What are climate stress tests and how effective are they?

Climate stress tests are frequently mentioned as a key element in central banks’ toolkit to address climate-related risks to the financial system. Several countries have exercises in the works, although some analysts say they fall short of the comprehensive climate stress tests that are required, and it could be many years before their benefits are felt.

Stress tests emerged after the financial crash and have since been applied to climate

Stress tests, led by prudential regulators, are assessments of how well banks are able to cope with financial and economic shocks. They allow supervisors to identify banks’ vulnerabilities and work with those institutions to address them. Stress testing was one of the measures institutionalised by the Basel Accords after the 2008 financial crisis to decrease damage to the economy resulting from banks taking on too much risk.

Pillar 2 of the Basel Framework, which governs stress tests, exists to reinforce the first pillar, which sets minimum capital requirements. It is intended to determine whether banks need additional capital buffers to withstand stressed situations.

Climate stress tests look at banks’ resilience to transition risks, due to new policies and technologies, as well as physical risks, due to acute and chronic extreme weather events. They are based on different predictions of the policies that might be implemented, and the ability of those policies to prevent critical temperature thresholds from being breached.

Some analysts draw a distinction between climate stress tests and climate scenario analyses. Scenario analysis involves assessing how a financial institution, or group of institutions, fares in a possible future state, and may be used by firms and regulators for a range of purposes. Stress tests are generally understood to look at institutions’ ability to withstand a particularly extreme set of conditions, and are often used to judge the amount of capital banks must hold.

Climate scenario analyses are underway in many jurisdictions

Supervisors in a number of jurisdictions have exercises underway, and the majority of these use scenarios developed by the Network for Greening the Financial System (NGFS). They are split more or less evenly between bottom-up approaches, which involve financial institutions directly, and top-down approaches, which are conducted by the supervisory authority.

The European Central Bank conducted an economy-wide stress test in 2021, and launched its first climate risk stress test for individual banks in January 2022. Although the test is not a pass or fail exercise, and will not directly impact banks’ capital requirements, the results of the test will feed into a broader supervisory review which could have implications for capital rules. The results of the exercise will be published in July.

Meanwhile, the Bank of England (BoE) has launched the second round of its biennial exploratory scenario. The BoE describes this as a learning exercise that will help banks to understand and manage their climate risks, and to inform its own approach to financial and supervisory policy. It has also stated that the exercise will not be used to set capital requirements, although it is looking into the merits of climate-adjusted capital requirements as part of a separate review.

The Federal Reserve is set to carry out its own exercise. Chair Jerome Powell recently told lawmakers that “it’s very likely that climate stress scenarios, as we like to call them, will be a key tool going forward”. He noted that these “are really about assuring that the large financial institutions understand all of the risks that they’re taking, including the risks that may be inherent in their business model regarding climate change over time”, rather than adjusting capital requirements.

The People’s Bank of China conducted a climate stress test involving 21 commercial banks and two development banks in the second half of 2021. It focused only on the effects of increased emissions costs on asset quality and capital adequacy levels. A report on the exercise warned of a growing risk of defaults on loans to high carbon sectors faced with rising emissions costs.

Authorities in Japan and South Korea have each also announced plans to carry out limited scenario exercises during 2022.

Analysts say the current exercises aren’t enough

Much of the criticism of supervisors’ current approach to scenario analysis centres around three broad arguments: that the process is taking too long; that the current exercises lack sophistication; and that they are a distraction from more impactful measures.

Although many supervisors have climate scenario exercises underway, a large number are in their pilot phase, with only a handful having concluded and reported. A need for more understanding and market data about climate risks is often cited as a reason for the limited progress, but analysts have questioned this rationale. “It does not make sense to expect research outcomes in 10 years from now to support policy decisions that must be taken in the present to have any impact,” said researchers at University College London and the New Economics Foundation.

Analysts have also stressed that in order for scenario analysis and stress-testing to be useful, it must use climate scenarios that are sophisticated enough to reflect the ways in which transition and physical risks are actually likely to materialise in the coming decades.

For example, various experts have said that they should account for not only global, but also regional transition and physical risks. Others have said they should not be based on overly-optimistic assumptions about the impact of unproven technologies like hydrogen energy and carbon dioxide removal. Furthermore, they should reflect the interdependence of the climate and biodiversity, and the potential for feedback loops between banks and the real economy.

Although there is a broad consensus that stress tests are a necessary part of supervisors’ toolkit to address climate risk, some experts have cautioned against relying on them too heavily. Academics, notably Ryan Collins et al at UCL, have warned that predicting the impact of global warming on economic and financial systems is impossibly complex, and that instead of relying on inadequate methodologies to try and measure those risks, central banks should take a precautionary approach towards addressing them.

This would involve proactively steering market actors towards a managed transition, so as to minimise harm to the financial system and the wider economy in the future. For example, Finance Watch sees climate-adjusted capital rules as being the most powerful element of supervisors’ toolkit, and has advocated for a ‘one-for-one’ rule for the financing of new fossil fuel projects, meaning that banks would have to do so solely from their own funds.