OeNB climate risk stress test – modeling a carbon price shock for the Austrian banking sector

The climate crisis is one of the most pressing global issues of our time. Policymakers across the field are challenged with the trade-offs of either taking insufficient action to tackle climate change and keeping the current economy humming or decisively addressing global warming and sending the economy into a tailspin. The introduction of a carbon pricing mechanism, one of the main policy instruments in the transition to a more climate-friendly economy, has been intensively discussed. In Austria, the government presented a tax reform package in September 2021, which also includes a carbon pricing scheme. In this article, we assess the impact of carbon pricing on the Austrian banking system in a forward-looking framework. We evaluate three scenarios over a horizon of five years: The baseline scenario is consistent with the current OeNB top-down solvency stress test and serves as a reference point. One transition scenario assumes an orderly increase of carbon emission costs for the economy, the other one envisages a disorderly increase. These two scenarios provide the empirical basis for our policy conclusions. Our stress test focuses on the transmission channels and the potential impact of transition risks on the banking system and should not be interpreted as a forecast of the development of the Austrian economy. We expand the OeNB’s top-down stress testing infrastructure with two additional models. First, we develop an enhanced multiregional input-output model to calculate cost and turnover changes for different economic sectors following the introduction of carbon pricing schemes. Second, we expand the OeNB’s corporate insolvency model introduced in 2020 to assess the impact of the COVID-19 pandemic to include shocks such as a carbon emissions-based shock. This allows us to assess the impact of the aforementioned policy measures on sectoral insolvency rates, which is then used as an approximation for stressed credit risk default probabilities. In addition, we use these stressed default rates to derive valuation losses in Austrian banks’ bond portfolios. Both inputs feed into the OeNB’s top-down stress testing framework ARNIE, making it possible to assess the impact on the Austrian banking system. Our results imply that especially the disorderly transition scenario can have a sizable impact on certain economic sectors, most importantly agriculture and transport, where default rates would rise sharply, affecting banks exposed to these sectors. The aggregate CET1 ratio for the Austrian banking system would decrease by 2.7 percentage points in the disorderly scenario and by 0.7 percentage points in the orderly scenario. Given initial capitalization levels, this seems manageable. Hence, while the introduction of a carbon pricing mechanism will certainly create additional costs for the Austrian banking system, our results indicate that the banks are well placed to withstand the indirect effects of measures to counter the climate crisis.