Colm Aodh Manning.
For the past three years, the Bank of England (the Bank) has carried out an annual ‘stress test’ of the UK’s largest banks. To do this, it designed a narrative-based stress scenario in 2014 and 2015. The goal was to determine the banking sector’s resilience to pertinent threats, like recessions or a sharp fall in house prices. However, changing scenarios each year makes it difficult to judge how banks’ overall vulnerability to risks changes over time. Since the crisis we learned that risks build in the good times and capital in the banking system should rise to reflect this. This is why – beginning this year – the Bank has also run an Annual Cyclical Scenario (ACS).
Stress testing is ubiquitous in today’s banking supervision regime. The stress test results are eagerly anticipated and received by the public and can have serious consequences for banks presenting ‘bad’ numbers. The public discussion of the stress scenarios seems to be focussed on their economic meaning (here is an example). The statistical smallprint relating to stress tests receives much less public attention. I pick up two modelling choices for closer inspection:
- Stress scenarios are meant to be point scenarios.
- Stress test results tend to be presented as single values.
I demonstrate that depending on the understanding of the scenario and the representation of the results, there is a wide range of plausible outcomes of a stress test.