EcoFlash

PDF
2
1 December 2018  
United Kingdom: Large UK banks could withstand a major  
shock under certain conditions  
Laure Baquero  
In 2018, the Bank of England (BoE) brought forward the  
publication of its stress test results so that MPs could have  
enough time to consider them before voting on the draft  
(PRA) and analysed by its Financial Policy Committee (FPC),  
which decides whether or not to adjust prudential  
requirements as a result.  
Brexit deal, which was initially scheduled to happen on  
1
UK banks increasingly resilient according to the Bank of  
England  
1
1 December 2018 .  
Evaluated banks started the 2018 BoE’s stress test with an  
aggregate Common Equity Tier 1 (CET1) 3.5 times higher  
than the level seen before the 2008 crisis according, to BoE  
estimates. It has been rising constantly since 2014, which  
means that UK banks have been strengthening their capital  
positions.  
Every year since 2014, the PRA has stress-tested large UK  
2
banks with three objectives: i) to carry out a quantitative and  
prospective analysis of capital levels in the UK banking  
system, ii) to make the BoE accountable for financial stability  
before Parliament and the general public and iii) to restore the  
publics confidence in financial stability.  
The BoE is satisfied with the 2018 results since each of the  
seven banks assessed would keep its CET1 capital above  
the minimum requirement even in the event of a shock  
deemed to be more severe than the 2008/09 crisis, and  
sufficiently severe to cover a disorderly Brexit scenario.  
The stress tests are a way for the BoE to check that bank  
capital levels are sufficient to deal with a stress scenario and  
to adjust capital requirements  in addition to Basel regulatory  
requirements  in the form of a buffer if it deems it necessary  
(see figure 1), either for an individual bank or for the whole  
banking system. The BoE emphasises that these are not tests  
that banks simply pass or fail, but reserves the right to require  
banks to take action to adjust their capital levels or address  
inadequacies in their capital management as the case may  
be.  
Based on these results, along with other data, the BoEs  
Financial Policy Committee maintained the level of its  
countercyclical capital buffer for the whole banking system  
on top of regulatory prudential requirements  at 1%.  
In the light of tests carried out in 2014 and 2015, the BoE  
made some methodological adjustments for tests in the  
following three years (introducing countercyclical tests,  
gradually increasing minimum CET1 thresholds, setting  
individual thresholds for each bank, increasing requirements  
for systemically important banks etc.). The capital buffer  
In 2013, the Bank of England decided to stress-test large UK  
banks every year, in addition to the stress tests carried out  
every two years by the European Banking Authority (EBA).  
The two tests are complementary, if only because they use  
different methodological approaches. In practice, the BoEs  
tests are carried out by its Prudential Regulation Authority  
2
Of which there were initially eight: Barclays, Co-Operative Bank, HSBC,  
1
BoE (20/11/2018), Change of publication date for Financial Stability Report  
Lloyds, Nationwide, RBS, Santander UK and Standard Chartered, but Co-  
and Bank of England stress testing results.  
Operative Bank left the group in 2015 following major restructuring.  
EcoFlash  
economic-research.bnpparibas.com  
2
required by the BoE consists of 4-6 elements, both bank-  
specific and general, and some may change depending on  
stress-test results (see figure 1). In addition to the annual  
stress tests, the PRA has introduced tests every two years to  
gauge the banking systems resilience to other risks that are  
not closely connected with the financial cycle. The first took  
place in 2017. It tested banksability to adjust to an  
environment of consistently low growth and interest rates.  
After that exercise, the PRAs view was that the banks  
assessed would manage to adjust without any major strategic  
changes or risk-taking, although they would have to cut costs  
Breakdown of CET1 regulatory requirements imposed  
by the BoE on UK banks  
PRA buffer  
(bank-specific)  
Buffers adjusted according  
to stress-test results  
Countercyclical capital buffer and sectoral capital  
requirements (system-wide)  
Capital conservation buffer  
(2.5% CET1)  
Buffers set with reference  
to the impact of a failure  
3
Systemic buffer (bank-specific)  
to offset lower margins .  
Pillar 2  
bank-specific)  
Since 2014, the BoE has declared itself to be satisfied overall  
with the stress test results, because the aggregate stressed  
CET1 level is higher than the minimum requirement, bearing  
(
4
in mind that it has risen each year since 2015 and is now  
Minimum requirements  
7
.8%, and that the macro-financial scenario imposed on  
Pillar 1  
banks can vary from one year to the next, except in 2018 (see  
figure 2). Looking at the details, several banks failed the test  
in the first few years because their CET1 capital was lower  
than the minimum requirement or because the PRA thought  
that their equity should be strengthened even though they  
complied with the regulatory thresholds.  
(4.5% CET1)  
Chart 1  
Source: BoE, BNP Paribas  
UK CET1 ratio before and after the BoE stress tests  
1
1
1
1
6
4
2
0
8
6
4
2
0
Hurdle rate  
Management actions and conversion of AT1  
1
4.5%  
Methodological  
change  
1
3.4%  
Co-Operative  
Bank no longer  
12.6%  
included in the group  
1.2%  
1
1
0.0%  
9
.7%  
8
.8%  
8
.3%  
7
.5%  
.3%  
7.6%  
7.2%  
7
.3%  
7.0%  
7
6
.5%  
BoE 2014  
Assuming a IFRS 9 transition period  
BoE 2015  
BoE 2017  
Identical macro scenario  
Source: BoE, BNP Paribas  
BoE 2018*  
BoE 2016  
*
Chart 2  
3
Bank of England (2017), Stress testing the UK banking system: 2017 results.  
The minimum threshold in aggregate terms corresponds to the average  
4
minimum threshold required of banks.  
EcoFlash  
economic-research.bnpparibas.com  
3
The PRA looked closely at changes in those banksCET1  
levels and their Tier 1 leverage ratios throughout the year  
before unveiling the stress-test results, and before urging  
certain banks to increase their equity if necessary. In 72% of  
cases, the PRA took the view that banks reached the  
minimum thresholds and had sufficient equity. That has been  
the case every year since 2014 for HSBC, Nationwide and  
Santander UK (see diagram 1). From 2017 onwards, the  
seven banks assessed have had enough equity to withstand  
the stress scenario devised by the BoE. However, that  
success relies on certain assumptions made by the BoE: i)  
that balance sheets are not dynamic, i.e. banks are  
authorised to take action to reorganise their activities to  
Success almost across the board for UK banks in the  
BoEs 2018 stress test  
6
In 2018 , the BoE tested the seven large UK banks using a  
stress scenario it regarded as more severe than the 2008  
7
crisis . It was more severe in terms of global GDP, jobs and  
residential real-estate prices in the UK, but not in terms of UK  
GDP. The BoE also regarded the scenario as sufficiently  
severe to cover a disorderlyBrexit and the resulting 4  
percentage-point (pp) fall in the CET1 ratio. The stressed  
CET1 ratio remains higher than the PRAs minimum threshold  
for each of the seven banks, even without any conversion of  
AT1 capital.  
5
absorb the shock and ii) that alternative Tier 1 (AT1)  
instruments can be converted in order to shore up equity if it  
is excessively affected by major stress.  
Summary of the BoE’s stress-test results between 2014 and 2018 in terms of CET1*  
Stress test  
Minimum threshold  
reached  
Minimum threshold  
not reached  
Equity increased during the year of the test  
at the bank's initiative  
Banks asked by the PRA to  
increase their equity  
Equity deemed  
insufficient by the PRA  
Equity deemed sufficient  
by the PRA  
Co-operative Bank (2014)  
RBS (2016)  
Loyds (2014)  
RBS (2015, 2017*)  
Standard Chartered (2015)  
Barclays (2015 and 2017*)  
RBS (2014)  
Barclays (2014, 2015, 2018)  
HSBC (2014 to 2018)  
Standard Chartered (2016)  
Lloyds (2015, 2016 to 2018)  
Nationwide (2014 to 2018)  
RBS (2018)  
Santander UK (2014 to 2018)  
Standard Chartered (2014, 2017,  
2018)  
*
In 2017, Barclays and RBS had enough equity to pass the test in the sense of reaching their  
basic hurdle rate, but not taking into account their systemically important status.  
Diagram  
Source: BoE, BNP Paribas  
6
Bank of England (2018), Financial Stability Report, Issue No. 44.  
GDP falling by 2.4% globally, 1.2% in China and 4.7% in the UK; the  
7
unemployment rate rising to 9.5%; real-estate prices falling 33% in the  
residential sector and 40% in the commercial sector; sterling falling 27% and the  
BoE's base rate rising to 4%.  
5
Depending on the year, this action could include reducing dividend payments,  
cutting staff levels and reducing the lending supply.  
EcoFlash  
economic-research.bnpparibas.com  
4
The BoE sets a minimum threshold specific to each bank,  
depending on whether it is regarded as systemically important  
or not, and also now taking into account its domestic  
exposure. In 2018, the minimum thresholds ranged between  
In 2018, most of the fall in CET1 capital caused by the BoE  
stress test was down to loan impairment charges (see  
figure 3). They were the result of lower loan production, lower  
asset prices and higher interest rates, bearing in mind that  
half of the increase in the cost of risk is connected to banks’  
exposure to the UK economy. Impairment charges were  
already the main factor depressing CET1 capital in previous  
stress tests, with an increasingly negative effect each year.  
Compared with 2017, the trend was made worse in 2018 by  
the adoption of IFRS 9 and its forward looking impairment  
model, although the effect was partly offset by the transition  
period designed for that purpose. Another factor pushing up  
loan impairment charges and of concern to the FPC is the  
rapid growth in leveraged loans, reflecting the deteriorating  
quality of loans granted in the broad market.  
6
.7% and 8.5% (see figure 4), compared with the 7% required  
by Basel III, excluding transitional provisions on the same  
8
date . The PRA and the PFC concluded from these results  
that these seven banks, which together grant 80% of loans in  
the UK, are resilient enough to withstand a major shock while  
continuing to fulfil their role of financing the real economy.  
Based on these results, along with other information on the  
9
UKs financial stability , the PRA did not required any increase  
in equity and the FPC kept its countercyclical buffer at 1%.  
Conversely, it was increased from 0% to 0.5% in June 2017  
and then from 0.5% to 1% in November 2017.  
The tricky comparison between the stress tests of BoE  
and those of the EBA  
It is important to note that these results are only valid  
assuming that the IFRS 9 transition period is used. The  
adoption of IFRS 9 and its forward-looking impairment model  
increases bankscost of risk, both at inception and during  
economic downturns. The transition period allows banks to  
Unlike the BoE, the EBA carries out stress tests every two  
years. The results of its 2018 tests appeared in early  
November and concern 48 European banks, including four of  
the seven UK banks assessed by the BoE: Barclays, HSBC,  
Lloyds and RBS. It is therefore tempting to compare the  
results of the two tests, assuming that, according to the EBAs  
stress test, the UK is bottom of the table in terms of stressed  
CET1 and its ratio is lower than the average of the 15  
European countries tested. The same is true of the four UK  
banks tested by the EBA, which rank between 27th and 48th  
depending on the assumption made regarding the IFRS 9  
transition period. However, any comparison between the BoE  
and EBAs stress tests is made more difficult by several major  
methodological differences.  
1
0
smooth out the cost of this change in accounting standards .  
If the transition period is not used, UK banks still pass the test  
but less convincingly and one of them (Barclays) would have  
to convert AT1 capital. However, the PRA states that the  
minimum thresholds based on the adoption of IFRS 9 with no  
transition period are hypothetical and need to be reworked.  
The full integration of IFRS 9 adoption is one of the  
methodological changes that is likely to take place from 2019  
onwards. The other major change is likely to relate to the  
framework imposed by the Vickers legislation on the UK  
banking sector (banks will have to ring-fence their UK retail  
banking activities, including setting up autonomous  
governance in the management of prudential ratios).  
Firstly, the EBAs 2018 test did not involve any minimum  
threshold to be attained, unlike the BoEs method.  
Factors behind movements in the aggregate CET1  
Secondly, while the BoE and EBA both reported results with  
and without an IFRS  
9 transition period, the BoEs  
ratio following the BoE’s 2018 test (pp)  
communication is focused on results with a transition period,  
since it promised to give banks the full benefit of the  
transitional period, including in its stress tests. Although their  
assessments of the cost of adopting IFRS 9 are fairly close in  
aggregate terms (0.10% for the EBA vs. 13 pp for the BoE),  
they sometime vary more widely for individual banks.  
In addition, the EBA assumes static balance sheets, while the  
BoE makes the opposite assumption. The EBAs approach  
has the advantage of being simpler but the disadvantage of  
being somehow unrealistic. For its part, the BoE admits that  
banks can take strategic management actions to deal with a  
major stress episode. If it is likely that banks would not take  
the measures assumed in the BoE’s stress tests when truly  
faced with significant macro-financial turbulence, this room for  
manoeuvring remains more acceptable than to abstract from  
it.  
The EBA also does not assume the conversion of AT1  
instruments to shore up equity if necessary, unlike the BoE. In  
that respect, the BoE has noted that UK banks used AT1  
instruments to increase their absorption capacity, and insists  
that the investors concerned should be well aware that those  
instruments could be converted if necessary.  
Chart 3  
Source: BoE, BNP Paribas  
8
Officially, Basel III provides for a seven-year transition period to give banks  
time to meet its capital requirements. The 7% threshold must be met by 2019,  
whereas the minimum requirement in 2018 is 6.4%.  
Bank of England (2018), Financial Stability Report, Issue No. 44  
0
T. Humblot (2018), IFRS 9 first time adoption: Significant cost differentials  
Finally, the macro-financial scenarios imposed by the EBA  
and BoE differ markedly, which strictly speaking means that  
the results of their stress tests cannot be compared.  
9
1
amongst banks, BNP Paribas.  
EcoFlash  
economic-research.bnpparibas.com  
5
At most, it is possible to compare the EBA and BoE stress  
tests by looking at the BoEs results using its macro-financial  
scenario, and the following two sets of assumptions, both put  
forward by the BoE:  
The first set of assumptions (A) gives better results than the  
second (B). Using the second set, Barclays, HSBC and  
Lloyds would fail the test because their CET1 ratios would be  
below the BoEs minimum requirement (see figure 4).  
A. with an IFRS 9 transition period, non-dynamic balance  
sheets and use of AT1 conversion. This set of  
assumptions is the one on which the BoEs  
communication is usually based.  
It would be risky to conclude that the BoE retains more  
favorable assumptions for the success of its stress tests.  
Above all, the more realistic nature of its assumptions makes  
them more operational.  
B. without an IFRS 9 transition period, static balance  
sheets and no use of AT1 conversion. This set of  
assumptions is one of the two put forward by the EBA.  
CET1 ratios before and after the BoE’s 2018 stress test in the UK (aggregate) and for each bank according to  
the two sets of assumptions A and B (%)  
32  
30  
28  
26  
24  
22  
20  
18  
16  
14  
12  
10  
8
6
4
2
0
Aggregate  
Barclays  
HSBC  
Lloyds  
Nationwide  
RBS  
Santander UK  
Standard  
Chartered  
CET1 ratios before the test, set of assumptions A (transition to IFRS 9, non-static balance sheet and use of AT1)  
CET1 ratios after the test, set of assumptions A (transition to IFRS 9, non-static balance sheet and use of AT1)  
CET1 ratios before the test, set of assumptions B (no transition to IFRS 9, static balance sheet and no use of AT1)  
CET1 ratios after the test, set of assumptions B (no transition to IFRS 9, static balance sheet and no use of AT1)  
Hurdle rate  
Chart 4  
Source: BoE, BNP Paribas .  
QUI SOMMES-NOUS ? Trois équipes d'économistes (économies OCDE, économies émergentes et risque pays, économie bancaire) forment la Direction des Etudes Economiques de BNP Paribas.
Ce site présente leurs analyses.
Le site contient 1994 articles et 547 vidéos