A climate stress-test of the financial system

An article, published in the journal “Nature Climate Change” on March 27th 2017. The authors, Stefano Battiston, Antoine Mandel, Irene Monasterolo, Franziska Schütze and Gabriele Visentin, have developed a “climate stress-test” of the financial system. The analysis looks beyond the fossil fuel and utility sector, by including energy-intensive sectors as well as indirect effects, and it differentiates between different types of investor.


There is growing consensus on the fact that climate change will have adverse effects on society in the long-term. However, so far it is unclear how climate change (and related policy action) might affect the financial system. Several organizations, including the G20 Financial Stability Board, have called for improving both data and methodologies to better disclose the climate risks faced by public and private investors in the financial markets.

Novelty of the research

The team has developed a novel climate stress-test methodology to assess climate risks of investment portfolios, conditional to policy scenarios. Their method allows to extend familiar financial statistics of risk for individual institutions (such as the Value at Risk) to account for the risks deriving from climate change and climate policies both through direct and indirect exposures across the network of financial contracts.


Results are based on

  1. micro-level data on equity holdings of all EU and US listed companies held by individual financial investors (from Bureau Van Dijk Orbis, 2014)
  2. balance-sheet data for the top 50 listed European banks (from Bureau Van Dijk Bankscope)
  3. financial exposures at the sectoral level from the ECB Data Warehouse


Exposures of all financial investors’ types to the fossil sector on their equity portfolios are limited (i.e. 4%-13%). In particular, the direct exposure of the top EU banks to the fossil-fuel and utility sectors is small (i.e. VaR on average 1% of banks’ capital and maximal loss about 7% of capital across banks), including when taking into account amplifications through the network of interbank obligations. This implies that climate policies would not directly cause defaults and systemic risks in the banking system.
However, combined exposures of financial investors’ equity portfolio to climate-policy relevant sectors (including energy-intensive sectors) are large (i.e. 36%-48% across investor types). Additionally, exposures of financial investors to each other also matter because they can amplify risks. In particular pension funds hold indirect exposures through their holdings in investment funds and banks through interbank lending.
Policy implications
The direct effect of climate policies on the fossil-fuels and utilities sectors is unlikely to cause banks’ defaults or even systemic risk in the financial system. Therefore EU banks should not fear the introduction of climate policies. In the context of the policy discussion around the guidelines from the G20’s Financial Stability Board Task Force on Climate-related Financial Disclosure, the results suggest that the disclosure of climate-relevant financial information is necessary to improve risk estimations and create the right incentives for investors.
However, because combined exposures are large, better disclosure may not be sufficient to mitigate risk. The timing and credibility of the implementation of climate policies matter: an early and stable policy framework would allow for a smoother adjustment of asset prices in the transition to a low-carbon economy.

Weblink to the paper: http://dx.doi.org/10.1038/nclimate3255
Weblink to infographics: https://simpolproject.eu/2016/06/10/climate-stress-test/

Correspondence e-mail: stefano.battiston(at)uzh.ch or Franziska Schütze

Funding: The work was funded by the of the EU grants FET Project SIMPOL nr. 610704, FET project DOLFINS nr 640772.