The results from the Bank of England’s 2018 UK bank stress test were always going to look ugly, not because of any new weakness in the banks’ capital levels, nor indeed from any increase in the overall level of stress applied, but because of a new accounting standard, IFRS 9, which pulls loan losses forward into the first two years of the test. Despite this, all the banks passed with an aggregate capital level of 9.2 percent.
IFRS 9 reduces the banks’ capital levels in the early years, but does not, of course, actually increase the real world losses incurred. To mitigate the effects of an accounting change, the PRA has adjusted the pass marks on the test, effectively allowing the banks to draw down on their capital buffers – that is, after all, what they are for. They have also introduced an extended phase-in period for the change. This means that the so called ‘fully loaded’ results will not be the ones that determine capital requirements for some years to come, allowing the banks to mitigate the effects. This is also important in the context of CoCo conversion – it is only under the ‘fully-loaded’ results that Barclays and Lloyd’s converted; however in the real world these banks have five years with which to adjust their balance sheets to ensure this doesn’t happen.
Messy, yes, but not game changing. We take enormous comfort from the fact that the regulators have already sanctioned increased dividends, share buybacks and other forms of capital repayment in the last year, which reemphasises their overall satisfaction in the capital levels in the system. Some of these approvals have been given in the very recent past, when the PRA would already have had a good idea of the test results.