This is the same reflex under which DNB considers a generic integrity norm to be a sufficient rule for providing customer tax integrity Good Practices and expectations. As long as DNB does not turn the expectations into hard rules, and leaves the implementation to the bank itself (on a proportionate and risk based basis), there seems to be no problem. The legal problem arises as soon as DNB treats the good practices as rules in itself, and takes measures to enforce them.

Secondly, the concreteness and measurability of the climate risks. By nature, climate risks are intrinsically difficult to predict. So only if there are actually measurable, scientifically substantiated and sufficiently probable financial impacts on a bank resulting from climate risk, it is prudent – and in line with CRD IV/CRR – to take those into account in determining a bank’s capital risk management. For now, it seems crucial to develop an internationally and commonly accepted climate risk rating, calculation and modelling framework.

Thirdly, the mandate of the banking supervisor and its rules. Prudential rules, and especially capital requirements, are intended to safeguard the financial solidity of banks. They are not intended to foster another moral/policy outcome, such as a more sustainable world. No matter how urgent or justifiable such outcome is. Nevertheless, if climate risks actually impact the financial solidity of banks, and thus pose an actual financial risk, the prudential rules should come into play.

Fourthly, and finally, risk categories must be applied to all banks within the SSM in a harmonized, equal and proportionate manner. Thus, within the Banking Union, a global risk such as climate risk should not be a prudential supervisory priority of one single national banking supervisor (such as DNB), but only of the ECB and the SSM as a whole. The question is also whether DNB even has the authority to impose these prudential requirements on Dutch significant banks, as these are directly supervised by the ECB. A harmonized Banking Union approach prevents that only Dutch less significant banks will be subject to the climate risk expectations.

All in all, banking supervisors have a crucial role to play. Not to save the world from a burning planet, but from a burning financial system. If, as a by-product, their supervision also has a positive effect on the planet in the process, even better.


P.S. If you liked this blog, you may want to check this 2019 Chapter that I wrote in a book on Sustainability and Financial Markets, dealing with this topic in much more detail.