How demanding and consistent is the 2018 stress test design in comparison to previous exercises?

14-06-2018

We provide an assessment of the design and calibration of the 2018 EU-wide stress test. The adverse scenario for the 2018 stress test is more severe than for previous stress tests in terms of the assumed GDP decline in the EU area. However, the test is less severe in terms of the losses that banks are expected to incur under the scenario. The adverse scenario has a highly asymmetric impact on different European countries, such that countries with a high degree of trade openness are affected considerably more. It seems unlikely that the assumed scenario constitutes the most plausible threat scenario for the EU economy. Since banks use heterogeneous models to forecast the stress scenario impact on loan losses and since the EBA does not publish its own respective benchmark parameters, the public cannot fully assess the true severity of the test in terms of its impact on banks’ capital. We argue that both the lack of transparency and the heterogeneity of banks’ practices to forecast stress scenario induced losses considerably weaken the credibility of the stress test and limit its usefulness in supporting market discipline among European banks.

We provide an assessment of the design and calibration of the 2018 EU-wide stress test. The adverse scenario for the 2018 stress test is more severe than for previous stress tests in terms of the assumed GDP decline in the EU area. However, the test is less severe in terms of the losses that banks are expected to incur under the scenario. The adverse scenario has a highly asymmetric impact on different European countries, such that countries with a high degree of trade openness are affected considerably more. It seems unlikely that the assumed scenario constitutes the most plausible threat scenario for the EU economy. Since banks use heterogeneous models to forecast the stress scenario impact on loan losses and since the EBA does not publish its own respective benchmark parameters, the public cannot fully assess the true severity of the test in terms of its impact on banks’ capital. We argue that both the lack of transparency and the heterogeneity of banks’ practices to forecast stress scenario induced losses considerably weaken the credibility of the stress test and limit its usefulness in supporting market discipline among European banks.

Ārējais autors

Rainer HASELMANN , Mark WAHRENBURG