11 Capital and loss absorbency |
231. The ICB recommended a package of reforms to
make the banking system more stable and make banks easier to resolve.
The ring-fence is a part of this package, but it is not sufficient,
as discussed in Chapter 6. In addition to the ring-fence, the
success of the proposed reforms to the banking system hinges on
whether capital and loss absorbency measures can be put in place
which reduce the likelihood of banks failing, and ensure that,
if banks do fail, the losses from failure are better aligned with
the rewards for success.
232. As set out in Chapter 2, the special characteristics
of banks necessitate treating them differently from other companies
when they are at or close to insolvency. The special resolution
regime created in the Banking Act 2009 provides some alternative
tools for dealing with failing banks, but as discussed in Chapter
2 the failure of a large and complex bank would pose a considerable
challenge for this regime and could pose an unacceptable risk
to public funds.
233. This means that a further tool is neededa
so-called 'bail-in regime'. This is a set of legal changes that
bestow powers on the resolution authority to put a bank through
a special, ultra-fast form of insolvency procedure, imposing losses
on bank creditors over a weekend rather than over weeks or months
and maintaining the continuity of provision of essential services.
This approach was recommended by the ICB and accepted by the Government.
234. There was broad agreement among witnesses that
further reforms were necessary to ensure that bank creditors bear
losses to preclude the need for the taxpayer to bail out banks
in future. Andrew Bailey explained that:
if you strip it back, all you are really saying about
bail-in is that you are asking to deal with the resolution of
a bank as you do any other company. Companies get refinanced by
their creditors. The point about banks is that, because of the
confidence issue, you have to do it very quickly. You only get
a weekend to do it. Therefore, you need legislation to effect
235. The Bank of England now sees bail-in a central
part of its strategy for dealing with failing banks:
[T]he size and complexity of the books of most global
wholesale banks greatly increases the challenge in rapidly separating
the critical economic functions in this manner without causing
severe systemic disruption... This is what led to the development
of the concept of bail-in resolution strategies, in order to ensure
that unsecured creditors are exposed to loss without having over
a resolution weekend to split up a bank into critical and other
parts that go into liquidation.
effective and credible bail-in tool would represent a major step
towards eliminating the implicit guarantee and ensuring that the
costs of resolving a failing bank are not borne by the taxpayer.
It is notable that bail-in is at the heart of the resolution strategies
currently being designed for large systemically important banks,
and will remain important even after the ring-fence is introduced.
237. One concern about relying on bail-in to make
banks resolvable is whether it is realistic to expect that in
a crisis the authorities will be willing to exercise their powers
to impose losses on creditors. Professor Rosa Lastra warned that
bail-in still needed to be shown to be a credible tool:
bail-in is a very useful instrument and the only
problem with it is that it still needs to be tested. It needs
to pass the market test of credibility and of not being stigmatised.
[...] The [European] Commission, when it was negotiating the Recovery
and Resolution Directive, had a group of legal experts discussing
the bail-in. It is clear that the 'bail-inable' instruments when
bailing in, the trigger points, the credibility and the stigma
are issues that need to be resolved.
Jessica Ground also expressed the concern that bail-in
"is one of those things that looks fantastic on paper",
but that when it came to the choice of "pushing the button",
authorities might not be willing to take the risk of triggering
a bail-in in a crisis because of the risk of contagion. She added:
"Bail-in [bonds] could be very good investments, but whether
they are that helpful at the end of the day in dividing up risk
and assigning the risk to that group of security holders, I remain
less convinced about."
238. Lord Turner noted that regulators were more
likely to exercise their bail-in powers when this does not involve
imposing losses on other banks:
this stuff [bail-inable debt] also has to be held
outside the banking system. If in those circumstances we press
the bail-in button, but we are worried that all this bail-in debt
is held by another bank, we will never be willing to do it. We
will be terrified that we are just setting off a domino set which
is going to keel over. That is a missing bit of it.
239. To make bail-in powers more usable and more
credible, it is proposed that banks must hold a certain amount
of debt that can easily absorb losses; the next section considers
how this might be implemented. As Sir John Vickers explained:
Can one move to a situation where it is absolutely
certain that the bondholders would bear loss? I think that total
certainty is, perhaps, not to be had, but I believe that one can
increase enormously the chance that bondholders would bear loss,
and our proposals were crafted with a view to maximising that
probability. You need the bail-in power of the regulator and a
significant or substantial slab of such debt [...] It needs to
be unsecured debt with appropriate maturity and the rest.
240. Some witnesses expressed concern that there
would not be sufficient demand for this kind of debt. For example,
Jessica Ground said:
A lot of traditional fixed-income investors are very
worried about them because of the potential for bail-in. A lot
of equity investors are not natural holders of these [bail-in
bonds], because your upside is capped. This is uncharted territory,
and the market, as far as I understand it, is very small.
John Grout agreed, indicating that he struggled to
identify potential buyers of bail-in debt:
It looks as though insurers and pension schemes may
not be suitable holders of bail-in bonds. That leaves you with
hedge funds, wealthy families, sovereign wealth fundsI
sort of run out when I get beyond that.
241. Others were more sanguine about the size of
the market for loss-absorbing debt. Erkki Liikanen told us that
investors would buy the debt in the right conditions,
while Paul Tucker said that:
The point you make about pension funds, and one could
say the same about life companies, is whether they should be allowed
to own regular senior unsecured bonds issued by banks. I do not
see why not, as long as they did not have concentrated exposures.
They own corporate bonds, and they sometimes default. Insurance
companies and pension funds do not only hold risk-free assets.
They lose money on their bond holdings, and their equity portfolios
go up and down in value.
It is quite important not to think about the world
investor base as completely averse to default risk. When I have
talked to some of the biggest asset managers in the world and
have put this to them as a question, they have said that they
would rather be able to take losses on their bank bond holdings
if that helped to insulate them from the mayhem caused by the
latest financial crisis, which devastated the value of their overall
remain about the design of a bail-in regime and whether it will
provide confidence that the authorities would actually use their
powers in the event of a crisis. The new tool risks being of particularly
limited utility if the authorities were required to impose losses
beyond the holders of specifically "bail-inable" debt
and move up the chain to, say, corporate depositors. The legal
and economic implications of bailing in a bank's creditors will
never be known until it is tried for the first time under stressed
conditions, and politicians and regulators will always face pressure
to incur the better-understood costs of a taxpayer bailout instead.
It should be a requirement that bail-inable debt is held outside
the banking system, to reduce contagion risks within the banking
system. The regulator should make early proposals on how best
to accomplish this. Uncertainty about the size and nature of market
for loss-absorbing debt will also mean that doubts will remain
over whether bail-in will function as intended and what its costs
will be. Parliament will need assurance that bail-in is not a
paper tiger, as will the markets. The Commission recommends accordingly
that the Bank of England be subject to a statutory requirement
under the new legislation to produce an annual report to Parliament
on the development and subsequent operation of bail-in to assist
in assessment of its feasibility, which should be required to
cover in particular:
- The quantity
of issued debt with characteristics which make it easily subject
to bail in;
- Whether bail-inable debt is
being issued out of the correct corporate entity within a banking
group to facilitate the preferred bail-in strategy;
- The distribution of holdings
of bail-inable bank debt within the rest of the financial system;
- The feasibility of mechanisms
for bailing in creditors other than long-term unsecured bonds,
such as corporate depositors, uninsured household depositors and
- Progress towards addressing
international legal barriers to the recognition of bail-in actions.
243. It is expected that a bail-in regime will be
introduced when the EU Directive on Recovery and Resolution (the
'RRD'), which is currently in draft, is implemented in UK law.
Several respondents noted the desirability of seeking agreement
on bail-in at a European or international level. For example,
Santander told us that:
Like other UK banks, Santander UK raises its funding
in international markets. Therefore, a UK-only statutory bail-in
power would not give clarity to market participants, many of whom
would be operating outside of the UK's jurisdiction. A bail-in
instrument of the kind described in the Government's White Paper
can only be viable if it is internationally agreed and implemented.
EU law also means that it would currently be difficult
for the UK to implement a bail-in regime unilaterally, because
the RRD "will remove some impediments to resolution arising
currently from other EU directives".
244. Martin Taylor urged the UK to go it alone should
international agreement founder:
I would certainly recommend to Parliament that, if
for any reason the European supervisors lose their nerve on bail-in
debt, Britain itself does something. There ought to be an international
standard there. A common international standard is also very desirable.
If we can do that through the European work, that would be the
best answer for that.
Commission supports the Government's endeavours to implement a
bail-in regime in the UK. The Government should also continue
to negotiate for a broad bail-in power to be applied across the
EU. Bail-in is an important tool for resolving bank failures in
a way that prevents the huge costs. The Commission is concerned
at the risk that the development of such a tool might be delayed
or watered down through negotiations at EU level and, given the
size of the financial services sector relative to the UK economy,
the Commission believes the Government should act at a UK level
in the event of EU discussions not resulting in the desired protection
for the taxpayer that bail-in aims to ensure. The Commission recommends
that the Government make provision in the forthcoming legislation
for bail-in powers at national level which could come into force
if the EU proposals were delayed or inadequate, on the understanding
that negotiations at European level would need to secure the subsequent
removal of any existing or prospective European legal obstacles
to the use of a more wider-ranging power at national level.
THE MAIN PLAC REQUIREMENTS
246. For a bail-in regime to work, it must be possible
for the authorities to impose losses on a bank's creditors without
excessive disruption to that bank's operations or to the rest
of the market. A deciding factor in whether this is the case is
the nature of the bank's liabilities, and in particular whether
there is enough unsecured debt of sufficiently long term to cover
bail-in requirements. If long-term debt is available to absorb
losses, short-term creditors are less likely to risk a run on
the bank by demanding their money back.
247. To this end, the ICB recommended that large
ring-fenced banks and UK-headquartered global banks issue the
equivalent of at least 17 per cent of their risk-weighted assets
(RWAs) in the form of primary loss-absorbing capacity (PLAC).
RWAs are calculated by multiplying each type of asset on a bank's
balance sheetgovernment bonds, mortgages, derivatives and
so onby a 'risk-weight' (for example, 0 per cent for assets
considered to be very safe, like OECD government bonds; and 100
per cent for assets judged to be riskier, such corporate loans),
then adding them up. When calculating the 17 per cent requirement,
the denominator is the total of RWAs. Therefore the riskier the
assets on a bank's balance sheet, the more PLAC a bank needs to
issue to achieve the 17 per cent.
248. PLAC is defined by the ICB as:
those liabilities that can be regarded as constituting
the best quality loss absorbing capacity. 'Primary loss-absorbing
capacity' is made up of (i) equity; (ii) non-equity capital; and
(iii) to reflect the fact that short-term liabilities are less
reliable as loss-absorbing capacity, those bail-in bonds with
a remaining term of at least 12 months.
Banks affected by the ICB's recommendations on capital
would also have minimum requirements on the components (i) and
(ii) of PLAC above, which would mean that they would have to hold
at most 6.5 per cent of RWAs as bail-in bonds to bring total PLAC
up to 17 per cent, and in fact more likely 3-3.5 per cent given
that most large banks will be subject to additional requirements
on equity. The Government has agreed with the ICB's PLAC recommendation,
defining PLAC as "regulatory capital and subordinated debt
and senior unsecured debt with at least twelve months' term remaining
and which the resolution authorities are confident could be bailed
249. The Government expects the forthcoming EU Directive
on Recovery and Resolution to include a minimum eligible liabilities
requirement as an adjunct to the bail-in powers it mandates, which
would have a similar effect to a PLAC requirement.
In anticipation of this, the draft Bill inserts a new section
142J of FSMA which would give the Government powers to instruct
the regulator how to impose debt requirements on banks.
PROPOSED EXEMPTION FROM PLAC REQUIREMENTS
250. The logic for making UK-headquartered banks
hold PLAC over and above the internationally agreed capital requirements
is that this provides an extra cushion on which the UK authorities
could impose losses if such a bank fails and needs resolving.
The Government has argued that it would not be appropriate to
impose this additional requirement on the overseas operations
of UK headquartered banks "where these do not pose a risk
to UK and/or EEA financial stability",
because the UK authorities would not be responsible for resolving
such entities. They argue that this would be "disproportionate",
presumably on the grounds that it would impose an unnecessary
cost on such operations that other international banks would not
face. The Chancellor of the Exchequer also warned that imposing
such a requirement would risk creating the perception that the
UK authorities would stand behind overseas operations of UK banks:
Let us imagine you have a large bank in the UK that
has a very large operation in Hong Kong, say. If you say that
that bank has to provide loss-absorbing capacity against the failure
of the Hong Kong operation then, by the way, the implication is
that if the Hong Kong operation fails, we are going to be on the
hook for it. I want it to be very clear we are not on the hook
for the failure of the Hong Kong operation.
In addition to using the powers in proposed section
142J to define PLAC, the Government also therefore intends to
use them to set out the conditions under which UK-headquartered
banks are exempted from holding PLAC in excess of international
minima against non-EEA assets.
251. The PLAC exemption was not considered in the
ICB's final report, although both Martin Taylor and Sir John Vickers
said in their evidence that they accepted the logic of the Government's
However, they did warn that care should be taken in designing
the conditions for exemption, Martin Taylor warning that "we
wait to see whether that is done properly. I would hold feet to
the fire on that one, if I were in Parliament."
EXEMPTION FROM PLAC REQUIREMENTS:
BURDEN OF PROOF
252. Sir John Vickers said that, while the correct
test for a PLAC exemption was whether a bank's overseas operations
were resolvable without posing a risk to UK financial stability
or the UK taxpayer, the burden of proof for demonstrating whether
that this test had been met ought to lie with the bank rather
than the regulator:
In its December response to our report, the Government
made the totally reasonable point that if a bank can demonstrate
that it was resolvable, etc [...] it would be disproportionate
to place that requirement on the bank. The logic of that seems
to me to be impeccable. But note that the onus of proof was on
the bank. In the June White Paper, the Government seemed to have
changed their position to the onus of proof being on the regulator.
I thought that was an unwise step.
In the text preceding the draft legislation in the
October document, the Government seem to be somewhere in between
their December and June positions. I am not completely sure how
to interpret it where it talks about the pros and cons and states
that a balance needs to be struck. Again, that is a hard thing
to disagree with, but I would flag it up as something to be alert
to in the secondary legislation.
253. Barclays also argued that involvement of the
host regulator (the regulator in the country where a bank is conducting
operations) was key:
Any exemption of assets held in overseas operations
must be subject to very stringent tests which require the 'host'
regulator to state unequivocally that the UK is in no way liable
if the entity of the relevant bank needs to be independently resolved
and that, in such circumstances, the parent bank cannot take steps
to rescue that unit that would in any way jeopardise the overall
254. Standard Chartered, a UK-headquartered bank
which conducts the overwhelming majority of its operations abroad,
argued that the global PLAC exemption should be at the discretion
of the regulator:
there are instances where the application of the
exemption renders the orderly wind-down of an international bank's
operations in resolution potentially unworkable. For instance
where the exemption has an impact on the UK's reputation or the
ability of the home regulator/resolution authority to discharge
its obligations to provide a coordinated resolution of a troubled
bank. Our view is that the exemption should be available for application
by the resolution authority on a discretionary basis where these
issues are addressed.
Standard Chartered also indicated that it would most
likely not take advantage of any exemption, preferring instead
a UK-based resolution plan. It stated that:
it is logical to give the regulatory authorities
scope to exempt banks from applying the PLAC requirement to their
international activities where they can demonstrate their failure
would not pose a risk to the UK's financial stability. Some international
banks will choose to take advantage of the exemption whilst others
may not. The choice is likely to be partly dependent on the approach
to resolution planning each individual bank develops with the
Financial Services Authority and Bank of England. For Standard
Chartered, we favour a whole bank resolution approach since we
believe this would ensure the optimal outcome in the extremely
unlikely event that we were to fail.
255. Witnesses from the Bank of England and the FSA
agreed that the burden of proof should be on the bank seeking
the exemption. Lord Turner said that:
we are very keenthis is an important point
for us on the Billthat the PRA should not be placed in
a position where it has to prove that this exemption from group
PLAC creates a risk to the taxpayer; the burden of proof, from
our point of view, must be the other way around.
Lord Turner set out two additional conditions he
believed should be required before an exemption is granted, using
the example of HSBC:
Secondly, the resolution plan should be agreed with
the resolution authorities across the world. We ought to be agreeing
it with the Hong Kong Monetary Authority and the Monetary Authority
of Singapore. Thirdly, the fact that those authorities have agreed
a regional break-up, rather than a group break-up, should be publicly
known and publicised so that the creditors of that bank in the
UK and in Hong Kong or Singapore are aware of that.
He also argued that under such conditions, it was
quite likely that overseas regulators would themselves require
additional capital to be held against local operations, once it
was fully clear that they would bear responsibility for resolution
in the event of a failure:
the banks involved would not get any benefit from
the proposed change to UK law, because I can absolutely predict
that we will demand at least 17 per cent of primary loss-absorbing
capacity, and so will the HKMA in Hong Kong, the MAS in Singapore
and the US authorities.
256. Andy Haldane and Paul Tucker noted that the
concept of agreeing a resolution plan with overseas authorities
was already part of the international recovery and resolution
planning process under the Financial Stability Board. Paul Tucker
said: "the UK authorities have an international obligation
to ensure that the worldwide group of any UK-domiciled banking
group is resolvable. That is not the same as saying we must be
257. The Chancellor of the Exchequer did not clarify
the Government's position fully, but emphasised that, once a decision
had been made, the regulator should not be able to change its
position without demonstrating its reasons for doing so:
We are clear that the firm will have to assure us
that the costs of the non-EEA bits of it, were they to fail, would
not fall on to the UK taxpayer. [...] However, once they have
assured us, what I don't want to have is, hanging over some extremely
large and important institutions in our country, a constant threat
that that judgment might change, that the regulator might suddenly
change its judgment, and that has an almost overnight impact on
the business. I think once the firm has assured the regulator,
it is then the obligation on the regulator, should it change its
mind, to demonstrate that it is right in changing its mind.
CONCLUSIONS ON THE PLAC EXEMPTION
from PLAC increase the risk that, in a crisis, the UK would need
to intervene in respect of overseas operations of a UK-based bank,
but would lack the level of PLAC necessary to shield the taxpayer.
The Commission recommends that the secondary legislation to be
made under to section 142J of the draft Bill place the burden
of proof for any exemption from PLAC requirements on the bank
seeking the exemption, rather than on the regulator. This would
mean that the regulator would only grant an exemption if a bank
had demonstrated to the regulator's satisfaction that there was
no risk to stability, rather than merely if the regulator could
not show that a risk existed, providing a greater level of protection
to the taxpayer. This should include the bank showing that the
resolution authorities in the areas in which they operate outside
the EEA would assume lead responsibility for resolving the operations
in those overseas territories in the event of the bank's failure,
in order to protect the UK taxpayer. The decision on whether to
grant an exemption should be made by the regulator with reference
to clear objectives, although in all cases it will need to involve
an exercise of judgment by the regulator. Decisions should be
subject to the same review and appeals processes as any other
decision by the regulator. The existence of exemptions should
be publicly disclosed. It will also be important for the regulator
to monitor the implications of exemptions in the case of each
firm affected on an ongoing basis. We would expect this monitoring
to be the subject of regular review by a strengthened Supervisory
Board of the Bank of England introduced in accordance with the
recommendations of the Treasury Committee.
ACCOUNTABILITY FOR USE OF POWERS
RELATING TO LOSS ABSORBENCY
259. Clause 4 of the draft Bill introduces a new
provision in proposed section 142J(4(d)) of FSMA which, after
an enabling order has been passed, confers a power on the Treasury
to issue further directions to the regulator in relation to loss
absorbency. The House of Lords Delegated Powers and Regulatory
Reform Committee noted that such an order would:
confer power on the Treasury to give directions,
not subject to any Parliamentary control, to the regulators. This
is mentioned, but not justified, [... in] the [Treasury's delegated
powers] memorandum. We normally require a full and convincing
justification for power which may be used to circumvent Parliamentary
control and we were not convinced on this occasion.
260. It is also notable that an order made under
section 142J(4(c)) could be used by the Treasury to "require
the regulator to consult the Treasury before imposing a requirement
in accordance with the order in a particular case".
This would appear to create the potential for the Treasury to
give itself a role in setting PLAC requirements for individual
firms, which if used would appear to conflict with the objective
of an independent regulator.
261. When asked why there were no principles in the
draft Bill to guide the use of the powers in section 142J, the
Chancellor of the Exchequer replied:
This is a significant economic decision; it is right
that Parliament is involved in that decision. It will be set out
in the legislation [...] but I think we are entitled as a Parliament
to say, first of all, that PLAC will be applied to certain institutions
and, secondly, how we define that PLAC. I am clear it has to be
equity or tier 1, tier 2 or unsecured longer-term debt. We have
set out in the White Paper what we think those requirements are.
But again, it comes down to a broader question: do you give complete
freedom to the regulators to make decisions that have a very real
economic impact on our constituents, or do you ask Parliament
to be involved and consulted on that decision? Here we have tried
to get the balance right, where Parliament will prescribe certain
minimum requirements and hard definitions, and then it will be
up to the regulator to go ahead and impose them.
broad, largely unconstrained powers contained in proposed section
142J of FSMA could be used by the Treasury to set a framework
which removes the regulator's discretion over whether to grant
a PLAC exemption. There is also a possibility that the Treasury
could use the power to intervene in individual decisions on exemptions
from PLAC requirements. If this was used to overrule the regulator's
decision on individual cases, this would be a highly inappropriate
263. The Commission
accepts that the Treasury should have certain powers to implement
the PLAC requirements, and that secondary legislation is the appropriate
vehicle: primary legislation is not appropriate for such technical
matters, and the changes will in some cases be too important to
be left solely to the discretion of the regulator. However, as
drafted, these powers are extremely wide-ranging, are subject
only to the negative resolution procedure, and need not be deployed
with reference to any particular policy objectives. Furthermore,
an order made under these provisions may confer a general power
to give further directions to the regulator without further parliamentary
oversight. This places an unacceptable level of unconstrained
power in the hands of the Treasury. The Commission recommends
- the Bill require
the powers of direction the Government acquires under proposed
section 142J to be exercised with reference to policy objectives
stated on the face of the statutory instrument which grants those
- the order-making powers under
proposed section 142J be subject to the affirmative resolution
procedure, rather than the negative resolution procedure, to ensure
a greater degree of parliamentary oversight; and
- the power under proposed section
142J(4)(d) to "confer power on the Treasury to issue directions
to the regulator as to specified matters" be removed from
the draft Bill altogether.
The Commission also notes that the
remaining powers of the Treasury to direct the regulator in relation
to the implementation of the PLAC requirements will need very
THE CONCEPT OF DEPOSITOR PREFERENCE
264. When a company goes into insolvency, its creditors
queue up to be paid out of the assets remaining in the firm. Preferring
one group of creditors means moving them toward the front of this
queue. Depositor preference is therefore relative: moving one
creditor towards the front necessarily pushes others back. Depositor
preference always favouring some depositors at the expense of
others. Depositor preference can, in principle, be applied to
certain categories of depositors (for example, households or SMEs)
or to certain deposits (such as deposits insured through the Financial
Services Compensation Scheme (FSCS)).
265. The ICB recommended in their final report that
"in insolvency (and so also in resolution), all insured depositors
should rank ahead of other creditors to the extent that those
creditors are either unsecured or only secured with a floating
In the White Paper preceding the draft Bill, the Government consulted
on whether certain other categories of debts
should be preferred alongside insured deposits.
Following the consultation, the Government has decided to propose
that only insured deposits should be preferred and proposals to
give effect to this are set out in Clauses 7 and 8 of the draft
266. In the UK, the deposits of households and small
businesses up to £85,000 are insured by the Financial Services
Compensation Scheme (FSCS), a body funded by the financial services
industry with a backstop from the Government. As the Institute
for Chartered Accountants pointed out, when only insured deposits
are preferred this does not benefit insured depositors themselves:
The Financial Services Compensation Scheme protects
individual deposits up to £85,000. So the proposals do not
affect the position of most individuals. Rather, insured-depositor
preference protects those who stand behind the FSCS schemethe
banks and, in some scenarios, HM Treasury.
The uninsured depositor loses at the expense of the
ARGUMENTS FOR INSURED DEPOSITOR
267. Uninsured creditors would currently bear losses
at the same rate as insured deposits if a bank failed. However,
as the ICB final report pointed out, "insured depositors
are not well-placed to exert market discipline on banks, and in
any case have no incentive to do so".
By moving insured deposits ahead of uninsured depositors in the
queue, depositor preference as proposed in the draft Bill concentrates
losses on uninsured creditors, "many of whom are better able
to exert such discipline by demanding higher returns if a bank
pursues riskier activities".
The measure is thus intended to create an incentive on uninsured
depositors to exercise an influence on the risk profile of banks.
268. If a bank does fail then any costs borne by
the FSCSfor example in compensating depositorsare
passed on to other banks through the FSCS levy. The ICB final
This requires safe, well-run banks that survive a
crisis to pay for the failure of risky banks (perhaps over a number
of years), and in so doing acts as a channel for contagion. If
surviving banks are unable to bear these costs, they will ultimately
fall on the taxpayer.
The risk to the taxpayer arises because in practice
the FSCS and its members do not have the financial capacity to
fund a sizeable resolution up front, so instead are likely to
be reliant on a loan from the Treasury. As of April 2012, the
FSCS still had obligations to repay the Treasury almost £18bn
as a result of bank failures in the recent crisis where it made
payouts or funded the transfer of deposits in resolution.
Insured depositor preference could reduce the risk to public funds
from any loans made to the FSCS by making it more likely that
full recovery will be made from the failed bank.
EXTENDING DEPOSITOR PREFERENCE BEYOND
269. Giving preference to insured deposits will result
in higher rates of loss in insolvency for other liabilities which
have been pushed down the creditor hierarchy. As an illustration,
in the case of a bank with £10bn of insured deposits and
£10bn of other deposits, these would currently rank equally
in insolvency. If the bank fails and there are £4bn of losses
left over after equity and any subordinated debt has been wiped
out, then both the FSCS faces a loss of £2bn and uninsured
depositors face a loss of £2bn20 per cent of their
claim. With insured depositor preference, all £4bn of losses
fall on uninsured depositors who lose 40 per cent of their claim,
while the FSCS faces no loss at all.
270. Some witnesses argued for extending preference
beyond insured deposits, to categories such as all retail deposits
or charities' deposits. A group of charity representatives argued
for extending preference to charities' deposits on the grounds
that the current proposals would place a greater and inappropriate
burden on charities:
The vast majority of charities manage their finances
and risk extremely well, yet most simply do not have the same
level of technical banking knowledge and expertise as other creditors
of comparable size. Managing increased banking risk is extremely
resource-heavy and investing in up-to-the-minute treasury management
diverts funds away from frontline services. It is therefore difficult
to justify using substantial charitable funds for this purpose,
particularly given the challenging funding environment.
They added that the nature of charity funding and
banking was unique and that "charities typically hold large
amounts on deposit (approximately £18bn across the sector)
as they require quick and easy access to cash".
271. Nationwide stated that it had "consistently
argued that all retail deposits in all ring-fenced institutions
should carry creditor priority, not just those up to the FSCS
limit". It added:
We are doubtful that this would have a material impact
on the cost of wholesale funding and believe that the consumer
message is much more powerful and comprehensible without qualification.
The Building Societies Association also said that
it "continues to support full retail depositor preference,
going beyond the FSCS coverage limit".
272. The Treasury stated that it had considered these
arguments in coming to its decision, and cited drawbacks of extending
depositor preference "such as the dilution of the benefits
of preference to the FSCS, and the effect of increasing the exposure
of other non-preferred groups to losses in the event the bank
fail". It argued that "no group or sector stands out
as an exceptional case".
It also noted:
it is expected that FSCS coverage will be extended
(under the Proposal for a Directive on Deposit Guarantee Schemes)
to include currently uncovered deposits. This will mean that all
individuals and most organisations will be eligible for FSCS protection
for amounts deposited up to the coverage limit.
Barclays similarly argued that extending to other
kinds of deposits was "a wholly inappropriate outcome"
for several reasons:
First, it is not equivalent to the insurance protection
provided to other depositors, as it will not guarantee payment.
Second, it creates a material communication issue with depositors
as to the nature of any protection to which they are exposed.
Third, in the context of communications, understanding how depositor
preference without insurance will work will require significant
sophistication on the part of customers. We believe that any expansion
of depositor protection should be dealt with best through changes
to the relevant insurance schemes and carefully coordinated across
at least Europe.
273. An additional reason for banks to oppose the
extension of depositor preference is that it is likely to increase
their costs. The Treasury impact assessment estimates that insured
depositor preference would cost UK banks between £0.3bn and
£0.7bn on aggregate each year, as a result of having to pay
higher interest rates to funding sources that would find themselves
subordinated, such as corporate deposits or unsecured bonds.
No estimates have been published by the Treasury of the funding
cost implications for banks of extended depositor preference,
but this cost could reasonably be expected to be higher because
the remaining corporate deposit holders and unsecured bond holders
would carry an even greater proportion of losses in the event
POTENTIAL PROBLEMS WITH DEPOSITOR
274. The ICAEW argued that many of the creditors
who would find themselves subordinated as a result of insured
depositor preference were not really the type of creditors who
it would be desirable to force losses upon:
The price of protecting the banks and, possibly,
the taxpayer would be borne by institutions such as larger charities,
hospitals, schools, local authorities, universities and lawyer/accountants'
client money accounts (which often temporarily hold, for example,
clients' deposits on house purchases), as well as mid-sized companies.
They will not be able to use ring-fenced banks with confidence
There are three ways this could be solved. Firstly,
extend the FSCS coverage to all 'end-user' of the banking system.
Secondly, extend depositor preference to all end-users. Or, drop
the 'depositor preference' proposal altogether.
Which? noted that uninsured retail depositors would
include those who have temporary high balances above the £85,000
It is also clear to us that there are many occasions
during a person's life when they need to hold deposits above the
level set by the FSCS. These could include house purchase/sale,
receiving an insurance pay-out, inheritance or pension lump-sum.
We believe that at such times, retail depositors are not well
placed to impose market discipline on banks and the loss of a
deposit would risk having a catastrophic effect on the individual.
275. The Treasury considered that it was appropriate
for such depositors to face the risk of losses:
Where individuals and organisations hold sums beyond
the FSCS limit, the Government believes it is right that they
should monitor and manage the risk in where they place deposits,
as all other unsecured creditors must do (including individuals
and small businesses).
276. Sir Mervyn King pointed out that it could be
politically difficult for the authorities to impose losses on
uninsured depositors, especially those with temporary high balances:
On any given day you will find thousands of people
in that position, and this is when I would understand if you in
Parliament stood up and said, "This is deeply unfair."
We have no means at present of handling that [...]
If you have a very small bank that fails and only
12 people hold deposits above £85,000, in three different
constituencies, I am convinced that the Chancellor will decide
not to intervene. If you have thousands of people involved, I
think the pressure on the Chancellor will be enormous. 
277. RBS expressed the view that depositor preference
was "of questionable benefit", and argued that "the
preservation of the creditor hierarchy is an important principle
agreed by the industry." It went on to suggest that as an
deposit guarantee schemes should, as in the EU RRD
proposal, be included in a bail-in regime as a senior unsecured
creditor, similar to their status in liquidation. This would ensure
that the pool of eligible bail-in liabilities is as broad as possible
to absorb potential losses. 
278. The Bank of England's explanation of how they
might expect to resolve a large ring-fenced bank in future also
suggested a need to be able to bail in the deposit guarantee scheme,
and Paul Tucker also argued for this in a speech in October 2012,
saying that bailing in deposit insurers would be a way of ensuring
bail-in could work for banks without enough bondholders.
HSBC noted that "Depositor preference is [...] inconsistent
with the scope for deposit guarantee scheme bail-in",
because seniority for the FSCS would mean that it could only be
bailed in if uninsured creditors had taken full losses first,
otherwise this would represent a departure from the creditor hierarchy.
Several witnesses also noted that implementation of depositor
preference would require amendment of the Recovery and Resolution
Directive, because the draft Directive currently prohibits it.
is crucial that deposit insurance be designed so as to avoid creating
irresistible political pressure for ad hoc extension in the event
of bank failure, as was the case in the last crisis. Implementation
of the proposal for preference for insured deposits, by increasing
prospective losses for others, has the potential to accentuate
such pressure. Depositor preference would also appear to be in
conflict with one of the resolution strategies favoured by the
Bank of England, involving bail-in of the deposit insurance scheme.
Both the above points weaken the credibility of the Government's
proposal. The Commission considers that the Treasury's case that
all non-insured creditors, including charities and small businesses
and temporary high deposits of households, would be treated alike
in the event of failure, is unconvincing. In view of these problems,
the Commission recommends that the Government and Bank of England
establish a joint group to prepare and publish a full report on
the implications for resolution of depositor preference and of
the scope and extent of depositor insurance. This report should,
in particular, consider the feasibility of establishing a voluntary
scheme of insurance for deposits over £85,000 with arrangements
for opt-out. This report should be published at least two weeks
before the House of Commons report stage of the Bill.
280. Banks fund their assets (such as mortgages)
with a mixture of equity and debt (such as customer deposits and
bonds). This chapter has so far addressed proposals relating to
the debt that a bank issues. These reforms involving bail-in,
loss-absorbing debt requirements and depositor preference are
all concerned with how debt-holders are treated when a bank approaches
or reaches insolvency. Witnesses acknowledged that the reforms
are important and promising, but also that they were untested.
There remains a chance that they will fail, in the sense that
the authorities will either not be willing or not feel able to
impose losses on certain classes of creditors when the time comes.
281. The remainder of this chapter addresses measures
relating to equity. Imposing a requirement on the level of equity
a bank issues is a more certain and tested safeguarding method.
The owners of equity (the shareholders) are the residual claimants
on the assets of the bank. They are the first to benefit from
any profits and, by the same token, the first to suffer any loss.
The more equity a bank has, the more money it can lose before
it becomes insolvent: equity is a shock-absorber. The prospect
of the shareholders bearing more of the losses from risky investments,
rather than shifting them to debt holders (or the State, if the
bank is bailed out) should dissuade them from taking too much
risk in the first place. Bank equity is therefore a stabilising
force in the financial system.
282. Requirements on equity are guided by the Basel
process. This is a system of international agreements which sets
minimum standards for how banks fund themselves. The main measure
used by the Basel process is "Tier 1 capital", which
includes equity and a limited class of other things.
283. Two types of requirement on capital are used
in the regulation of banks. The first is setting a minimum ratio
of capital to 'risk-weighted assets', which was considered earlier
in this chapter. The second is setting a minimum on the total
amount of capital as a percentage of total unweighted assets -
a 'leverage ratio'. The leverage ratio treats all assets the same
rather than attempting to weight them by their relative riskiness.
284. Capital requirements against risk-weighted
assets are being tightened considerably by the Basel 3 process.
The Basel 3 rules are being implemented in the UK via the draft
EU Capital Requirements Directive and Capital Requirements Regulation
IV (CRDIV/CRR). The standards being implemented through CRDIV/CRR
require that all banks should issue Tier 1 capital of at least
8.5 per cent of total RWAs. They also set a leverage ratio as
an additional requirement, requiring that banks fund themselves
with a minimum of 3 per cent of Tier 1 capital relative to total
(i.e. unweighted) assets.
285. The leverage ratio which is based on
total assets and the capital ratio which is based on RWAs
are connected, but not the same. Two banks that had the same ratio
of capital to RWAs might easily have different leverage ratios.
The bank that had assets judged to be safer, and which therefore
had lower risk weights, could have a higher leverage ratio than
the bank whose assets were held to be riskier, and which therefore
had higher risk weights. If a bank were required to have capital
equal to at least 8.5 per cent of its RWA, it would need the average
risk rating of its assets to be at least 35 per cent (= 3 divided
by 8.5, expressed as a percentage) in order for it also to have
a leverage ratio above 3 per cent. The addition of the leverage
ratio as a supplement to the capital ratio has the effect of preventing
a bank from holding too many assets, even if the assets are judged
to be low risk. The assets judged to be low risk do not give rise
to a corresponding obligation to issue capital against them. The
leverage ratio acts a safety net. If the assets with low risk
weightings turn out to be greatly overvalued, an appropriate leverage
ratio can ensure that the bank has sufficient capital to absorb
286. The ICB proposed to raise Tier 1 capital requirements
for large ring-fenced banks from 8.5 per cent to 11.5 per cent
of RWAs. The Government has agreed to implement this proposal.
The ICB also proposed to increase the leverage ratio requirement
on these banks by the same proportion, from 3 per cent to 4.06
per cent, so that the backstop has the same effect as under Basel
3. The Government has rejected this latter proposal.
287. Some witnesses questioned whether the Government's
decision not to increase the leverage ratio in line with the ICB's
recommendation was the right one. One argument against the Government's
position was that measures using RWAs are held to be deeply flawed.
A robust back-stop was therefore needed. Andy Haldane told us
that "both Basel II and Basel III are built on the shakiest
of foundations, because on the denominator of that capital ratio
is a measure of the assets of the bank, weighted by risk".
Setting risk-weights is not an exact science: for example, both
German and Greek government bonds could be assigned zero risk
weights under Basel II, so that a capital requirement based on
risk-weighted assets would require no capital at all to be held
288. Another major problem with measures of RWAs
is that Basel II and III both allow some banks to calculate their
own risk weights using internal models, subject to some constraints.
Andy Haldane told us that "choice over those risk weights
moved from the regulator to the firm. The firm was then setting
its own standards through the risk models".
In 2009, the FSA asked banks to evaluate the capital requirements
on a common portfolio of assets. According to Andy Haldane,
the range of reported capital requirements held against
this common portfolio was striking. For wholesale exposures to
banks, capital requirements differed by a factor of over 100 per
cent. For corporate exposures, they differed by a factor of around
150 per cent. And for sovereign exposures, they differed by a
factor of up to 280 per cent. Those differences could equate to
a confidence interval around reported capital ratios of 2 percentage
points or more.
Michael Cohrs echoed this point:
the FSA sent out a portfolio to its constituents
and asked them to risk-weight it. It was pretty simple. I think
the expectation was that there would be a very narrow spread among
the banks when they came back with this relatively simple portfolio.
Lo and behold, it came back and there was a massive spread between
the bank that was most conservative and the bank that was most
aggressive. Risk-weighted assets are at the heart of the Basel
ratio system. Therefore, you just throw your hands up. You start
from there and say, "It doesn't really work".
In evidence, Paul Tucker drew attention to the progress
being made in improving disclosure of the processes underpinning
It is very important that banks have to disclose
a lot more about where they are taking risks, because what you
described plainly happened and could happen again, and the underlying
problem is that the asset management industry is not looking at
risk-adjusted returns. I think they could fairly say at the moment
that it is quite hard to do that when they cannot see what is
inside the banks. The industry has come up with a disclosure framework
internationally that goes a lot further than anything the regulators
have demanded so far, and we and the international community have
289. Sir Mervyn King joined Andy Haldane in cautioning
against "relying too much on risk-weighted assets".
He said that risk weights were set using an "international
process which is very hard to negotiate and therefore extremely
inflexible", that risk weights "change over time"
and that in a major financial crisis "it is a time when the
things you thought were risky or less risky turn out not to be".
He suggested that the leverage ratio turned out to be "a
far better predictor of the institutions that failed in the crisis"
than measures of risk-weighted assets.
290. The banks and building societies supported the
Government's decision not to increase the leverage ratio. HSBC
noted that, "when the proposed increased capital requirements
for G-SIBs were announced by the Basel Committee and FSB, they
did not recommend any changes to the leverage requirements"
and went on to say:
the original leverage ratio of 3 per cent devised
by the Basel Committee was intended to apply across all banks
including universal banks. Therefore it was a blended rate recognising
that banks hold low risk liquidity pools and mortgages as well
as higher risk-weighted corporate and wholesale banking assets.
Applying that blended rate to retail ring-fenced banks, with their
concentration of lending to lower risk mortgages and with larger
tranches of liquid assets, would in and of itself constitute a
more onerous standard. To go beyond this would mean that the leverage
ratio would becoming a binding capital restriction rather than
the backstop measure which the Basel Committee originally intended.
Related to these points is the contention that a
4 per cent minimum leverage ratio would be tighter than international
minimum standards, tilting the playing field against UK banks.
291. In opposition to an increase in the leverage
ratio it was also pointed out that a leverage cap would hit certain
banks and certain forms of lendingi.e. those with low risk
weightsparticularly hard. Mortgage lenders who conduct
what is widely regarded as low risk lending may be disproportionately
applies particularly to building societies, which cannot raise
capital quickly and whose assets often have low risk weights.
Nationwide contended that one response might be "perversely,
to increase our risk taking".
Nationwide argued that if
the requirement were tightened to 4 per cent, it should be "calibrated
based on institutions' risk profiles [...] The key issue therefore
is to determine the extent to which a retail ring-fenced bank
and a building society have different risk profiles [...] For
building societies, we would anticipate it [the leverage ratio]
being no more than 3 per cent, otherwise it will become a primary
Nationwide suggested that raising the leverage ratio might restrict
292. Sir Mervyn King addressed the issue of whether
mortgage lenders would be hit disproportionately, noting that
Northern Rock had a "staggering" degree of leverage
before it failed, and that "It would not be sensible to allow
a mortgage lender to build up to that degree of leverage. It is
important to constrain the degree of leverage."
Calibrating an otherwise flat leverage ratio requirement for different
institutions works against the premise of the simple leverage
ratiothat it is a mistake to rely too heavily on judgements
about relative riskiness. It is noteworthy that a number of building
societies have been taken over or gone into resolution during
the present crisis.
On the question of propensity to lend, Sir Mervyn King argued
that if as a result of increased leverage ratios a bank was "lending
less than it otherwise would, it means that we are minimising
the risk that something will go badly wrong."
293. In March 2012, following HM Treasury's earlier
request, the interim FPC agreed unanimously a statement outlining
its advice on potential powers of Direction for the statutory
FPC. This included that the FPC should seek powers of Direction
over a countercyclical capital buffer, sectoral capital requirements
and a leverage ratio. The Government set out its reply to this
request in September 2012. Citing consistency with international
standards, the Government stated that it intends to provide the
FPC with a time-varying leverage ratio tool, but no earlier than
2018 and subject to a review in 2017 to assess progress on international
precise design of the tool will depend on the provisions of the
relevant European legislation and will be set out in secondary
legislation to be introduced by the Government at the time.
on capital requirements based on risk weighted assets alone is
not sufficient. The leverage ratio is an important part of banks'
minimum capital requirements. If a 3 per cent leverage ratio is
a backstop when the requirement in terms of RWAs is 8.5 per cent,
raising the leverage ratio broadly in line with a higher requirement
in terms of RWAs is logical. The Commission is not convinced about
the appropriateness of the Government's decision to reject the
ICB's recommendation to limit leverage at 25 times rather than
33 times. We believe that high leverage was a
significant contributor to the crisis. The Commission considers
it essential that the ring-fence should be supported by a higher
leverage ratio, and would expect the leverage ratio to be set
substantially higher than the 3 per cent minimum required under
Basel III. Not to do so would reduce the effectiveness
of the leverage ratio as a counter-weight to the weaknesses of
the leverage ratio is a complex and technical decision, and one
which is ultimately best made by the regulator.
The FPC cannot be expected to work with one hand tied behind its
back. The FPC should be given the duty of setting the leverage
ratio from Spring 2013. An early change to the leverage ratio
would pose particular problems for some building societies. In
view of their special characteristics, the regulator should carefully
consider the case for longer transition arrangements for them.
Changes to leverage ratios might be mitigated by changes to the
tax treatment of debt and equity for banks, a matter to which
we will return in our Report in the New Year. We took little evidence
on the effects on regulatory arbitrage and passporting held to
be a possible consequence of setting higher capital or leverage
ratio at a national level than are required under Basel III. We
will consider this as part of our wider work on regulatory arbitrage
issues in our final Report.
leverage ratios have the drawback that they incentivise banks
to hold the highest-yielding and therefore presumably riskiest
assets that they can, and to offload as many lower-yielding and
safer assets as they can into other companies. Risk-weighting
of assets was introduced as a remedy. Risk-weighting has, however,
been unsatisfactory and arguably dangerous in practice. Banks
were allowed to set their own risk weights using their own models.
Some of the weights were much too low. The zero or low weights
attached to government securities have encouraged banks to acquire
large amounts of what were in some cases very risky assets. Many
governments have an incentive not to address this, because of
their need to fund large deficits. Parliament needs to be assured
that the work to improve risk-weighting is being given the highest
priority. The Commission recommends that the new Bill require
the Bank of England to provide an annual assessment to be laid
before Parliament of progress of risk-weighting and that the assessment
should examine in particular the possible operation of floors
for risk-weights, and steps taken with regard to simplification
of risk-weights and trading exposures. If a more independent and
more skilled Supervisory Board of the Bank of England is established
in accordance with the recommendations of the Treasury Committee,
it would be important for this Board to provide regular oversight
of the work by the Bank of England in this area.
330 HM Treasury, Sound banking: delivering reform,
Cm 8453, October 2012 p 15 Back
Q 1003 Back
Ev w181 Back
Q 705 Back
Q 710 Back
Q 1003 Back
Q 796 Back
Q 716 Back
Q 727 Back
Q 1169 Back
HM Treasury, Sound banking: delivering reform, Cm 8453,
October 2012, p 5 Back
Ev w127 Back
Ev w180 Back
Q 443 Back
Independent Commission on Banking, Final Report, September 2011,
para 4.78 Back
HM Treasury, Sound banking: delivering reform, Cm 8453,
October 2012, para 2.57 Back
Ibid., para 2.56 Back
Ibid., para 2.6 Back
HM Treasury, Banking reform: delivering stability and
supporting a sustainable economy, June 2012, Cm 8356, p 38 Back
Q 1100 Back
Qq 443, 788 Back
Q 443 Back
Q 788 Back
Ev w33 Back
Ev w151 Back
Q 986 Back
Q 1216 Back
Q 1099 Back
Ev w54 Back
Draft Financial Services (Banking Reform) Bill, Clause 4 Back
Q 1097 Back
Independent Commission on Banking, Final Report, September 2011,
para 4.135 Back
Banks' liabilities to their own pension schemes, overseas deposits
up to the FSCS coverage limit, and deposits placed by groups such
as charities or local authorities. Back
HM Treasury, Banking reform: delivering stability and
supporting a sustainable economy, June 2012, Cm 8356, consultation
question 4 Back
HM Treasury, Sound banking: delivering reform, Cm 8453,
October 2012, para 2.42 Back
Ev w70 Back
Independent Commission on Banking, Final Report, September 2011,
para 4.89 Back
Ibid., para 4.89 Back
Ibid., para 4.90 Back
HC Deb, 16 April 2012, col 2WS [Written Ministerial Statement]
Ev w4 Back
Ev w95 Back
Building Societies Association, Submission to Government Consultation
on the Future of Building Societies, para 72 Back
HM Treasury, Sound banking: delivering reform, Cm 8453,
October 2012, para 2.51 Back
Ibid., 2.52 Back
Ev w33 Back
HM Treasury, Sound banking: delivering reform, Cm 8453,
October 2012 Back
Ev w70 Back
Ev w168 Back
HM Treasury, Sound banking: delivering reform, Cm 8453,
October 2012, para 2.52 Back
Q 1175, Q 1176 Back
Ev w108 Back
Ev w182, para 11 [Bank of England] Back
Bank of England, Speech by Paul Tucker to International Association
of Deposit Insurers Annual Conference, 25 October 2012 Back
Ev w137 Back
Ev w100, para 14 [Professor Rosa M. Lastra]; Ev w137, para 1.71
[HSBC Holdings] Back
Qq 705, 710, 727, 728, 731. Back
Tier 1 capital includes common shares, retained earnings and other
instruments which can absorb losses on a going-concern basis.
These other instruments must comply with a range of criteria relating
to features such as subordination, coupon payments and redemption.
Source: www.bis.gov.uk Back
Strictly speaking, the leverage ratio is calculated using 'exposures'
rather than 'assets'. Back
Q 628 Back
Q 629 Back
Bank of England, Remarks by Andrew Haldane, 9 January 2011 Back
Uncorrected transcript of oral evidence before Parliamentary
Commission on Banking Standards Panel on Regulatory Approach on
11 December 2012, HC 821-i, Q 8 Back
Q 1214 Back
Q 1209 Back
Ev w131 Back
Ev w42, para 11 [Building Societies Association], para 11 Back
Ev w92, para 3 Back
Ibid., para 6 Back
Ibid., para 3 Back
Q 1213 Back
See, for example, "Financial Services Authority confirms
Chelsea Building Society merger with the Yorkshire Building Society",
FSA press notice 050/2010, 19 Mar 2010 Back
Q 1213 Back
HM Treasury, The Financial Services Bill: the Financial Policy
Committee's macro-prudential tools, Cm8434, September 2012 Back