It’s November again and time for the annual Bank of England stress test results.
To be honest I don’t take too much notice of these. The tests are hardly going to show that the UK banking system is not resilient to deep simultaneous recessions in the UK etc etc., although they may show Brexit is the worst thing ever.
But I was sad to see one chart has gone missing.
From 2014 to 2017 the report included a chart of ‘projections for aggregate profits before tax’ for UK banks, which no one with a previous career in back-testing (i.e. me) could resist. How would the actual results compare with the projected ones? One only has to wait a few years to find out.
The results are in the table below. The publication year is the date on which the forecasts were published, the forecast year is the year each forecast was for.
Sadly the four projections (2014-17), turned out to be a bit wrong. The first report in 2014 includes forecasts for 2014 (£20bn), 2015 (30bn) and 2016 (£45bn), definitely upward sloping. However the corresponding actuals are £20bn, £18bn, £10bn, i.e. they are downward sloping. Furthermore, the forecasts themselves have been successively revised downwards. E.g. the 2014 forecast for 2016 starts at £45bn, then falls by £10bn each publication year until it reaches £10bn.
Given the forecasts were meant to support the narrative that the banks could easily rebuild their capital by withholding dividends, this is a bit disappointing.
Fortunately the chart seems to have been dropped from this year’s report, or if not, it has been well hidden.
| Publication year | ||||
| Forecast year | 2014 | 2015 | 2016 | 2017 |
| 2014 | £20bn | £20bn (A) | ||
| 2015 | £30bn | £30bn | £18bn (A) | |
| 2016 | £45bn | £30bn | £20bn | £10bn (A) |
| 2017 | £40bn | £30bn | £20bn | |
| 2018 | £50bn | £45bn | £38bn | |
| 2019 | £50bn | £50bn | ||
| 2020 | £50bn | £55bn |
Table: aggregate pre-tax profit projections by publication year and forecast year