ECB climate models are built on obsolete scenarios

The author is professor of environmental studies at the University of Colorado

The late Kenneth Arrow, who won the Nobel Prize in Economics in 1972, previously worked as a long-term weather forecaster in the US military. When he realized that this weather forecast was no more accurate than random guesses, Arrow passed the message on to his superiors. “The commanding general is well aware that the forecast is not good,” Arrow was told. “However, he needs it for planning purposes.”

Much like the commanding general of Arrow, the European Central Bank is currently conducting an economy-wide climate stress test. This makes sense because climate change is real and serious. The ECB test, which aims to assess “the exposure of euro zone banks to future climate risks. . . under different climate scenarios ”, is also ambitious and far-reaching. As ECB Vice President Luis de Guindos wrote in March, it looks 30 years away and covers around 4 million companies worldwide and 2,000 banks – “almost all monetary financial institutions in the area euro ”.

Unfortunately, the scenarios on which the ECB relies are obsolete, calling into question the entire exercise. Although somewhat technical, come into the rabbit hole with me on this one as things get interesting quickly.

The ECB and more than 60 other central banks belong to the Network for Greening the Financial System. This volunteer group covers most of the global economy and advocates for climate stress tests to assess financial resilience “to what-if, extreme, but plausible scenarios.” However, these scenarios are accompanied by a health warning, namely: the most commonly used “were designed about 10 years ago and do not correspond well to recent trends in emissions”.

In fact, as my colleagues and I have documented, these scenarios have dramatically overstated carbon dioxide emissions since 2005. Moreover, even the most recent NGFS scenarios are out of step with real world data. This could lead to bad decisions and possibly even higher systemic risks.

The NGFS uses three main scenarios to guide climate assessments. “Orderly” assumes that policy change begins now, leading to net zero emissions by 2070, with emissions reaching 2100 and severe environmental consequences.

The NGFS is clear that compromises are needed to reduce emissions. A carbon tax could accelerate the transition to net zero and promote carbon capture technologies. At the same time, however, it would increase the costs of some raw materials and force companies to review their operations. This could lead to lower short-term growth but much higher long-term growth – the opposite of the greenhouse scenario.

Unfortunately, some of the other assumptions underlying the three NGFS scenarios are implausible.

For example, they assume global emissions of carbon dioxide from fossil fuels in 2020 of about 36 gigatonnes. Yet in 2020, these emissions were closer to 33 Gt, a gap that is only partly due to the dire economic consequences of the pandemic. By 2040, this little mistake becomes very important. The “greenhouse” scenario predicted around 45 Gt of fossil fuel emissions. Yet even before the commitments made at last month’s climate summit, the International Energy Agency is projecting around 35 Gt of emissions in 2040. The gap widens as you move forward. future. According to a recent analysis in Nature, fossil fuel emissions would be around 25 Gt by 2100, under assumed 2019 policies and technologies. Yet the greenhouse scenario projects 81 Gt, or more than three times that.

Stress tests use “what-if, extreme” scenarios and, as the NGFS clearly shows, these must be “plausible”. Yet it is simply not convincing that the world collectively decides to convert much of its energy supply to coal, which would be necessary for fossil fuel emissions to exceed 80 Gt by 2100.

The ECB uses its climate scenarios in the same way that Arrow General used his ‘bad’ weather forecast – because there is no alternative for planning purposes. Nevertheless, it is urgent to take stock. Because these climate scenarios are usually out of sight and technical, those who rely on them may not know how much they have diverged from the real world. Because they are institutionalized, changing these stress tests will also be difficult.

Nonetheless, if financial institutions are to perform meaningful climate stress tests, it is imperative that they be based on the latest scientific data and not on improbable scenarios.