Financial Stability Report
June 2017
|
Issue No. 41
Financial Stability Report
Presented to Parliament pursuant to Section 9W(10) of the Bank of England
Act 1998 as amended by the Financial Services Act 2012.
June 2017
BANK OF ENGLAND
© Bank of England 2017
ISSN 1751-7044
BANK OF ENGLAND
Financial Stability Report
June 2017 | Issue No. 41
The primary responsibility of the Financial Policy Committee (FPC), a committee of the Bank of England, is
to contribute to the Bank of England’s objective for maintaining financial stability. It does this primarily by
identifying, monitoring and taking action to remove or reduce systemic risks, with a view to protecting and
enhancing the resilience of the UK financial system. Subject to that, it supports the economic policy of
Her Majesty’s Government, including its objectives for growth and employment.
This Financial Stability Report sets out the FPC’s view of the outlook for UK financial stability, including its
assessment of the resilience of the UK financial system and the current main risks to financial stability, and
the action it is taking to remove or reduce those risks. It also reports on the activities of the Committee
over the reporting period and on the extent to which the Committee’s previous policy actions have
succeeded in meeting the Committee’s objectives. The Report meets the requirement set out in legislation
for the Committee to prepare and publish a Financial Stability Report twice per calendar year.
In addition, the Committee has a number of duties, under the Bank of England Act 1998. In exercising
certain powers under this Act, the Committee is required to set out an explanation of its reasons for
deciding to use its powers in the way they are being exercised and why it considers that to be compatible
with its duties.
The Financial Policy Committee:
Mark Carney, Governor
Jon Cunliffe, Deputy Governor responsible for financial stability
Ben Broadbent, Deputy Governor responsible for monetary policy
Sam Woods, Deputy Governor responsible for prudential regulation
Andrew Bailey, Chief Executive of the Financial Conduct Authority
Alex Brazier, Executive Director for Financial Stability Strategy and Risk
Anil Kashyap
Donald Kohn
Richard Sharp
Martin Taylor
Charles Roxburgh attends as the Treasury member in a non-voting capacity.
This document was delivered to the printers on 26 June 2017 and, unless otherwise stated, uses data available
as at 16 June 2017. This page was revised on 11 April 2018.
The Financial Stability Report is available in PDF at www.bankofengland.co.uk.
Foreword
Executive summary i
Box 1 The FPC’s 2017 Q2 UK countercyclical capital buffer rate decision vi
Box 2 Possible financial stability implications of the United Kingdom’s withdrawal from
the European Union vii
Part A: Main risks to financial stability
The FPC’s approach to addressing risks from the UK mortgage market 1
Box 3 PRA Supervisory Statement on underwriting standards for buy-to-let mortgages 11
Box 4 Powers of Direction over LTV limits 12
Box 5 The affordability test Recommendation 13
UK consumer credit 14
Box 6 Overview of the UK consumer credit market 18
Global environment 20
Asset valuations 23
Part B: Resilience of the UK financial system
Banking sector resilience 27
Box 7 Building cyber resilience in the UK financial system 32
Market-based finance 34
Box 8 The UK High-Value Payment System 39
The FPC’s medium-term priorities 41
Annex 1: Previous macroprudential policy decisions 42
Annex 2: Core indicators 45
Index of charts and tables 49
Glossary and other information 51
Contents
E
xecutive summary
i
Executive summary
The Financial Policy Committee (FPC) aims to ensure the UK financial system is resilient to the wide range of risks it
faces.
The FPC assesses the overall risks from the domestic environment to be at a standard level: most financial stability
indicators are neither particularly elevated nor subdued.
As is often the case in a standard environment, there are pockets of risk that warrant vigilance. Consumer credit has
increased rapidly. Lending conditions in the mortgage market are becoming easier. Lenders may be placing undue
weight on the recent performance of loans in benign conditions.
Exit negotiations between the United Kingdom and the European Union have begun. There are a range of possible
outcomes for, and paths to, the United Kingdom’s withdrawal from the EU.
Some possible global risks have not crystallised, though financial vulnerabilities in China remain pronounced.
Measures of market volatility and the valuation of some assets — such as corporate bonds and UK commercial real
estate — do not appear to reflect fully the downside risks that are implied by very low long-term interest rates.
To ensure that the financial system has the resilience it needs, the FPC is:
• Increasing the UK countercyclical capital buffer rate to 0.5%, from 0%. Absent a material change in the
outlook, and consistent with its stated policy for a standard risk environment and of moving gradually, the FPC
expects to increase the rate to 1% at its November meeting.
• Bringing forward the assessment of stressed losses on consumer credit lending in the Bank’s 2017 annual stress
test. This will inform the FPC’s assessment at its next meeting of any additional resilience required in aggregate
against this lending. The FPC further supports the intentions of the Prudential Regulation Authority and Financial
Conduct Authority to publish, in July, their expectations of lenders in the consumer credit market.
• Clarifying its existing insurance measures in the mortgage market, designed to prevent excessive growth in the
number of highly indebted households. This will promote consistency across lenders in their application of tests to
assess whether new mortgage borrowers can afford repayments.
• Consistent with its previous commitment, restoring the level of resilience delivered by its leverage ratio
standard to the level it delivered in July 2016 before the FPC excluded central bank reserves from the leverage ratio
exposure measure. The FPC intends to set the minimum leverage requirement at 3.25% of non-reserve exposures,
subject to consultation.
• Overseeing contingency planning to mitigate risks to financial stability as the United Kingdom withdraws from
the European Union.
• Building on the programme of cyber resilience testing it instigated in 2013, by setting out the essential elements
of the regulatory framework for maintaining cyber resilience. It will now monitor that each element is being
fulfilled by the relevant UK authorities.
i
i Financial Stability Report
J
une 2017
The Financial Policy Committee (FPC) assesses the overall
risks from the domestic environment to be at a standard
l
evel: most financial stability indicators are neither
particularly elevated nor subdued.
As is often the case in a standard environment, there are
pockets of risk that warrant vigilance. Consumer credit has
i
ncreased rapidly. Lending conditions in the mortgage
market are becoming easier. Lenders may be placing undue
weight on the recent performance of loans in benign
conditions.
The FPC is increasing the UK countercyclical capital buffer
(CCyB) rate to 0.5%, from 0% (see Box 1). Absent a
material change in the outlook, and consistent with its
stated policy for a standard risk environment and of moving
gradually, the FPC expects to increase the rate to 1% at its
November meeting.
• The action will supplement banks’ already substantial
ability to absorb losses (Chart A).
• At its November meeting, the FPC will have the full set of
results from the 2017 stress test of major UK banks.
• In line with its published policy, the FPC stands ready to cut
the UK CCyB rate, as it did in July 2016, if a risk materialises
t
hat could lead to a material tightening of lending
conditions. Banks’ capital buffers exist to be used as
necessary to allow banks to support the real economy in a
downturn.
T
he FPC supports the intentions of the Prudential
Regulation Authority (PRA) and Financial Conduct
Authority (FCA) to publish, in July, their expectations of
lenders in the consumer credit market. Firms remain the
first line of defence. Effective governance at firms should
ensure that risks are priced and managed appropriately and
benign conditions do not lead to complacency by lenders.
The Bank’s annual stress test assesses banks’ resilience to
risks in consumer credit. Given the rapid growth in
consumer credit over the past twelve months, the FPC is
bringing forward the assessment of stressed losses on
consumer credit lending in the Bank’s 2017 annual stress
test. This will inform the FPC’s assessment at its next
meeting of any additional resilience required in aggregate
against this lending.
• Consumer credit grew by 10.3% in the twelve months to
April 2017 (Chart B) — markedly faster than nominal
household income growth. Credit card debt, personal loans
and motor finance all grew rapidly.
5
0
5
10
15
20
25
2013 14 15 16 17
Percentage points
Total consumer credit
(b)
Credit card
(b)
Dealership car finance
(a)
Other (non-credit card and non-dealership car finance)
(b)(d)
Nominal household income growth
(c)
+
–
Chart B Consumer credit has been growing much faster
than household incomes
Annual growth rates of consumer credit products and household
income
Sources: Bank of England, ONS and Bank calculations.
(a) Identified dealership car finance lending by UK monetary financial institutions (MFIs) and
other lenders.
(b) Sterling net lending by UK MFIs and other lenders to UK individuals (excluding student
loans). Non seasonally adjusted.
(c) Percentage change on a year earlier of quarterly nominal disposable household income.
Seasonally adjusted.
(d) Other is estimated as total consumer credit lending minus dealership car finance and credit
card lending.
Chart A Major UK banks have continued to strengthen
their capital positions
Major UK banks’ capital ratios
Sources: PRA regulatory returns, published accounts and Bank calculations.
(a) Major UK banks’ core Tier 1 capital as a percentage of their risk-weighted assets. Major UK
banks are Banco Santander, Bank of Ireland, Barclays, Co-operative Banking Group, HSBC,
Lloyds Banking Group, National Australia Bank, Nationwide, RBS and Virgin Money. Data
exclude Northern Rock/Virgin Money from 2008.
(b) Between 2008 and 2011, the chart shows core Tier 1 ratios as published by banks, excluding
hybrid capital instruments and making deductions from capital based on FSA definitions.
Prior to 2008 that measure was not typically disclosed; the chart shows Bank calculations
approximating it as previously published in the Report.
(c) Weighted by risk-weighted assets.
(d) From 2012, the ‘Basel III common equity Tier 1 capital ratio’ is calculated as common equity
Tier 1 capital over risk-weighted assets, according to the CRD IV definition as implemented in
the United Kingdom. The Basel III peer group includes Barclays, Co-operative Banking Group,
HSBC, Lloyds Banking Group, Nationwide, RBS and Santander UK.
(e) CET1 ratio less the aggregate percentage point fall projected under the Bank of England’s
2016 annual cyclical stress scenario for the six largest UK banks.
E
xecutive summary
iii
• Loss rates on consumer credit lending are low at present.
Partly as a result, banks’ net interest margins on new
l
ending have fallen and major lenders are using lower risk
weights to calculate the capital they need to hold. The
current environment is also likely to have improved the
credit scores of borrowers.
•
Other things equal, these developments mean lenders have
less capacity to absorb losses, either with income or capital
buffers. In this context, a review by the PRA has found
evidence of weaknesses in some aspects of underwriting
and a reduction in resilience.
• The short maturity of consumer credit means that the
credit quality of the stock of lending can deteriorate
quickly. Lenders expect to continue to grow their portfolios
this year, at the same time as real household income
growth is expected to remain particularly weak.
The FPC has clarified its existing insurance measures in the
mortgage market, designed to prevent excessive growth in
the number of highly indebted households. Lenders should
test affordability at their mortgage reversion rate —
typically their standard variable rate — plus 3 percentage
points. This will promote consistency across lenders in their
application of tests to assess whether new mortgage
borrowers can afford repayments.
• Historically, the build-up of mortgage debt has been a
significant risk to financial and economic stability. Because
highly indebted borrowers need to cut spending sharply in a
downturn, recessions become deeper. And looser
underwriting standards expose banks to bigger losses.
• The FPC put policies in place to guard against these risks
in 2014. These Recommendations were: a limit on lending
at loan to income multiples at 4.5 or above; and guidance
to lenders to assess whether new borrowers would be
able to afford their repayments if interest rates were to
rise.
• Following a review (see The FPC’s approach to addressing
risks from the UK mortgage market chapter), the FPC
expects its measures to remain in place for the foreseeable
future.
• Mortgage lending at high loan to income ratios is increasing
and the spreads and fees on mortgage lending have fallen.
If lenders were to weaken underwriting standards to
maintain mortgage growth, the FPC’s measures would
limit growth in the number of highly indebted households.
This would have material benefits for economic and
financial stability by mitigating the further cutbacks in
spending that highly indebted households make in
downturns.
Consistent with its previous commitment, the FPC is
restoring the level of resilience delivered by its leverage
r
atio standard to the level it delivered in July 2016, before
the FPC excluded central bank reserves from the leverage
ratio exposure measure. The FPC therefore intends to set
the minimum leverage requirement at 3.25% of
non-reserve exposures, subject to consultation.
• In July 2016, the FPC excluded central bank reserves from
the measure of banks’ exposures used to assess their
leverage. This change reflected the special nature of central
bank reserves and was designed to avoid a situation in
which the Committee’s leverage standards impeded the
transmission of monetary policy.
• The FPC committed last year that it would make an
offsetting adjustment to ensure that the amount of capital
needed to meet the UK leverage ratio standard would not
decline. The FPC did not intend for there to be a
permanent loosening of the standard.
• By raising the minimum leverage standard from 3% to
3.25%, the FPC intends to ensure that the original standard
of resilience is restored, while also preserving the benefits
of excluding central bank reserves from the exposure
measure.
Exit negotiations between the United Kingdom and the
European Union have begun. There are a range of possible
outcomes for, and paths to, the United Kingdom’s
withdrawal from the EU. The FPC will oversee contingency
planning to mitigate risks to financial stability as the
withdrawal process evolves (see Box 2).
Irrespective of the particular form of the United Kingdom’s
future relationship with the European Union, and consistent
with its statutory responsibility, the FPC will remain
committed to the implementation of robust prudential
standards in the UK financial system. This will require a
level of resilience to be maintained that is at least as great
as that currently planned, which itself exceeds that required
by international baseline standards.
• The United Kingdom’s position as the leading
internationally active financial centre, with a financial
centre that is, by asset size, around ten times GDP, means
that the FPC’s statutory responsibility of protecting and
enhancing the resilience of the UK financial system is
particularly important for both the domestic and global
economies.
• Absent consistent implementation of standards
internationally and appropriate supervisory co-operation,
the FPC would need to assess how best to protect the
resilience of the UK financial system.
i
v Financial Stability Report
J
une 2017
Some possible global risks have not crystallised, though
financial vulnerabilities in China remain pronounced.
Measures of market volatility and the valuation of some
assets — such as corporate bonds and UK commercial real
e
state — do not appear to reflect fully the downside risks
that are implied by very low long-term interest rates.
Banks’ ability to withstand these risks is being tested in the
2017 stress test scenario.
• Euro-area sovereign bond spreads have fallen as some
political uncertainties have been resolved. Further progress
has been made in strengthening European bank capital
positions, and a domestically significant bank in Spain was
resolved in an orderly fashion.
• In China, capital outflows have stabilised, but economic
growth continues to be accompanied by rapid credit
expansion (Chart C).
• Measures of uncertainty implied by options prices are low
(see Asset valuations chapter). Often in periods of low
volatility, risks are building and later become apparent.
• In the United Kingdom, ten-year real government bond
yields are at around -2% (Chart D). Long-term real rates
are low across the G7. These levels are consistent with
pessimistic growth expectations and high perceived tail
risks.
• Some asset valuations, particularly for some corporate
bonds and UK commercial real estate assets, appear to
factor in a low level of long-term market interest rates but
do not appear to be consistent with the pessimistic and
u
ncertain outlook embodied in those rates (Chart E).
0
10
20
30
40
50
0
50
100
150
200
250
300
2006 08 10 12 14 16
Non-financial sector
(a)
(right-hand scale)
A
djusted total social financing
(b)
(left-hand scale)
H
eadline total social financing
(c)
(left-hand scale)
P
ercentage changes on a year earlier Per cent of GDP
Chart C Credit continues to grow rapidly in China
China non-financial sector debt and growth of total social
financing
Sources: BIS total credit statistics, CEIC and Bank calculations.
(a) Non-financial sector debt data are to 2016 Q4. Includes lending by all sectors at market
value as a percentage of GDP, adjusted for breaks.
(b) Total social financing adjusted for net issuance of local government bonds.
(c) The People’s Bank of China stock of total social financing used from December 2014
onwards. Prior to this the stock of total social financing is estimated using monthly ‘newly
increased’ total social financing flows.
3
2
1
0
1
2
3
4
2006 07 08 09 10 11 12 13 14 15 16 17
United Kingdom
United States
E
uro area
Per cent
+
–
Chart D Advanced-economy risk-free real interest rates
remain close to historically low levels
International ten-year real government bond yields
(a)
Sources: Bloomberg and Bank calculations.
(a) Zero-coupon bond yields derived using inflation swap rates. UK real rates are defined
relative to RPI inflation, whereas US and euro-area real rates are defined relative to CPI and
HICP inflation respectively.
40
50
60
70
80
90
100
110
120
130
140
150
03 05 07 09 11 13 15 17
Indices: 2007 Q2 = 100
2001
London West End
office prices
Aggregate
CRE prices
Ranges of sustainable valuations
(a)
Upper part of ranges: low rental yields persist.
Lower part of ranges: rental yields rise,
consistent with a fall in rental growth
expectations or a rise in risk premia.
Chart E UK commercial real estate prices look stretched
based on ranges of sustainable valuations
Commercial real estate prices in the United Kingdom and ranges
of sustainable valuations
Sources: Bloomberg, Investment Property Forum, MSCI Inc. and Bank calculations.
(a) Sustainable valuations are estimated using an investment valuation approach and are based
on an assumption that property is held for five years. The sustainable value of a property is
the sum of discounted rental and sale proceeds. The rental proceeds are discounted using a
5-year gilt yield plus a risk premium, and the sale proceeds are discounted using a 20-year,
5-year forward gilt yield plus a risk premium. Expected rental value at the time of sale is
based on Investment Property Forum Consensus forecasts. The range of sustainable
valuations represents varying assumptions about the rental yield at the time of sale: either
rental yields remain at their current levels (at the upper end), or rental yields revert to their
15-year historical average (at the lower end). For more details, see Crosby, N and Hughes, C
(2011), ‘The basis of valuations for secured commercial property lending in the UK’, Journal of
European Real Estate Research, Vol. 4, No. 3, pages 225–42.
E
xecutive summary
v
• These asset prices are therefore vulnerable to a repricing,
whether through an increase in long-term interest rates or
an adjustment of growth expectations, or both. The impact
of this could be amplified given reduced liquidity in some
m
arkets.
Progress has been made in building resilience to cyber
attack, but the risk continues to build and evolve.
Regulators are nearing completion of a first round of cyber
resilience testing for all firms at the core of the UK financial
system, in line with the Recommendation from the FPC in
2015.
• The FPC’s concern is to mitigate systemic risk — the risk of
material disruption to the economy.
• With 31 out of 34 firms at the core of the UK financial
system, including banks representing more than 80% of the
outstanding stock of PRA-regulated banks’ lending to the
UK real economy, so far having completed penetration
testing and having action plans in place, the FPC is satisfied
that its 2015 Recommendation has been met.
• Consistent with that, the FPC is also setting out the
essential elements of the regulatory framework for
maintaining cyber resilience and will now monitor that each
element is being fulfilled by the relevant UK authorities.
• Alongside the Bank, PRA and FCA, the FPC will now
consider its tolerance for the disruption to important
economic functions of the financial system in the event of
cyber attack.
The FPC has updated its medium-term priorities (see
The FPC’s medium-term priorities chapter).
• The FPC’s primary responsibility is to identify, monitor and
t
ake action to remove or reduce systemic risks, with a view
to protecting and enhancing the resilience of the UK
financial system. It aims to ensure the financial system
does not cause problems for the rest of the economy and, if
and when problems arise in the economy, the financial
system can absorb rather than amplify them.
• To help to meet its objectives, alongside its ongoing
assessment of the risk environment, the FPC is prioritising
three initiatives over the next two to three years:
• Finalising, and refining if necessary, post-crisis bank
capital and liquidity reforms.
• Completing post-crisis reforms to market-based finance
in the United Kingdom, and improving the assessment of
systemic risks across the financial system.
• Preparing for the United Kingdom’s withdrawal from the
European Union.
Part A of this Report sets out in detail the Committee’s
analysis of the major risks and action it is taking in the light of
those risks. Part B summarises the Committee’s analysis of
the resilience of the financial system.
v
i Financial Stability Report
J
une 2017
Box 1
The FPC’s 2017 Q2 UK countercyclical capital
buffer rate decision
The FPC is increasing the UK countercyclical capital buffer
(CCyB) rate from 0% to 0.5%, with binding effect from
27 June 2018. Absent a material change in the outlook, and
consistent with its stated policy for a standard risk
e
nvironment and of moving gradually, the FPC expects to
increase the rate to 1% at its November meeting, with
binding effect a year after that. At that point, it will have the
full set of results from the 2017 stress test of major UK banks.
The increase to 0.5% will raise regulatory buffers of common
equity Tier 1 capital by £5.7 billion. This will provide a buffer
of capital that can be released quickly in the event of an
adverse shock occurring that threatens to tighten lending
conditions. The increase in the CCyB rate will also lead to a
proportional increase in major UK banks’ leverage
requirements via the countercyclical leverage buffer (CCLB).
The Committee’s decision to increase the UK CCyB rate to
0.5% — with an expectation of a further increase to 1% in
November — reflects its assessment of the current risk
environment and its intention to vary the buffer in gradual
steps.
In its published strategy for setting the CCyB, the FPC
signalled that it expects to set a UK CCyB rate in the region of
1% in a standard risk environment. The FPC assesses the
overall risks from the domestic environment to be at a
standard level: most financial stability indicators are neither
particularly elevated nor subdued. Domestic credit has grown
broadly in line with nominal GDP over the past two years
(Chart A). Within the overall risk environment, some
indicators are more benign. For example, despite high levels of
indebtedness, private sector debt-servicing costs are low,
supported by the low level of interest rates. In contrast, risk
levels in some sectors are more elevated, notably so in the
consumer credit market (see UK consumer credit chapter).
Global risks — which could influence the risks on UK exposures
indirectly via their potential effects on UK economic growth —
are also judged to be material, as are risks from some asset
valuations.
The FPC’s measured approach is likely to decrease the risk that
banks adjust by tightening credit conditions, thereby
minimising the cost to the economy of making the banking
system more resilient.
In line with its published policy, the FPC stands ready to cut
the UK CCyB rate, as it did in July 2016, if a risk materialises
that could lead to a material tightening of lending conditions.
The cut in the CCyB rate in July 2016 was a response to
greater uncertainty around the UK economic outlook and an
increased possibility that material domestic risks could
crystallise in the near term. The FPC’s action served to ensure
banks did not hoard capital and restrict lending in those
conditions. Banks’ capital buffers exist to be used as necessary
to allow banks to support the real economy in a downturn.
Under EU law, the UK CCyB rate applies automatically (up to a
2.5% limit, and currently subject to a transition timetable) to
the UK exposures of firms incorporated in other European
Economic Area (EEA) states. The FPC expects it to apply also
to internationally active banks in jurisdictions outside the EEA
that have implemented the Basel III regulatory standards.
Consistent with this, recent CCyB actions by Czech Republic,
Hong Kong and Norway have been reciprocated.
5
0
5
10
15
2
000 02 04 06 08 10 12 14 16
Nominal GDP
growth rate
(b)
B
ank credit
(a)
Per cent
+
–
C
hart A Credit directly financed by the banking system
h
as grown broadly in line with nominal GDP over the
past two years
Growth in credit to households and firms compared with nominal
GDP growth
Sources: ONS and Bank calculations.
(a) Quarterly twelve-month growth rate of monetary financial institutions’ sterling net lending
to private non-financial corporations and households (in per cent) seasonally adjusted.
(b) Twelve-month growth rate of nominal GDP.
E
xecutive summary
vii
Box 2
Possible financial stability implications of the
United Kingdom’s withdrawal from the
European Union
In March 2017, the UK Government notified the European
Council of the United Kingdom’s intention to withdraw from
the European Union. This initiated, under Article 50 of the
Treaty on European Union, a two-year period for the
United Kingdom and the European Union to negotiate and
conclude a withdrawal agreement. The exit negotiations have
now begun.
As the FPC stated in September 2016, irrespective of the
particular form of the United Kingdom’s future relationship
with the European Union, and consistent with its statutory
responsibility, the FPC will remain committed to the
implementation of robust prudential standards in the UK
financial system. This will require a level of resilience to be
maintained that is at least as great as that currently planned
and which itself exceeds that required by international
baseline standards.
(1)
In addition, consistent with its statutory duty, the FPC will
continue to identify and monitor UK financial stability risks,
so that preparations can be made and action taken to
mitigate them.
There are a range of possible outcomes for the
United Kingdom’s future relationship with the European Union
and possible paths to that relationship. Consistent with its
remit, the FPC is focused on scenarios that, even if they may
be the least likely to occur, could have most impact on
UK financial stability. This includes a scenario in which there is
no agreement in place at the point of exit. Such scenarios are
where contingency planning and preparation will be most
valuable.
The Bank, FCA and PRA are working closely with regulated
firms and financial market infrastructures (FMIs) to ensure
they have comprehensive contingency plans in place. The FPC
will oversee contingency planning to mitigate risks to financial
stability as the withdrawal process unfolds.
Through this work, the FPC is aiming to promote an
orderly adjustment to the new relationship between the
United Kingdom and the European Union.
Without contingency plans that can be executed in the
available time, effects on financial stability could arise both
through direct effects on the provision of financial services,
and indirectly, through macroeconomic shocks that could test
the resilience of the financial system.
(1) Direct effects on the provision of financial services
A very large part of the United Kingdom’s legal and regulatory
f
ramework for financial services is directly or indirectly derived
from EU law. The United Kingdom’s financial services law
must therefore become domestic at the point of
withdrawal. The Government plans to execute this through
the Repeal Bill. Once enacted, this will ensure there is no legal
o
r regulatory vacuum in respect of financial services when the
United Kingdom leaves the European Union.
The European Union’s framework for financial services
establishes the right of financial companies within the
European Economic Area (EEA) to provide services across
national borders and to establish local branches in other
Member States without local authorisation.
This promotes substantial cross-border provision of a wide
range of financial services. Around £40 billion of UK financial
services revenues relate to EU clients and markets.
(2)
These
cross-border connections have resulted in more efficient
financial services for businesses and households across the
European Union.
There is no generally applicable institutional framework for
cross-border provision of financial services outside the
European Union. Globally, liberalisation of trade in services
lags far behind liberalisation of trade in goods. So without a
new bespoke agreement, UK firms could no longer provide
services to EEA clients (and vice versa) in the same manner as
they do today, or in some cases not at all. This creates two
broad risks. First, services could be dislocated as clients and
providers adjust. Second, the fragmentation of service
provision could increase costs and risks.
In the United Kingdom, the flow of new banking and
insurance services to UK customers could be disrupted if
EEA firms are unable to operate in the United Kingdom in
the same manner as they do today. Around 10% of the
outstanding stock of loans to private non-financial
corporations in the United Kingdom is extended by
UK branches of EEA banks.
(3)
Around 7% of general insurance contracts undertaken in the
United Kingdom and 3% of life insurance contracts are
written by EEA insurers.
(4)
As well as disrupting new business
from these providers, fragmentation could require the existing
contracts to be transferred to a UK-authorised firm in order to
address any legal uncertainties as to the status of, and ability
to perform, such contracts.
(1) www.bankofengland.co.uk/publications/Pages/news/2016/033.aspx.
(2) Source: Oliver Wyman, 2016.
(3) Source: Bank of England calculations.
(4) Sources: Firms’ published accounts, regulatory data and Bank calculations. Based on
premiums relating to insurance contracts.
v
iii Financial Stability Report
J
une 2017
There could also be material dislocation of some services
supplied from the United Kingdom to the European Union.
EU clients would need to source substitute services from banks
and FMIs established in the EEA or other countries recognised
b
y the European Commission as ‘equivalent’. This is
particularly relevant to new debt and equity issuance and
derivatives business. These dislocations could also disrupt
the provision of services to UK clients who rely on EU
counterparties.
UK-located banks underwrite around half of the debt and
equity issued by EU companies.
(1)
EU companies could need
to find alternative providers of this service to sustain their
capital market issuance.
UK-located banks are counterparty to over half of the
over-the-counter (OTC) interest rate derivatives traded by
EU companies and banks.
(2)
To support EU-based derivatives
trading, substantial operational capacity may need to be
established in the European Union and additional capital and
balance sheet capacity would probably be needed.
Central counterparties (CCPs) located in the
United Kingdom provide services to EU clients in a range of
markets. The United Kingdom houses some of the world’s
largest CCPs. For example, LCH handles over 90% of
cleared interest rate swaps globally.
In addition to the potential disruption to new clearing business
for EU firms, if EU firms are unable to move their existing
derivatives contracts to EU authorised or recognised CCPs,
they would face capital charges that are up to ten times
higher. Moreover, to move a large stock of existing trades will
pose substantial and complex operational and legal
challenges.
In addition to the dislocation of services, fragmentation of
market-based finance could result in higher costs and
greater risks for both EU and UK companies and households.
Separation of derivatives clearing would reduce the benefits of
central clearing. It would impair the ability to diversify risks
across borders and, by increasing costs, reduce incentives for
firms to hedge risks. Industry estimates suggest that a single
basis point increase in cost resulting from splitting clearing of
interest rate swaps could cost EU firms €22 billion per year
across all of their business.
Delegation of asset management across borders is a
well-established practice. For example, 40% of the assets
managed in the United Kingdom are managed for overseas
clients; around half of this activity is on behalf of clients
outside Europe.
(3)
UK-located asset managers account for
37% of all assets managed in Europe.
(4)
If asset management
were to fragment between the United Kingdom and Europe,
material economies of scale and scope that are currently
achieved by pooling of funds and their management would be
reduced.
Together, these effects could increase the reliance of both
t
he UK and EU economies on their banking systems and
reduce the diversification and resilience of finance.
(2) Macroeconomic shocks that could test the resilience of
the financial system
To maintain consistent provision of financial services to the
UK economy, the financial system must be able to absorb
the impacts on their balance sheets of any adverse
economic shocks that could arise in some scenarios for the
United Kingdom’s withdrawal from the European Union.
The Bank of England’s regular stress testing aims to ensure
that the banking system has the strength to withstand, and
continue to lend in, a broad and severe economic and market
shock.
The United Kingdom’s withdrawal from the European Union
has the potential to affect the economy through supply,
demand and exchange rate channels.
(5)
The supply side of the economy could be disrupted by abrupt
increases in the costs of, or obstacles to, cross-border trade.
Demand could be impacted by the abrupt introduction of
restrictions on exports of financial and other services and
tariffs on trade in goods with the European Union. A reduction
in economic activity in high tax-paying sectors could affect
public finances and spending.
In some scenarios, heightened uncertainty could also reinforce
the existing risk of a fall in appetite of foreign investors for
UK assets. The United Kingdom relies on inflows of overseas
capital to finance its current account deficit — the excess of
investment over domestic saving. That deficit, which stood at
4.4% of GDP in 2016, is financed largely through direct
investment and portfolio investment in the form of long-term
debt and equity (Chart A).
A material reduction in the appetite of foreign investors to
provide finance to the United Kingdom would tighten
financing conditions for UK borrowers and reduce asset prices
(1) Based on Bank analysis of UK-located investment banks’ revenues in 2015 for M&A
and debt/equity issuance activities, using multiple sources.
(2) Based on Bank calculations and multiple sources, including Bank for International
Settlements triennial survey data (2016) which show UK-based dealers account for
82% of European trading in OTC single currency interest rate derivatives.
(3) Sources: Investment Association Annual Survey (2015–2016) and Bank calculations.
(4) Source: Investment Association Annual Survey (2015–2016).
(5) See the May 2017 Inflation Report;
www.bankofengland.co.uk/publications/Documents/inflationreport/2017/may.pdf.
E
xecutive summary
ix
and investment. The effect could be most pronounced in
markets that have recently had greater reliance on access to
overseas capital, such as commercial real estate (CRE).
Around half of the investment in UK CRE since 2015 has been
financed by overseas investors (Chart B).
All else equal, economic shocks like these would probably
depress the exchange rate, putting upward pressure on
inflation. The combination of shocks could therefore possibly
create a more challenging trade-off for monetary policy. The
Monetary Policy Committee would have to make careful
judgements about the net effect of these influences on
demand, supply and inflation.
In these circumstances, the maintenance of financial stability
would require banks to be able to withstand, and continue
lending in, an environment of higher loan impairments,
increased risk of default on other assets, and lower asset prices
and collateral values.
Mitigating risks to financial stability
The FPC will continue to assess the resilience of the UK
financial system to adverse economic shocks that could arise.
The FPC will use the information from its regular stress testing
of major UK banks and building societies. These test banks’
resilience to a range of relevant scenarios, including a snap
back of interest rates, sharp adjustment in UK property
markets, and severe stress in the euro area.
The FPC will continue to assess the suitability of firms’
contingency plans for emerging risks, in the context of
progress on agreements and the continuity of the domestic
regulatory framework. This will draw on reviews by the Bank,
PRA and FCA of firms’ plans, including responses from banks,
insurers and designated investment firms to the PRA’s
April 2017 letter requesting that they summarise their
contingency plans for the full range of possible scenarios
following the United Kingdom’s withdrawal from the
European Union.
2
0
10
0
10
20
30
2011 12 13 14 15 16
Portfolio investment
(b)
F
oreign direct investment (FDI)
Other investment
(c)
Reserves and net derivatives
T
otal net inward financing flow
(d)
Current account deficit
+
–
Per cent of GDP
C
hart A The United Kingdom has relied on material
i
nflows of portfolio investment and FDI to finance its
current account deficit in recent years
Net inward financing flows
(a)
Sources: ONS and Bank calculations.
(
a) This is the change in UK foreign liabilities, less the change in UK foreign assets, for each
category of investment. These data are presented as annual series using four-quarter
averages.
(b) Portfolio investment consists of debt securities (including government debt), equities and
investment fund shares.
(c) Other investment consists mostly of loans and deposits.
(d) The total net inward financing flow is equal in magnitude to the current account deficit
(
plus errors and omissions).
0
5
1
0
15
2
0
25
2003 04 05 06 07 08 09 10 11 12 13 14 15 16 17
B
y overseas investors
B
y domestic investors
Total United Kingdom
£ billions
C
hart B Overseas investors have accounted for around
h
alf of total investment in UK CRE since 2015
UK CRE transactions (gross quarterly flows)
(a)
Sources: The Property Archive and Bank calculations.
(a) Final data points are the sum of three months to May 2017.
The FPC’s approach to addressing risks
from the UK mortgage market
P
art A
T
he FPC’s approach to addressing risks from the UK mortgage market 1
Summary
Buying a house is the biggest investment that many people
will make in their lives, and it is typically financed by debt.
In the United Kingdom, mortgages are households’ largest
liability and lenders’ largest loan exposure.
The FPC’s concern is with risks to the resilience of both
borrowers and lenders that arise from high levels of household
debt. While a significant factor contributing to high levels of
house prices relative to incomes in the United Kingdom has
been the relatively limited growth in the stock of housing, the
main drivers of housing supply are not under the control of the
Bank of England or the FPC. Consumer protection, meanwhile,
remains the responsibility of the FCA.
Historically, the build-up of mortgage debt has been a key
source of risk to financial and economic stability:
• Highly indebted households are more vulnerable to
unexpected falls in their incomes or increases in their
mortgage repayments.
• In an economic downturn, they do all they can to pay their
mortgages, but — as a result — may have to cut spending
sharply, making the downturn worse.
• In doing so, they also increase the risk of losses to lenders,
not just on mortgages, but on other lending too.
Build-ups of mortgage debt can be self-reinforcing. Lenders’
underwriting standards can turn quickly from responsible to
reckless:
• Housing is the main source of collateral in the real economy,
so higher house prices tend to lead to higher levels of
mortgage lending, feeding back into higher valuations.
• In an upturn, when risks are perceived to be low, lenders’
underwriting standards can loosen quickly, as they seek to
maintain or build market share. This increases the supply of
credit further.
To insure against these risks, in June 2014 the FPC introduced
a policy package, designed to prevent a significant increase in
the number of highly indebted households and a marked
loosening in underwriting standards. The two FPC
Recommendations were to:
• limit the proportion of mortgages extended at high
loan to income ratios; and
• promote minimum standards for how lenders test
affordability for borrowers.
These measures were not expected to restrain housing market
activity unless lenders’ underwriting standards eased. They
were put in place as insurance and complement the annual
stress tests of major lenders, which test that lenders can
withstand sharp economic downturns, including large falls in
house prices, while also continuing to lend.
The FPC views its Recommendations as likely to remain in
place for the foreseeable future. It judges that their benefits
would increase if they became more binding relative to
lenders’ underwriting standards. The FPC will continue to
review their calibration on a regular basis.
The FPC has clarified its affordability test Recommendation to
ensure consistency in its application across lenders. The
Committee has recommended that:
• When assessing affordability, mortgage lenders should apply
an interest rate stress test that assesses whether borrowers
could still afford their mortgages if, at any point over the
first five years of the loan, their mortgage rate were to be
3 percentage points higher than the reversion rate specified
in the mortgage contract at origination.
This Recommendation can alternatively be interpreted as
introducing a ‘safety margin’ between current mortgage
payments and current income, also ensuring that the
household sector as a whole is better able to withstand
adverse shocks to income and employment.
The housing market can be a key source of risk to
financial stability
Housing accounts for nearly half of the total assets of
UK households. And around two thirds of house purchases are
financed by mortgage debt.
Housing has been at the heart of many financial crises. Since
the 1950s, there has been a material rise in mortgage debt
across advanced economies, driving a major expansion of the
balance sheet of the financial sector. More than two thirds of
the 46 systemic banking crises for which house price data are
2
Financial Stability Report
J
une 2017
available were preceded by housing boom-bust cycles.
(1)
Mortgage booms have also tended to be followed by periods
of considerably slower economic growth than non-mortgage
credit booms, irrespective of whether a financial crisis
o
ccurred or not.
(
2)
Mortgages are the largest liability of UK households.
They can be a source of risk for borrowers’ resilience and
broader economic activity.
Over the past 20 years, house prices have risen significantly
relative to incomes, so households have to borrow more to
buy a house. The resulting high levels of household debt
expose the UK economy to the risk of sharp cuts in
consumption in the face of shocks to income or interest
rates.
A significant factor contributing to high levels of house prices
relative to incomes in the United Kingdom (Chart A.1)
(3)
has
been the relatively limited growth in the stock of housing. The
absolute level of house prices may further reflect a protracted
decline in interest rates.
Over the past 50 years, the number of new houses built each
year in the United Kingdom has more than halved, from a peak
of 426,000 in 1968 (Chart A.2). Since 1982, this number has
averaged less than 190,000, while the UK population has
increased by around 265,000 per year.
(4)
Partly as a result, the
cost of renting a property — as well as buying it — can be high
relative to household incomes in parts of the country. In the
2016 NMG survey, around one third of respondents who lived
in rented accommodation reported that their rental payments
were 30% or more of their pre-tax incomes. The main drivers
of housing supply are not under the control of the Bank of
England or the FPC, and are partly related to the planning
system.
(5)
Long-term real interest rates have been declining across
advanced economies for several decades. Global structural
factors — such as demographics — are likely to have been the
primary driver of these falls, which have contributed to a rise
in the level of house prices.
(6)
In recent months, annual house
price inflation has weakened; in May 2017 house prices rose at
the slowest annual rate since May 2013.
(7)
But while
respondents to the RICS survey in May 2017 expected the
slowdown to continue in the near term, they expected prices
to rise over the next year.
Building up a deposit to buy a house has become more
difficult. The average house price for first-time buyers
increased from around £50,000 in 1997 to over £200,000 in
2016. Over the same period, the size of a typical deposit for a
first-time buyer increased from less than £5,000 to over
£30,000. The Wealth and Assets Survey
(8)
suggests that only
(1) Crowe, C, Dell’Ariccia, G, Igan, D and Rabanal, P (2011), ‘How to deal with real estate
booms: lessons from country experiences’, IMF Working Paper 11/91;
www.imf.org/external/pubs/ft/wp/2011/wp1191.pdf.
(2) See Jordà, O, Schularick, M and Taylor, A M (2014), ‘The great mortgaging: housing
finance, crises and business cycles’, Federal Reserve Bank of San Francisco Working
Paper 2014–23;
www.frbsf.org/economic-research/publications/working-papers/wp2014-23.pdf.
(3) The national average masks regional differences. At end-2015, the house price to
household income ratio was 3.5 in the North of England and 6.2 in London, and
averaged 4.2 for the United Kingdom.
(4) Over 1981–2016, the size of the average UK household has also fallen, incrementing
the pressure from population growth. For evidence on the impact of supply on
affordability, see Hilber, C A L and Vermeulen, W (2015), ‘The impact of supply
constraints on house prices in England’;
http://onlinelibrary.wiley.com/doi/10.1111/ecoj.12213/abstract.
(5) For more details see Barker, K (2004), Review of housing supply;
https://web.archive.org/web/20080513212848/http:/www.hm-treasury.gov.uk/
consultations_and_legislation/barker/consult_barker_index.cfm#report and
Hilber, C A L and Vermeulen, W (2015), op cit.
(6) See the box ‘Explaining the long-term decline in interest rates’ on pages 8–9 of the
November 2016 Inflation Report; www.bankofengland.co.uk/publications/
Documents/inflationreport/2016/nov.pdf and the box ‘The long-term outlook for
interest rates’ on pages 6–7 of the May 2017 Inflation Report;
www.bankofengland.co.uk/publications/Documents/inflationreport/2017/may.pdf.
(7) Based on the average of the Halifax and Nationwide house price indices.
(8) The Wealth and Assets Survey is a household survey that gathers information on,
among other things, level of savings and debt, saving for retirement, how wealth is
distributed across households and factors that affect financial planning.
0
1
2
3
4
5
6
1973 78 83 88 93 98 2003 08 13
House price to household income
Ratio
Chart A.1 UK house prices have risen significantly
relative to households’ incomes
UK house price to household income ratio
(
a)(b)
Sources: Department for Communities and Local Government, Halifax, Nationwide, ONS and
Bank calculations.
(a) The ratio is calculated as average UK house price divided by the four-quarter moving sum of
gross disposable income of the UK household and non-profit sector per household.
Aggregate household disposable income is adjusted for financial intermediation services
indirectly measured (FISIM) and changes in pension entitlements.
(b) House price is an average of the Halifax and Nationwide indices.
0
50
1
00
150
200
2
50
3
00
350
4
00
450
1952 62 72 82 92 2002 12
C
ompletions of new dwellings
Thousands
C
hart A.2 Over the past 50 years, the number of houses
b
uilt each year has more than halved
Completions of new dwellings in the United Kingdom
(a)
Sources: Department for Communities and Local Government and Bank calculations.
(a) Total number of permanent dwellings completed in the United Kingdom per calendar year.
I
ncludes completions from private enterprises, housing associations and local authorities.
Data for 2016 Q4 and 2017 Q1 assume that completions of new dwellings in the
United Kingdom as a whole have grown in line with those in England. The diamond shows
the 2017 Q1 outturn on an annualised basis for 2017. Data are seasonally adjusted.
P
art A
T
he FPC’s approach to addressing risks from the UK mortgage market 3
around 6% of renters aged 45 or younger have financial assets
worth £30,000 or more, and that only 11% have £15,000 or
more.
T
he increase in house prices relative to incomes in recent
decades has contributed to a rise in UK household
indebtedness, which is currently high by historical
standards. Since the late 1980s, the outstanding stock of
mortgage debt has nearly doubled and represents the majority
of the aggregate household debt to income (DTI) ratio
(Chart A.3).
During the financial crisis, countries that had initially higher
levels of household debt relative to income saw larger falls in
aggregate consumption (Chart A.4), putting downward
pressure on broader economic activity. Analysis of
household-level data also suggests that individual households
with higher mortgage debt relative to income adjust spending
more sharply in response to shocks. For example, data from
the Living Costs and Food Survey show that, during the
financial crisis, the fall in consumption relative to income
among UK households with loan to income (LTI) ratios above
four was around three times larger than the fall for those with
ratios between one and two (Chart A.5). Econometric studies
confirm these results, even after controlling for other
household characteristics.
(1)
Given the ‘full-recourse’ nature of UK mortgage contracts,
borrowers in the United Kingdom typically do all they can
to pay their mortgages rather than default, including
cutting back sharply on spending. In the United Kingdom, if
a borrower defaults on a mortgage and the value of the house
does not cover the outstanding mortgage, the lender has a
claim against other assets of the debtor. This contrasts with
some other jurisdictions, such as the United States, where
‘non-recourse’ mortgages are more widespread.
(1) See Bunn, P and Rostom, M (2015), ‘Household debt and spending in the
United Kingdom’, Bank of England Staff Working Paper No. 554;
www.bankofengland.co.uk/research/Documents/workingpapers/2015/swp554.pdf.
0
20
40
60
80
1
00
1
20
140
160
91 95 99 2003 07 11 15
P
er cent
1987
Household debt to income ratio (excluding mortgages)
H
ousehold debt to income ratio (of which mortgages)
Total household debt to income ratio
Chart A.3 UK household indebtedness is high by
historical standards
UK household debt to income ratio
(
a)(b)(c)(d)
Sources: ONS and Bank calculations.
(a) Total household debt to income ratio is calculated as gross debt as a percentage of a
four-quarter moving sum of gross disposable income of the UK household and non-profit
sector. Includes all liabilities of the household sector except for unfunded pension liabilities
and financial derivatives of the non-profit sector.
(b) Mortgage debt to income ratio is calculated as total debt secured on dwellings as a
percentage of a four-quarter moving sum of disposable income.
(c) Non-mortgage debt is the residual of mortgage debt subtracted from total debt.
(d) The household disposable income series is adjusted for FISIM and changes in pension
entitlements.
10
8
6
4
2
0
2
4
6
8
1
0
0100200300400
H
ousehold debt to income ratio in 2007, per cent
United Kingdom
Adjusted consumption growth 2007–12, per cent
+
–
C
hart A.4 Countries with higher levels of household
d
ebt relative to income saw larger consumption falls in
the crisis
Household debt to income ratio and consumption growth over
2007–12
(a)
S
ources: Flodén (2014) and OECD National Accounts.
(a) Change in consumption is adjusted for the pre-crisis change in total debt, the level of total debt
and the current account balance. See Flodén, M (2014), ‘Did household debt matter in the
Great Recession?’, available at http://martinfloden.net/files/hhdebt_supplement_2014.pdf.
25
20
15
10
5
0
0 to 1 1 to 2 2 to 3 3 to 4 ≥4
P
ercentage point change in consumption to income ratio
Mortgage LTI ratio
–
Chart A.5 UK households with higher levels of mortgage
debt relative to income adjusted spending more sharply
during the crisis
Change in consumption relative to income among mortgagors
with different LTI ratios between 2007 and 2009
(a)(b)(c)
Sources: Living Costs and Food (LCF) Survey, ONS and Bank calculations.
(a) Change in average non-housing consumption as a share of average post-tax income (net of
mortgage interest payments) among households in each mortgage LTI category between
2007 and 2009.
(b) LCF Survey data scaled to match National Accounts (excluding imputed rental income,
income received by pension funds on behalf of households and FISIM). LTI ratio is calculated
using secured debt only as a proportion of gross income.
(c) Repeat cross-section methodology used, with no controls for other factors, or how
households may have moved between LTI categories between 2007 and 2009.
4
Financial Stability Report
J
une 2017
Given the prevalence of short-term fixed-rate mortgage
contracts, UK households are also particularly exposed to
the risk of unexpected changes in interest rates.
(1)
Almost
80% of new mortgage lending in 2016 was either on a fixed
r
ate for a period of less than five years or on a floating rate.
In summary, mortgage debt can be a source of risk for
borrowers’ ability to weather downturns without substantial
cutbacks in their spending. UK household indebtedness is
high, which can be a threat to the wider economy. Highly
indebted households can cut back sharply on spending in
response to adverse shocks to incomes or interest rates,
putting downward pressure on economic activity and reducing
the resilience of the financial system.
Mortgages are the largest loan exposure for UK lenders.
They can be a source of risk for lenders’ resilience,
impairing the provision of credit.
In a severe downturn, some borrowers will be unable to
repay their mortgages even after cutting back on spending,
for example, in the event of a rise in unemployment. The
resulting defaults can affect lenders’ resilience, with
mortgages accounting for around two thirds of major
UK banks’ loans to UK borrowers.
(2)
The proportion of households experiencing repayment
difficulties can rise sharply as the share of income spent on
servicing mortgage debt — also known as the mortgage
debt-servicing ratio (DSR) — increases beyond a certain level
(Chart A.6). Both the average DSR of the UK household
sector as a whole and the proportion of households with high
mortgage DSRs have fallen since the crisis, supported by the
low level of interest rates. But households’ ability to service
their mortgage debt could be challenged by either a rise in
mortgage rates or a fall in incomes. As an illustration, Bank
staff estimate that in the event of a rise in unemployment to
8%, the proportion of households with high DSRs would
double, reaching a level close to that seen in 2007 (Chart A.7).
During the global financial crisis, loss rates on mortgages were
contained, reflecting the sharp fall in interest rates and a rise
in unemployment that was relatively modest given the fall in
economic activity.
(3)
But in the 1990s recession, which was
marked by a significant rise in interest rates and
unemployment, loss rates were more material.
(4)
And results
from stress tests of major UK banks suggest that they could
reach similar levels in future periods of severe stress (Chart
A.8), particularly if house prices were to fall significantly,
increasing lenders’ losses in the event of borrower default.
Significant falls in house prices are highly correlated with
economic downturns, when borrowers are also more likely to
become unemployed and default on their mortgages. In such
a severe stress, lenders are likely to incur larger losses on
lending originally extended at high LTV ratios. This is because
such mortgages are more likely to experience ‘negative equity’
in the event of a fall in house prices, meaning that the value of
the housing collateral will be less likely to cover the mortgage
loan.
0
4
8
12
16
2
0
0
–5 5–10 10–15 15–20 20–25 25–30 30–35 35–40 40–50 50+
NMG survey
(2014–15, arrears 2 months+)
Wealth and Assets Survey
(2010–12, arrears 2 months+)
P
ercentage of mortgagors in arrears
(a)
Mortgage DSR, per cent
C
hart A.6 Households with high debt-servicing ratios
(
DSRs) are more likely to experience repayment difficulties
Households in two-month arrears by mortgage DSR
Sources: NMG Consulting survey, Wealth and Assets Survey and Bank calculations.
(a) The share of mortgagors who have been in arrears for at least two months. The mortgage
DSR is calculated as total mortgage payments (including principal repayments) as a
percentage of pre-tax income. Calculation excludes those whose DSR exceeds 100%.
R
eported repayments may not account for endowment mortgage premia.
(1) This has been a feature of the UK mortgage market for many years and was discussed
in Miles, D (2004), The UK mortgage market: taking a longer-term view;
http://webarchive.nationalarchives.gov.uk/20071204181447/hm-
treasury.gov.uk/consultations_and_legislation/miles_review/consult_miles_index.cfm.
(2) Unless otherwise stated, ‘banks’ or ‘lenders’ refer to all UK banks and building
societies.
(3) The distribution of unemployment was also skewed towards younger households,
among whom the owner-occupier rate is lower.
(4) Less developed credit scoring and credit risk management practices relative to today
also help explain the high loss rates in the early 1990s.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
1991 96 2001 06 11 16
P
ercentages of households
Households with mortgage DSR ≥ 40%
— unemployment at 8%
Households with mortgage DSR ≥ 40% (BHPS/US)
Households with mortgage DSR ≥ 40% (NMG)
Chart A.7 An increase in unemployment could double the
proportion of vulnerable households
Percentage of households with mortgage debt-servicing ratios of
40% or greater
(
a)(b)(c)
Sources: British Household Panel Survey (BHPS), NMG Consulting survey, ONS, Understanding
Society (US) and Bank calculations.
(a) Mortgage DSR calculated as total mortgage payments as a percentage of pre-tax income.
(b) Percentage of households with mortgage DSR above 40% is calculated using British
Household Panel Survey (1991 to 2008), Understanding Society (2009 to 2013), and
NMG Consulting survey (2011 to 2017).
(c) A new household income question was introduced in the NMG survey in 2015. Data from
2011 to 2014 surveys have been spliced on to 2015 data to produce a consistent time series.
Data for 2017 come from the spring survey, while data from previous years come from the
autumn survey.
P
art A
T
he FPC’s approach to addressing risks from the UK mortgage market 5
The provision of high LTV lending has increased from its
post-crisis lows recently, but both the annual average total
volume of high LTV lending and its share of total mortgage
lending remain lower than at any point between 1982 and
2008 (Chart A.9). At the same time, LTV ratios for
outstanding loans have fallen as a result of house price growth
and mortgagors repaying existing debt. As a result, for
example, only 2% of the major six lenders’ stock of mortgages
had an LTV above 90% at end-2016 (Chart A.10).
High mortgage debt can also affect lenders’ resilience
indirectly. In a downturn, highly indebted households
prioritising mortgage payments may default on their other
credit commitments, such as credit cards or personal loans.
Sharp cuts in consumption that amplify the downturn can also
lead to credit losses on other types of lending, for example
loans to businesses.
Overall, in a severe stress, high levels of mortgage debt could
lead to significant losses (both directly and indirectly) and
reduce the resilience of lenders, impairing the provision of
credit to the real economy and further intensifying an
economic downturn.
Self-reinforcing feedback loops between levels of
mortgage lending and house prices can amplify risks to
both borrowers and lenders.
Housing is the main source of collateral in the real
economy, so higher house prices tend to lead to higher
levels of mortgage lending, feeding back into higher
valuations. As valuations increase, rising wealth for existing
homeowners and higher collateral values for lenders can
increase both the demand for, and supply of, credit, leading to
a self-reinforcing loop between levels of mortgage lending and
house prices. Expectations of future house price increases can
also prompt prospective buyers to bring forward house
purchases. The resulting rapid growth in mortgage lending can
amplify risks to both lenders and borrowers.
0.0
0.5
1
.0
1
.5
2.0
2
.5
3.0
1991–95 2009–13 2014 stress
test
2015 stress
test
2016 stress
test
P
er cent
Cumulative five-year loss rates
E
stimated loss rates incurred by insurers
C
hart A.8 Loss rates on UK mortgages could reach
m
aterial levels in a severe stress
Cumulative five-year loss rates on UK mortgages in past
downturns and in stress tests
(a)(b)(c)(d)(e)
Sources: Acadametrics, Bank of England, lenders’ stress-testing submissions and
Bank calculations.
(a) Cumulative loss rates are calculated as cumulative losses divided by average balances.
(b) Losses defined as write-offs for the 1991–95 and 2009–13 periods and impairment charges
for stress-test results.
(c) 1991–95 write-offs include all banks and building societies and an estimate of the losses
borne by the UK insurance industry on loans originated by banks and building societies.
Based on data published by MIAC-Acadametrics.
(
d) Losses in 2009–13 and stress tests include the six major lenders.
(e) Impairments in the 2014 stress test are cumulative over three years.
0
50
100
150
2
00
250
3
00
3
50
400
10
20
3
0
40
50
60
83 87
9
1
95
9
9 2003
07 11
1
5
Percentage of mortgages
N
umber of mortgages (thousands)
1
979
0
Completions <90% LTV (right-hand scale)
Completions ≥90% LTV
(d)
(right-hand scale)
P
roportion ≥90% LTV (left-hand scale)
(e)
C
hart A.9 High LTV mortgage lending remains lower
t
han at any point between 1982 and 2008
New mortgage lending by LTV at origination
(a)(b)(c)
Sources: Council of Mortgage Lenders (CML), FCA Product Sales Database (PSD) and
Bank calculations.
(a) Data are shown as a four-quarter moving average.
(
b) Data include loans to first-time buyers, council/registered social tenants exercising their right
to buy and home movers.
(c) The PSD includes regulated mortgage contracts only.
(d) The number of mortgage loans with ≥90% LTV is calculated using the aggregate number of
mortgages from the CML and the proportion of mortgages with ≥90% LTV from the PSD.
(e) PSD data are only available since 2005 Q2. Data from 1993 to 2005 are from the Survey of
Mortgage Lenders, which was operated by the CML, and earlier data are from the 5% Sample
S
urvey of Building Society Mortgages. The data sources are not directly comparable: the
PSD covers all regulated mortgage lending whereas the earlier data are a sample of the
mortgage market. Data for the first three quarters of 1992 are missing, chart values are
interpolated for this period.
0
10
20
30
40
50
60
70
80
90
100
10 11 12 13 14 15 16
Per cent of mortgage book
2009
0% < LTV ≤ 50%
50% < LTV ≤ 75%
75% < LTV ≤ 90%
90% < LTV ≤ 95%
95% < LTV ≤ 100%
100% and above
Chart A.10 The LTV distribution of the stock of
mortgages has improved since the crisis
UK mortgage books by indexed LTV
(
a)
Sources: PRA regulatory returns and Bank calculations.
(a) Peer group accounts for around 74% of total UK mortgages and includes the major
UK lenders.
6
Financial Stability Report
J
une 2017
In an upturn, when risks from credit losses are perceived to
be low, the underwriting standards lenders apply to decide
on what terms to lend can deteriorate quickly as they seek
to maintain or build market share. This increases the
s
upply of credit further. Underwriting standards on
UK mortgages weakened in the lead-up to the financial crisis,
contributing to the growth in mortgage lending and house
prices. Across a range of metrics, underwriting standards are
now more robust relative to the period before the crisis. But
market contacts suggest that lending conditions in the
mortgage market are becoming easier and competitive
pressures in the mortgage market remain. Mortgage spreads
over risk-free rates have fallen materially since their peak
in 2012 (Chart A.11). Lenders are also extending an increasing
proportion of mortgages without fees. Forty-six per cent
of mortgages were extended without fees in the first part
of 2017, compared to 37% in 2016 and just 12% in 2011
(Chart A.12).
A similar feedback loop between house prices and credit also
arises in a downturn. An economic slowdown can reduce
house prices. Due to the role of housing as collateral, lower
house prices reduce the demand for, and supply of, credit.
Expectations of further price reductions, which can result in
sales of houses at heavily discounted prices (‘fire sales’), can
further amplify house price falls, reinforcing the adverse
feedback loop. The resulting deterioration in borrowers’
and lenders’ resilience will intensify a downturn
(Figure A.1).
Growth in the private rental sector in recent years may
have led to growing risks of amplified house price cycles
from leveraged buy-to-let investors.
(1)
The share of
households in the private rental sector rose from around 10%
in 2002 to 20% in 2016. Buy-to-let investors do not live in
the house that they rent out and their behaviour is more likely
to be driven by their expected returns on their housing
investment than that of owner-occupiers. But if either house
prices or the income received from rental payments were to
fall materially, there is a risk that some leveraged investors
may look to sell their properties quickly, reinforcing house
price falls in a downturn.
The size of the buy-to-let segment of the mortgage market
has almost doubled since the period before the crisis
(Chart A.13). So the impact of a growing share of leveraged
investors on the dynamics of the broader market in a stress
Expectations of further price
reductions and ‘fire sales’
Collateral effect
Adverse
s
hock
House
p
rice fall
R
eduction in
demand for and
supply of credit
Figure A.1 Feedback loops between mortgage credit and
house prices can amplify a downturn
0
5
10
15
20
25
30
3
5
4
0
45
50
2011 12 13 14 15 16 17
P
er cent
Proportion of £0 fee new mortgages
C
hart A.12 The proportion of mortgages with no fees
has increased
Proportion of new mortgages with no fees
(a)
Sources: Moneyfacts and Bank calculations.
(a) The proportion of £0 fee products in each year is calculated relative to the total number of
new mortgages offered during the year. The proportion in 2017 is calculated based on data
from January to April 2017.
0
50
100
150
200
250
3
00
350
4
00
450
05 06 07 08 09 10 11 12 13 14 15 16 17
B
asis points
2004
M
ortgage spreads
— owner-occupier
(a)
Mortgage spreads
— buy-to-let
(b)(c)
Chart A.11 Mortgage spreads have fallen
Mortgage rates on owner-occupier and buy-to-let lending relative
to risk-free rates
Sources: Bank of England, Bloomberg, Council of Mortgage Lenders, FCA Product Sales
Database, Moneyfacts and Bank calculations.
(a) The overall spread on residential mortgage lending is a weighted average of quoted
mortgage rates over risk-free rates, using 90% LTV two-year fixed-rate mortgages and 75%
LTV tracker, two and five-year fixed-rate mortgages. Spreads are taken relative to gilt yields
of matching maturity for fixed-rate products. Spreads are taken relative to Bank Rate for the
tracker product. Weights are based on relative volumes of new lending. The Product Sales
Database includes regulated mortgages only.
(b) The spread on new buy-to-let mortgages is the weighted average effective spread charged
on new floating and fixed-rate non-regulated mortgages over risk-free rates. Spreads are
taken relative to Bank Rate for the floating-rate products. The risk-free rate for fixed-rate
mortgages is calculated by weighting two-year, three-year and five-year gilts by the number
of buy-to-let fixed-rate mortgage products offered at these maturities.
(c) Buy-to-let data are only available from 2007 as they are sourced from the Bank of England’s
Mortgage Lenders and Administrators Return (MLAR) which started being collected in 2007.
(1) See Section 3.3 of Bank of England (2016), ‘The Financial Policy Committee’s powers
over housing policy instruments’, A draft Policy Statement;
www.bankofengland.co.uk/financialstability/Documents/fpc/draftpolicystatement181
116.pdf.
P
art A
T
he FPC’s approach to addressing risks from the UK mortgage market 7
has yet to be tested. But there is evidence of this channel
operating in the United States in the financial crisis. In those
US states that experienced the largest housing booms and
busts, at the peak of the market almost half of mortgage
o
riginations were associated with investors.
(
1)
The FPC has taken action to mitigate risks to both
borrowers’ and lenders’ resilience
Given the importance of the UK mortgage market for
financial stability, the FPC and the Bank have taken action
in recent years to mitigate risks to both borrowers’
resilience, where this can impact broader economic activity,
and to lenders’ resilience.
In June 2014, the FPC introduced two Recommendations,
which: limit the proportion of mortgages with high LTI ratios;
and promote minimum standards for how banks test
affordability for borrowers.
Also in 2014, following a Recommendation by the FPC,
the Bank introduced annual stress tests of the UK banking
system.
In September 2016, the PRA published a Supervisory
Statement setting out its expectations for minimum
underwriting standards on buy-to-let mortgages, in particular
on how lenders test affordability.
Parliament has also given the FPC powers of Direction,
(2)
which can cover both owner-occupier and buy-to-let
mortgage lending.
Table A.1 summarises the Bank’s toolkit to deal with risks
from the UK mortgage market.
Insuring against risks to borrowers’ resilience
The FPC’s Recommendations insure against a further
significant rise in the number of highly indebted households
and a marked loosening in underwriting standards.
The ‘affordability test’ Recommendation was designed to
insure against a loosening in lenders’ standards for assessing
mortgage affordability. It builds on the FCA’s rules that
require lenders to assess whether prospective borrowers could
afford their mortgage, taking into account their income,
spending patterns and potential future interest rate increases.
At the time of the original Recommendation in 2014, most
major lenders tested whether prospective borrowers could
afford their mortgages assuming a stressed mortgage rate of
around 7%. That compared with prevailing mortgage
reversion rates
(3)
in the region of 4%–4½%. In order to insure
against a relaxation of those standards, the FPC recommended
that all mortgage lenders should assess whether borrowers
could still afford their mortgages if Bank Rate were to increase
by 3 percentage points — the idea being that this increase
would feed through to lenders’ reversion rates, to result in a
stressed mortgage rate in the region of 7%.
The affordability test can alternatively be interpreted as
introducing a ‘safety margin’ between current mortgage
payments and current incomes. This margin seeks to ensure
that the household sector is better able to withstand
fluctuations in income and employment. Bank staff estimate
that the margin of safety created by assessing affordability
(1) See Haughwout, A, Lee, D, Tracy, J and van der Klaauw, W (2011), ‘Real estate
investors, the leverage cycle, and the housing market crisis’, Federal Reserve Bank of
New York Staff Report No. 514;
www.newyorkfed.org/medialibrary/media/research/staff_reports/sr514.pdf.
(2) The FPC has two main powers. It can make Recommendations to anybody, including
to the PRA and FCA. It can also, where the Government has given the FPC a power of
Direction, direct the regulators to implement a specific measure to further the FPC’s
objectives.
(3) The reversion interest rate is the (typically floating) rate to which a mortgage reverts
after an initial contractual period that is often based on a fixed interest rate.
0.0
0
.2
0.4
0
.6
0
.8
1
.0
1
.2
1.4
0
2
4
6
8
10
12
14
16
1
8
2
0
2000 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17
£ trillionsPer cent
Owner-occupiers (right-hand scale)
Buy-to-let (right-hand scale)
Buy-to-let share (left-hand scale)
Chart A.13 The size of the buy-to-let segment of the
mortgage market has almost doubled since the period
before the crisis
Composition of the outstanding mortgage stock
(a)
Sources: Bank of England, Council of Mortgage Lenders and Bank calculations.
(a) Lending to owner-occupiers is calculated as outstanding lending to individuals secured on
d
wellings less outstanding lending secured on buy-to-let properties.
T
able A.1 The Bank has an extensive toolkit to address risks from
t
he UK mortgage market
Owner-occupier Buy-to-let
L
oan to value policies
L
oan to value limit Loan to value limit
Affordability policies Loan to income limit
*
Affordability test
*
A
ffordability test
*
I
nterest coverage ratio limit
D
ebt to income limit
Capital policies Stress-testing framework
*
UK countercyclical capital buffer (CCyB) rate
*(a)
S
ectoral capital requirements
* Policies marked with an asterisk are currently in place.
(a) The CCyB is not a power of Direction, but the FPC is the designated authority to set the UK CCyB rate.
Colour scheme:
FPC’s power of Direction
FPC’s Recommendation
PRA’s Supervisory Statement
The Bank’s annual stress test is conducted under the guidance of the FPC and
the PRC.
8
Financial Stability Report
J
une 2017
against a 300 basis point rise in Bank Rate is equivalent to that
needed by the household sector to be better able to withstand
a 2–3 percentage point rise in unemployment.
(1)
T
he ‘LTI flow limit’ Recommendation limits the number of
mortgages extended at LTI ratios of 4.5 or higher to 15% of a
lender’s new mortgage lending. The 4.5 multiple was
calibrated to ensure that, at a stressed mortgage rate of 7%
and a typical mortgage term of around 25 years, mortgagors’
stressed DSRs would not exceed 35%–40%. That is the point
beyond which the proportion of mortgagors that start
experiencing repayment difficulties can rise sharply
(Chart A.6). The 15% flow at or above the 4.5 threshold
ensures that access to high LTI mortgages remains for those
borrowers who can afford it.
The affordability test and the LTI flow limit complement
each other in protecting households’ ability to service their
debt. The affordability test effectively sets an LTI cap for each
borrower that depends primarily on the term of the mortgage,
or ‘tenor’, and the borrower’s spending commitments. The
relationship between the effective LTI cap and the mortgage
term is illustrated in Chart A.14. The swathe reflects
variations in the effective cap, depending on the borrower’s
spending commitments and the precise stressed interest rate
used by lenders to assess affordability.
For borrowers seeking a relatively short mortgage term, the
affordability test effectively places a lower cap on LTIs than
the threshold implied by the LTI flow limit. This is because,
at short tenors, a given loan amount will have higher
debt-servicing costs due to higher capital repayments.
In 2016, less than 20% of new mortgages had a tenor of
fifteen years or less.
Chart A.14 also shows that the LTI flow limit can act as a
simple backstop to the more complex affordability test: the
LTI flow limit would be more likely to bind if mortgage terms
increased, or if lenders loosened the standards through which
t
hey assess affordability (eg the approach to accounting for
other spending commitments). Other things equal, such
changes would move the effective LTI cap implied by the
affordability test towards the top of the swathe.
The FPC judges that its Recommendations have had
only a modest impact on mortgage lending to date.
In November 2016, the FPC reviewed the two
Recommendations. The Committee assessed that, since
their introduction in June 2014, they had had only a modest
impact on mortgage lending, as lenders’ own underwriting
standards had not loosened. The Committee also judged that
the Recommendations continued to insure against a
deterioration in underwriting standards.
(
2)
The Committee
retains both of these judgements.
LTI flow limit. In aggregate, lenders are advancing around 10%
of new mortgages at LTIs at or above 4.5 (Chart A.15). So
there remains headroom for further high LTI lending in
aggregate. In February 2017, the PRA changed the relevant
measure from a fixed quarterly limit to a four-quarter rolling
limit, which should further help lenders manage their business
pipeline.
(3)
One feature of recent lending has been a ‘bunching’ of loans
just below the FPC’s 4.5 LTI limit. In part, this is likely to
represent some individuals being constrained to smaller loans
than they would have otherwise obtained. Bank staff estimate
the size of this impact to be small in aggregate: for example, if
the share of borrowers with an LTI between 4 and 4.5 were to
be returned to its level before the FPC Recommendations were
made, and the remaining borrowers in that category were to
obtain an LTI of 5 instead, the value of total mortgage lending
would increase by less than 1%.
Affordability test. The impact of the FPC’s affordability test is
more difficult to assess because the total number of
prospective borrowers who fail the test is not directly
observable. Nevertheless, in November 2016 the FPC judged
that the Recommendation had not been excluding a
significant number of borrowers. This was based on an
0
1
2
3
4
5
6
7
0510 15 20 25 30 35
Maximum LTI implied by FPC affordability test
FPC LTI flow limit
LTI
Mortgage term
Chart A.14 The affordability test and LTI flow limit
complement each other in protecting households
Relationship between the affordability test and the LTI flow limit
in constraining lending
(a)(b)
Source: Bank of England.
(a) Swathe for affordability test assumes borrowers have 30% to 50% of gross income available
to support mortgage repayments, and lenders assess affordability at stress interest rates of
6.75% to 7%.
(b) The FPC flow limit restricts the share of new mortgages at LTIs of 4.5 or greater to 15%.
(1) For example, analysis by Bank staff suggests that the proportion of households that
would have a DSR greater than 40% in the face of an interest rate shock of 300 basis
points is broadly equivalent to the proportion of households that would have a DSR
greater than 40% in the face of an unemployment shock of around 3 percentage
points. And empirical relationships between aggregate arrears and macroeconomic
variables suggest that the proportion of households that would be in arrears if
interest rates were to rise by 300 basis points is broadly equivalent to the proportion
of households that would be in arrears if unemployment increased by just under 2%.
(2) See the November 2016 Report for a more detailed discussion of the impact of the
policies; www.bankofengland.co.uk/Pages/reader/index.aspx?pub=fsrnov16&page=1.
(3) For further details on the PRA’s change, see Bank of England (2017), ‘Amendments to
the PRA’s rules on loan to income ratios in mortgage lending’, PRA Policy Statement
PS5/17; www.bankofengland.co.uk/pra/Documents/publications/ps/2017/ps517.pdf.
P
art A
T
he FPC’s approach to addressing risks from the UK mortgage market 9
analysis of trends in mortgage applications and information
received from a sample of lenders, suggesting that the
calibration of the affordability test was not resulting in a
material proportion of mortgage enquiries being rejected, even
prior to the formal application stage. Two further pieces of
evidence support this conclusion:
• First, while there has been a long-run trend towards longer
mortgage terms since the crisis, there has been no
acceleration in that trend since the introduction of the
affordability test (Chart A.16). Were the policy to be
excluding a large number of prospective mortgagors,
borrowers could seek to pass the test by lengthening
mortgage tenor, which lowers monthly repayments.
• Second, first-time buyers, who might have been expected to
be most affected by any measure that restricts loan size
relative to income, have maintained their share of total
mortgage lending at around a third since 2014.
The FPC has further considered the possible impact of its
Recommendations in different hypothetical scenarios.
The Committee judges that, in the event that they were to
become more binding relative to lenders’ own underwriting
standards, their benefits would also increase. The FPC
expects them to remain in place for the foreseeable future.
As an example, in a hypothetical ‘upside’ scenario where a
loosening of underwriting standards pushes mortgage lending
higher and leads to an increase in the aggregate house price to
income ratio, the absence of policies would lead to
a significant increase in the number of highly indebted
households. Specifically, the share of new mortgages
extended at an LTI multiple at or above 4.5 would be 40% by
the end of the scenario, compared with 13% if the policies
were in place (Chart A.17). Over time, this would lead to a
significant deterioration in the distribution of the stock of
household debt.
The benefits of the FPC’s Recommendations under this
scenario include:
• By limiting the number of highly indebted households, the
policies reduce the potential for cuts in consumption in
response to adverse shocks. There are significant
uncertainties around the relationship between household
indebtedness and cuts in consumption. But mapping
estimates based on international evidence (Table A.2) onto
the distribution of debt with and without policies suggests
0
2
4
6
8
10
12
1
4
1
6
1
8
2005 06 07 08 09 10 11 12 13 14 15 16
Per cent of new mortgages
4 ≤ LTI < 4.5
LTI ≥ 4.5
LTI ≥ 5
C
hart A.15 There remains headroom for further high LTI
l
ending in aggregate
Flow of new mortgages by LTI
(a)(b)
Sources: FCA Product Sales Database and Bank calculations.
(a) The Product Sales Database includes regulated mortgages only.
(b) LTI ratio calculated as loan value divided by the total reported gross income for all named
borrowers. Chart excludes lifetime mortgages, advances for business purposes and
remortgages with no change in the amount borrowed.
0
10
20
30
40
50
60
70
2005 06 07 08 09 10 11 12 13 14 15 16 17
Per cent of new mortgages
Mortgage term ≥ 35
30 ≤ Mortgage term < 35
25 ≤ Mortgage term < 30
Chart A.16 There has been a long-run trend towards
longer mortgage terms, but no acceleration more
recently
Share of new mortgages by mortgage term
(a)(b)
Sources: FCA Product Sales Database and Bank calculations.
(a) The Product Sales Database includes regulated mortgages only.
(b) Chart excludes lifetime mortgages, advances for business purposes and remortgages with no
change in the amount borrowed.
2016 Q4
024
Loan to income ratio (LTI)
6
0.0
0.2
0.4
0.6
S
hare of new mortgage lending
LTI = 4.5
(FPC flow limit)
2023 Q4 (no policy)
2023 Q4 (with policy)
Chart A.17 In an ‘upside’ scenario, the flow of mortgage
lending would be skewed towards higher LTIs without
FPC policies in place
LTI distribution of new mortgage lending
(a)(b)
Sources: FCA Product Sales Database and Bank calculations.
(a) The Product Sales Database includes regulated mortgages only.
(b) LTI distribution of new mortgage lending in 2016 Q4 and at the end of an ‘upside’ seven-year
scenario, with or without FPC Recommendations in place.
1
0 Financial Stability Report
J
une 2017
that the fall in aggregate consumption in the event of an
adverse shock could, in this scenario, be up to around 20%
larger.
• By limiting the deterioration in the stock of household debt,
the policies further reduce the probability that households
default on their mortgages.
• Finally, the Recommendations can also reduce the size of
some adverse shocks in the future. For example, by
preventing a marked loosening in underwriting standards,
they reduce the risk of a self-reinforcing feedback loop
between mortgage lending and house prices, which could
amplify any fall in house prices.
The macroeconomic costs of the FPC’s Recommendations
arise from their impact on housing market activity — and
therefore broader economic activity — in these scenarios.
• Because they constrain mortgage approvals in scenarios in
which underwriting standards loosen, the FPC’s policies may
also have some effect on consumer spending. For example,
moving house is often associated with a greater propensity
to purchase durable goods, such as furniture and household
appliances.
(1)
In the hypothetical ‘upside’ scenario, the
Recommendations are estimated to reduce the level of
nominal GDP by around 0.2% by the end of the scenario.
• The Committee further judges that it is unlikely that a
restriction on mortgage credit supply would have a material
effect on the economy’s longer-term growth rate or
productive capacity. So the costs of the policies would be
temporary, while the benefits of increased resilience would
persist.
In reaching its judgement, the FPC considered different
scenarios constructed by staff. These pointed to the benefits
of policy rising as it became increasingly binding relative to
lenders’ own underwriting standards.
The FPC’s existing Recommendations cover the
owner-occupier mortgage market. The FPC also has a
power of Direction over interest coverage ratio (ICR)
limits on buy-to-let mortgages. It has not yet used this
power as it judges that the PRA Supervisory Statement
currently provides adequate insurance against a
loosening of underwriting standards in the buy-to-let
market (Box 3).
Insuring against risks to lenders’ resilience
The FPC’s Recommendations protect the resilience of
lenders by reducing the probability that new borrowers will
default on their mortgages. They complement the
framework of bank capital requirements, which seeks to
ensure that lenders can withstand sharp economic
slowdowns, including large falls in house prices, while
continuing to lend.
Banks are required to hold more capital against riskier loans.
(
2)
For example, lenders allocate more capital to mortgages with
higher LTV ratios (Chart A.18) because, in the event of a
severe house price fall, higher LTV ratios would result in larger
losses on defaulted mortgages.
The Bank’s stress-testing framework assesses whether banks
hold enough capital to withstand a severe stress and continue
to lend to the real economy. The ‘annual cyclical scenario’
includes sharp falls in house prices (eg a 33% fall in the 2017
scenario). These stress scenarios are more severe if risks from
the housing and mortgage market are judged to have
increased, other things equal.
(3)
And, for a given
macroeconomic shock, losses in the stress test will increase
with the riskiness of lenders’ portfolios (eg due to more high
LTV lending). Taken together, this leads to a countercyclical
capitalisation of the banks, strengthening their resilience
against risks from the mortgage market.
Where stress tests show banks need bigger buffers of capital
to absorb the stress scenarios, the FPC can take action by
raising the UK countercyclical capital buffer rate, and the
T
able A.2 More highly indebted mortgagors made larger
s
pending cuts during the crisis
Cuts in consumption between 2007 and 2009 among mortgagors with
different LTI ratios
L
TI ratio United Kingdom
(a)
Denmark
(b)(c)
Norway
(b)(d)
(per cent) (per cent) (per cent)
0 to 1 -1.4 1.2 1.9
1
to 2 -4.2 1.9 -6.3
2
to 3 -7.0 1.0 -11.5
3 to 4 -9.8 -2.3 -21.3
4 to 5 -12.6 -5.8 -28.9
5
to 6 n.a. -7.9 n.a.
Sources: Andersen, Duus and Jensen (2014), Bunn and Rostom (2015), Fagereng and Halvorsen (2016) and
Bank calculations.
(a) Predicted change in non-housing consumption between 2006–07 and 2009–10 associated with mortgage
LTI ratio in 2006–07. Estimated using a synthetic panel approach with a range of control variables. See
Table 2 in Bunn, P and Rostom, M (2015), ‘Household debt and spending in the United Kingdom’,
Bank of England Staff Working Paper No. 554;
www.bankofengland.co.uk/research/Documents/workingpapers/2015/swp554.pdf.
(b) Average predicted change in consumption between 2007 and 2009 as a share of income in 2007 for
households. LTI calculated using total mortgagor debt, including unsecured loans. Estimated using
econometric analysis of household-level administrative data featuring a range of control variables.
(
c) See Chart 4 in Andersen, A L, Duus, C and Jensen, T L (2014), ‘Household debt and consumption during the
financial crisis’, Danmarks Nationalbank Monetary Review, 1st Quarter;
www.nationalbanken.dk/en/publications/Documents/2014/03/Household_MON1_2014.pdf.
(d) See Fagereng, A and Halvorsen, E (2016), ‘Debt and household consumption responses’, Norges Bank Staff
Memo No. 1; http://static.norges-bank.no/contentassets/1dae87a5ddd94b4d871712bcf8791196/
staff_memo_1_2016.pdf. Figures provided by the author to allow comparison with Andersen, Duus and
Jensen (2014).
(1) See the box ‘The housing market and household spending’ on pages 18–19 of the
November 2016 Inflation Report; www.bankofengland.co.uk/publications/
Documents/inflationreport/2016/nov.pdf.
(2) Lenders have to be funded by capital in proportion to the risks they take. To compute
the required amount of capital, lenders ‘risk weight’ their loans. They calculate risk
weights either by using a ‘standardised approach’ set internationally or, if allowed by
their national supervisor, by using their own risk parameters — the ‘internal ratings
based’ (IRB) approach’.
(3) See Bank of England (2015), ‘The Bank of England’s approach to stress testing the
UK banking system’; www.bankofengland.co.uk/financialstability/Documents/
stresstesting/2015/approach.pdf.
P
art A
T
he FPC’s approach to addressing risks from the UK mortgage market 11
Prudential Regulation Committee (which is responsible for
promoting the safety and soundness of individual banks)
(1)
can raise the regulatory capital buffers of individual lenders.
The FPC has not yet used its powers of Direction over
LTV limits. The Committee judges that the measures in
place are proportionate to current risks (Box 4).
The FPC regularly reviews the calibration and
implementation of its Recommendations
The FPC has withdrawn and replaced the affordability test
Recommendation in order to promote consistency of
implementation across lenders, with the aim of insuring
against a future loosening in underwriting standards
(Box 5). The new Recommendation states that:
When assessing affordability, mortgage lenders should apply an
interest rate stress test that assesses whether borrowers could
still afford their mortgages if, at any point over the first five years
of the loan, their mortgage rate were to be 3 percentage points
higher than the reversion rate specified in the mortgage contract
at the time of origination (or, if the mortgage contract does not
specify a reversion rate, 3 percentage points higher than the
product rate at origination). This Recommendation is intended
to be read together with the FCA requirements around
considering the effect of future interest rate rises as set out in
(1) On 1 March 2017, as the PRA ceased to be a subsidiary of the Bank and its functions
were fully transferred to the Bank, the PRA Board became the Prudential Regulation
Committee.
0
10
20
30
40
50
60
70
0–50 50–60 60–70 70–80 80–90 90–100 100+
Mortgage LTV (per cent)
IRB weighted average
± 1 standard deviation
(b)
IRB weighted average
(c)
Per cent
Chart A.18 Lenders allocate more capital to mortgages
with higher LTV ratios
Risk weights on UK ‘prime’ mortgages by LTV
(a)
Sources: PRA regulatory returns and Bank calculations.
(a) Chart shows risk weights on ‘prime’ mortgages calculated under the internal ratings based
(IRB) approach; it excludes buy-to-let mortgages, which tend to have higher risk weights.
Non-defaulted mortgages only.
(b) Shows ± 1 standard deviation based on the sample of lenders.
(c) Average of UK lenders that use the IRB approach to calculate mortgage risk weights
(ten UK lenders including the six major lenders) weighted by exposure.
Box 3
PRA Supervisory Statement on underwriting
standards for buy-to-let mortgages
(1)
In 2016 the PRA undertook a review of lenders’ underwriting
standards in the buy-to-let market.
(
2)
The review revealed
that some lenders were applying standards that were
somewhat weaker than those prevailing in the market as a
w
hole. A number of lenders and other firms planned to grow
their gross buy-to-let lending significantly, so there was a risk
that competitive conditions would lead more firms to relax
underwriting standards to implement their plans.
In response, the PRA Board issued a Supervisory Statement
to clarify its expectations for underwriting standards in the
buy-to-let market. It set the baseline minimum stressed
interest rate to be used in the affordability test at the higher of
5.5% or a 2 percentage point increase in buy-to-let mortgage
interest rates.
Although the stressed interest rate is lower than that applied
to owner-occupier mortgages in the FPC’s Recommendation,
lenders tend to test affordability using ICR thresholds of at
least 125% (and, more recently, industry standards have been
moving to 145% due to tax changes). So, effectively, these
affordability assessments require borrowers’ expected
monthly rental income from the buy-to-let property to
include at least a 25% buffer over the monthly mortgage
interest payment estimated using the stressed interest rate
above.
In addition, LTV ratios at origination in excess of 75% are
much less common in buy-to-let than in owner-occupier
mortgages; in 2016, they accounted for around 10% and 45%
of new mortgages respectively. Buy-to-let loans therefore
typically start with a larger equity cushion, providing greater
resilience to credit risk for lenders.
The FPC considers that no action beyond the PRA Supervisory
Statement is warranted at this stage for macroprudential
purposes. The growth of buy-to-let mortgage lending slowed
significantly in April 2016 and has been weak since, due to the
combination of changes to stamp duty land tax, changes to
mortgage interest tax relief and the implementation of the
PRA Supervisory Statement. The FPC will continue to monitor
developments in the buy-to-let market and any potential
threats to financial stability.
(1) Bank of England (2016), ‘Underwriting standards for buy-to-let mortgage contracts’,
P
RA Supervisory Statement SS13/16; www.bankofengland.co.uk/pra/Documents/
publications/ss/2016/ss1316.pdf. See the Record of the FPC meeting on 23 March 2016
for more details on the FPC’s reaction to the PRA’s review; www.bankofengland.co.uk/
publications/Documents/records/fpc/pdf/2016/record1604.pdf.
(2) The review assessed the lending plans of the top 31 lenders in the industry, which
represented over 90% of buy-to-let lending.
1
2 Financial Stability Report
J
une 2017
Box 4
Powers of Direction over LTV limits
The FPC has not yet used its powers of Direction over
LTV limits, either for owner-occupier or buy-to-let mortgages.
High LTV lending primarily poses risks to lenders. Resilience of
l
enders to these risks is afforded by the Bank’s stress-testing
framework, the broader capital regime and the FPC’s existing
Recommendations.
The Committee is also mindful that the LTI flow limit can
effectively constrain the proportion of high LTV lending in
some instances. An individual borrower’s LTI and LTV are
mechanically linked through the house price to income ratio:
Mortgage loan
=
Mortgage loan
X
Value of house
Income Value of house Income
So for a given house price to income ratio, the greater a
borrower’s LTI, the greater their LTV.
Due to substantial variation in the ratio of house prices to
incomes across different regions of the country, the LTI flow
limit will not constrain high LTV lending in all circumstances.
For example, in the South East, where house prices are higher
relative to incomes, borrowers tend to have higher LTI ratios.
Limits on high LTI lending effectively constrain high LTV
lending too. In other regions, where house prices are lower
relative to incomes, underwriting standards on LTI ratios are
less constraining. This is illustrated in Chart A, which shows
that more mortgagors in London and the South East have
higher LTI ratios.
The FPC will keep under review the risks to lenders stemming
from high LTV mortgage lending. It could in future consider
employing LTV limits to insure against risks on owner-occupier
or buy-to-let mortgages, as other macroprudential authorities
have done in a number of countries.
(1)
However, the FPC
judges that the combination of stress testing and bank capital
requirements, alongside its existing Recommendations in the
mortgage market, build a degree of lender resilience that is
proportionate to current risks.
MCOB 11.6.18(2). This Recommendation applies to all lenders
which extend residential mortgage lending in excess of
£100 million per annum.
The FPC will continue to review the calibration of both
the affordability test and the LTI flow limit regularly.
While the current policy package is likely to remain in place for
the foreseeable future, the FPC will continue to review the
calibration of the affordability test and LTI flow limit on a
regular basis.
(1)
The following principles will guide its approach
to considering the calibration of the policies in the future.
The FPC will review the calibration of the two
Recommendations together. The affordability test and the
LTI flow limit complement one another. For a given mortgage
term and stressed mortgage rate, the two provide a broadly
equivalent constraint on the amount of borrowing prospective
mortgagors could take on relative to their incomes
(Chart A.14). As Bank Rate rises in the future, the affordability
test could become more constraining on lending relative to
the LTI flow limit. The Committee therefore intends to
consider the balance between the two policies if and when
Bank Rate rises to a level close to 1%.
The calibration of the policies will depend on the FPC’s
judgement around risks to both interest rates and incomes.
Increases in Bank Rate lead to higher mortgage rates and, so,
higher mortgage payments. Higher unemployment raises the
(1) For international evidence on the impact of macroprudential measures in the housing
market, see Box 1 in Bank of England (2016), ‘The Financial Policy Committee’s
powers over housing policy instruments’, A draft Policy Statement;
www.bankofengland.co.uk/financialstability/Documents/fpc/draftpolicystatement181
116.pdf.
LTV (per cent)
0
30
60
9
0
120
Other UK regions
L
ondon and South East
0510
House price to income ratio
15 20
LTI = 4.5
L
TIs increase with
higher LTV and
higher HPI
C
hart A There is greater scope for an increase in mortgage
L
TVs in regions outside London and the South East
Mortgages’ LTV ratio and house price to income ratio
(a)(b)
Sources: FCA Product Sales Database and Bank calculations.
(
a) The Product Sales Database includes regulated mortgages only.
(b) The chart shows the house price to income ratio (HPI) and LTV combinations for a
representative sample of new mortgages (excluding remortgages) in 2016. It splits off London
and the South East (red dots) from other UK regions (blue dots). Each dot represents a
mortgage. It also shows a constant-LTI line (LTI = 4.5) for various HPI and LTV combinations,
reflecting their relationship: LTI = HPI * LTV. In London and the South East, where HPIs are
higher, limits on high LTI lending effectively constrain LTVs.
(1) The FPC has a duty to review its Recommendations at regular intervals and consider
whether they should remain in place or be withdrawn.
P
art A
T
he FPC’s approach to addressing risks from the UK mortgage market 13
probability that borrowers suffer a reduction to their income,
reducing the available resources to meet mortgage costs.
Both types of shock increase mortgagors’ DSRs, other things
equal, and high DSRs have historically been associated with
increases in arrears rates and falls in consumption.
The FPC will draw upon a range of indicators to inform its
judgements around risks to interest rates and incomes.
These include, but are not limited to: the distance between
variables’ current values and their estimated equilibrium
values; and historical and international evidence on the scale
of potential shocks. When assessing potential future
changes to interest rates, the Committee is more likely to
be guided by slow-moving, ‘structural’ measures of interest
rates than by market expectations of future interest rates.
Given the long-term nature of mortgage contracts, it judges
that it would be imprudent to rely too heavily on potentially
volatile market-implied measures.
The FPC will consider the overall volume of mortgage
lending in calibrating the LTI flow limit. The LTI flow limit is
expressed as a share of new mortgage lending. If the total
volume of mortgage lending were to increase materially, the
FPC could consider recalibrating the LTI flow limit to ensure
that the overall number of highly indebted households does
not become excessively high.
Box 5
The affordability test Recommendation
The new Recommendation states that lenders should test
affordability by considering a 3 percentage point increase in
their current reversion rate (for many lenders this is the
standard variable rate, or ‘SVR’), while the previous
R
ecommendation stated that lenders should consider a
3 percentage point increase in Bank Rate.
So far, lenders have been using a range of approaches to
calculate the stressed interest rate at which they test
affordability — so there has been a lack of consistency across
the market. Because the previous Recommendation was
expressed as a 3 percentage point change in Bank Rate, it was
open to different interpretations by lenders. For example,
lenders could make different assumptions about whether the
appropriate rate to use was the one at origination or the
reversion rate. Indeed, there has been significant variation
across lenders on the stressed mortgage rate used to assess
affordability compared to their current SVRs. Around half of
the mortgages extended in 2016 Q4 were tested using a
stressed interest rate of SVR plus 2.75–3.25 percentage points.
About 30% of mortgages were tested at a lower rate, and
about 20% at a higher rate.
Lending conditions in the mortgage market are becoming
easier and competitive pressures in the market remain. So
there is a risk that lenders loosen the standard at which they
test affordability, especially if there is significant scope for
interpretation of the policy.
The new Recommendation promotes consistency of
implementation across lenders and insures against the risk of
loosening underwriting standards. It also ensures that
borrower affordability is tested in the event that the borrower
is unable to refinance their mortgage at the end of the
fixed-rate period, which is appropriate given that — in times of
stress — some borrowers may be unable to do.
Although the new Recommendation will require some lenders
to increase the stressed interest rate at which they test
affordability, the aggregate impact on current mortgage
lending is expected to be small. The amendment reinforces
the insurance role of the FPC’s measures.
• In 2016 Q4, the average stressed rate (weighted by the
volume of new lending) was just over 6.8%. Bank staff
estimate that with the new Recommendation it would have
been just over 7%.
(1)
• Bank staff’s central estimate is that, had the new
Recommendation been in place in 2016, it would have
reduced mortgage approvals by less than 0.5% relative to
the previous Recommendation, with a slightly larger impact
on smaller lenders than on the major lenders.
• The aggregate impact on actual lending is estimated to be
small because, even if a lender increases its stressed interest
rate, borrowers whose mortgage payments (calculated at
the stressed rate) are low relative to their incomes will still
pass the test.
Consistent with the LTI flow limit, the new affordability test
Recommendation includes a ‘de minimis’ threshold (set at
£100 million of mortgage lending per annum) that exempts
some lenders, to ensure proportionality.
(2)
(1) The 2016 Q4 quarter has been chosen rather than the whole of 2016 to avoid
comparability issues due to the Bank Rate cut in August 2016.
(2) The FPC will monitor lending done by firms below this threshold.
1
4 Financial Stability Report
J
une 2017
UK consumer credit
Consumer credit has been growing rapidly in recent years…
Lending to individuals in the form of consumer credit grew by
10.3% in the twelve months to April 2017. Though a little
slower than in 2016 Q4, this was close to its fastest annual
growth rate since 2005.
Consumer credit has been growing much faster than
household incomes in recent years (Chart A.19). During that
period, dealership car finance has seen the fastest expansion,
though other forms of consumer credit (mainly credit cards
and personal loans) accounted for more than half of consumer
credit growth in the past year. Around half of net consumer
credit lending in the twelve months to April 2017 was from
banks, with major UK banks and smaller banks making broadly
equal aggregate contributions to the annual growth rate.
Lenders expect to continue to grow their portfolios through
2017, at the same time as real household income growth is
expected to remain particularly weak. For an overview of the
consumer credit market, see Box 6.
…reflecting a shift in the supply of credit…
Strong consumer credit growth in recent years partly reflects
structural changes in the provision of consumer finance,
including a shift towards the use of dealership car finance.
Technological changes, such as online shopping and
contactless cards, have also encouraged greater credit card use
for transactional purposes, with transactional credit card
balances growing at broadly the same rate as overall credit
card balances between 2012 and 2016.
UK consumer credit has been growing rapidly. Loss rates on consumer credit lending are low at
present. Partly as a result, banks’ net interest margins on new lending have fallen and major lenders
are using lower risk weights to calculate the capital they need to hold. Other things equal, these
developments mean lenders have less capacity to absorb losses. In this context, a review by the PRA
has found evidence of weaknesses in some aspects of underwriting and a reduction in resilience.
The FPC supports the intentions of the PRA and FCA to publish, in July, their expectations of lenders
in the consumer credit market. Firms remain the first line of defence. Effective governance at firms
should ensure that risks are priced and managed appropriately and benign conditions do not lead to
complacency by lenders. The Bank’s annual stress test assesses banks’ resilience to risks in consumer
credit. Given the rapid growth in consumer credit over the past twelve months, the FPC is bringing
forward the assessment of stressed losses on consumer credit lending in the Bank’s 2017 annual
stress test. This will inform the FPC’s assessment at its next meeting of any additional resilience
required in aggregate against this lending.
5
0
5
10
15
20
25
2013 14 15 16 17
Percentage points
Total consumer credit
(
b)
Credit card
(b)
Dealership car finance
(a)
Other (non-credit card and non-dealership car finance)
(b)(d)
Nominal household income growth
(c)
+
–
Chart A.19 Consumer credit has been growing much
faster than household incomes
Annual growth rates of consumer credit products and household
income
Sources: Bank of England, ONS and Bank calculations.
(a) Identified dealership car finance lending by UK monetary financial institutions (MFIs) and
other lenders.
(b) Sterling net lending by UK MFIs and other lenders to UK individuals (excluding student
loans). Non seasonally adjusted.
(c) Percentage change on a year earlier of quarterly nominal disposable household income.
Seasonally adjusted.
(d) Other is estimated as total consumer credit lending minus dealership car finance and credit
card lending.
P
art A
U
K consumer credit 15
Respondents to the Bank’s quarterly Credit Conditions Survey
reported a loosening of unsecured credit availability to
households in every quarter from end-2012 to end-2016. This
partly reflects intense competition among lenders, as reported
b
y market contacts, including those of the Bank’s Agents, and
by major lenders in the Credit Conditions Survey.
Interest rates on new personal loans have fallen since 2013
(Chart A.20), squeezing interest margins. Credit card lenders
have been offering longer interest-free periods. The average
interest-free period offered to new credit card customers on
balance transfer offers has doubled since 2011 (Chart A.21).
…and relatively benign economic conditions recently.
The fall in pricing may also reflect lenders incorporating a
relatively benign macroeconomic environment in their
assessment of risk. Loss rates on consumer credit lending are
low at present, with arrears rates on UK banks’ consumer
credit books in 2016 materially lower than their post-crisis
average. If lenders place undue weight when assessing risk on
the recent performance of loans, this could mask a build-up of
vulnerabilities in the consumer credit market.
These developments leave lenders more vulnerable to losses
in a stress…
All else equal, falling interest margins mean that less interest
income is available for lenders to absorb losses on consumer
credit lending. Average risk weights for consumer credit have
also fallen in recent years, reducing the loss-absorbing capital
required to fund these exposures (Chart A.22).
(1)
Falling margins do not appear to have been accompanied by a
corresponding improvement in the underlying credit quality of
new lending. For example, qualitative responses to the Bank’s
Credit Conditions Survey indicate that lenders have been
loosening their credit scoring criteria for non-credit card
unsecured lending since 2013. All else equal, this makes it
easier for households with lower credit scores to access
consumer credit.
Joint Bank/FCA analysis of data from credit reference agencies
further shows a slight deterioration in credit scores on new
consumer credit lending between 2015 and 2017. This
deterioration has occurred despite the fact that credit scores
— which are used as an input to underwriting decisions by a
number of lenders — can display elements of procyclicality. In
a benign macroeconomic environment, more borrowers can
access credit and are able to make regular payments, leading
to an improvement in credit scores, all else equal.
The effects of any loosening in underwriting standards could
be exacerbated by the rapid turnover of lenders’ consumer
credit portfolios. As a result of short payment terms, the stock
0
2
4
6
8
10
12
1
4
2004 06 08 10 12 14 16
Percentage points
Personal loans
cross-subsidised
by PPI income in
this period
(c)
C
hart A.20 Interest rates on new personal loans have
b
een falling relative to risk-free rates
Spread between effective interest rates on new personal loans and
Bank Rate
(a)(b)
Sources: Bank of England and Bank calculations.
(a) The Bank’s effective interest rate series are currently compiled using data from up to
19 UK MFIs. Data are non seasonally adjusted.
(b) Effective rates are sterling-only monthly averages.
(
c) Income from cross-selling of payment protection insurance (PPI) substantially offset low
margins on personal loans during this period. For more details see
www.bankofengland.co.uk/publications/Documents/quarterlybulletin/qb100301.pdf.
5
10
15
20
25
30
3
5
4
0
4
5
2011 12 13 14 15 16 17
Maximum 0% balance transfer offer
(b)
Average 0% balance transfer offer
(c)
N
umber of months
0
Chart A.21 Credit card lenders are offering longer
interest-free periods
Interest-free periods of credit card balance transfer offers
(
a)
Sources: Moneyfacts and Bank calculations.
(a) Whole market end-month data, excluding values of zero and nil returns.
(b) The maximum 0% balance transfer term available across all lenders.
(c) The average 0% balance transfer term is the average of the maximum 0% balance transfer
term available for each lender.
(1) Risk weights have not fallen for firms using the standardised approach to risk weights.
1
6 Financial Stability Report
J
une 2017
of most forms of consumer credit turns over much faster than
mortgages: at end-2016, around 75% of the major UK banks’
outstanding personal loans had been issued in the previous
two years, compared with about 30% of their mortgages
(
Chart A.23). The credit quality of the stock of lending can
therefore deteriorate quickly.
…and could pose a risk to lenders’ resilience.
Consumer credit accounts for less than 10% of major
UK banks’ stock of lending to UK real-economy borrowers,
compared with around 70% for mortgage lending. But losses
on consumer credit are far higher than for mortgages. That is
because, in the face of adverse shocks, borrowers are much
more likely to default on their consumer credit loans than
their mortgages. And as the majority of consumer credit
lending is unsecured, lenders cannot rely on the value of
collateral to cushion their losses. Over the past ten years,
UK banks’ total write-offs on UK consumer credit lending have
been ten times higher than on mortgages (Chart A.24).
Losses on consumer credit lending are highly correlated with
changes in unemployment (Chart A.25). In the 2016 stress
test, which included a 4.5 percentage point increase in the
UK unemployment rate, stressed impairments on
UK consumer credit exposures totalled around £18.5 billion.
This compared with £11.8 billion of impairments for
UK mortgages. Aggregate cumulative impairments on
consumer credit were 19% of total exposures in the five years
of the stress scenario, compared with 1% for mortgages.
In this context, vigilance is warranted.
The PRA has undertaken a targeted review into the credit
quality of consumer credit lending, and the FCA has been
undertaking a review of its rules and guidance on
creditworthiness assessments used in the consumer credit
market. The PRA review has found evidence of weaknesses in
some aspects of underwriting and a reduction in resilience.
The FPC supports the intentions of the PRA and FCA to
publish, in July, their expectations of lenders in the
consumer credit market. Firms remain the first line of
defence. Effective governance at firms should ensure that
risks are priced and managed appropriately and benign
conditions do not lead to complacency by lenders.
The Bank’s annual stress-test exercise assesses banks’
resilience to risks in consumer credit. Given the rapid
growth in consumer credit over the past twelve months, the
FPC is bringing forward the assessment of stressed losses
on consumer credit lending in the Bank’s 2017 annual stress
test. This will inform the FPC’s assessment at its next
meeting of any additional resilience required in aggregate
against this lending.
0
2
0
4
0
60
80
1
00
120
C
redit card Other consumer credit
2
014
2
015
2
016
P
er cent
C
hart A.22 Average consumer credit risk weights have
f
allen since 2014
Major UK banks’ average risk weights on consumer credit
exposures
(a)
Source: Bank of England.
(a) Risk-weighted assets include both IRB and standardised exposures and are measured as a
p
ercentage of drawn balances for end-year periods 2014 to 2016.
0
25
50
75
100
Personal loans Mortgages
Pre-2015
2015
2016
Per cent
Chart A.23 The stock of consumer credit turns over
quickly
End-2016 mortgage and personal loan portfolios of major
UK banks, by year of issuance
(a)(b)(c)
Source: Bank of England.
(a) Personal loan calculations have been made using vintaged data.
(b) Peer group for personal loan data is made up of the major UK banks, representing around
75% of the total UK market.
(c) Peer group for mortgage data represents around 74% of the total UK market, and includes
the major UK banks.
P
art A
U
K consumer credit 17
0
2
4
6
2
1012
Annual write-off rate (per cent)
(a)(b)
P
ercentage point change in the annual unemployment rate
(twelve-month lag)
+–
C
hart A.25 Consumer credit losses are highly correlated
w
ith changes in unemployment
Historical relationship between changes in unemployment and
write-offs on non-credit card consumer credit exposures
S
ources: Bank of England, ONS and Bank calculations.
(
a) Write-offs by UK MFIs on all currency other unsecured loans to UK individuals, expressed in
sterling. Write-offs are net of recoveries. Non seasonally adjusted.
(b) These series are calculated as annualised quarterly write-offs divided by the corresponding
loans outstanding at the end of the previous quarter. These data are presented as annual
series using four-quarter averages.
0
10
20
30
40
50
6
0
7
0
80
90
100
1993 96 99 2002 05 08 11 14 17
C
onsumer credit
(b)
Mortgages
(c)
P
ercentage share
C
hart A.24 Consumer credit losses are far higher than
f
or mortgages
UK banks’ sterling write-offs on lending to individuals
(a)
Sources: Bank of England and Bank calculations.
(a) Write-offs of sterling lending by UK MFIs to UK individuals. Write-offs are net of recoveries.
Non seasonally adjusted.
(b) Consumer credit consists of credit card lending and other unsecured lending (other loans and
advances) and excludes student loans.
(c) Lending secured on dwellings.
1
8 Financial Stability Report
J
une 2017
Box 6
Overview of the UK consumer credit market
In April 2017, the total stock of UK consumer credit was
£198 billion. By comparison, the total amount of outstanding
mortgage debt is around seven times larger at £1.3 trillion.
Consumer credit accounts for less than 10% of UK banks’ stock
o
f lending to UK real-economy borrowers, compared with
around 70% for mortgage lending. But it accounts for a much
higher proportion of losses: since 2007, UK banks’ total
write-offs on UK consumer credit have been ten times higher
than on mortgages.
The consumer credit market can be divided into three broad
categories:
• Credit cards (34% of the stock of consumer credit).
• Dealership car finance (30% of the stock).
• ‘Other’ (36% of the stock), which is mainly made up of
personal loans. This category also includes overdrafts,
peer-to-peer lending, store credit, and lending from credit
unions and small non-bank money lenders such as
pawnbrokers and payday lenders.
(1)
The consumer credit market is highly diverse, with a wide
range of products and lenders. But there are some common
features across product types:
• Short terms: most consumer credit lending is at maturities of
five years or less (apart from revolving facilities such as
credit cards or overdrafts).
• Higher interest rates: while interest rates vary considerably
across product types, they are typically materially higher
than for mortgages. This reflects a lack of collateral for most
consumer credit products, and is intended to compensate
the lender for the higher risk of losses.
• Fixed rates: most consumer credit products have interest
rates that are fixed for the entire term of the loan.
Banks provide around 80% of lending in the credit card and
‘other’ categories, but less than half of dealership car finance
(Chart A).
Credit cards
There are two main ways of using credit cards: ‘transactional’
and ‘revolving’. Transactional credit card users repay their full
balances at the end of the month, and so incur no interest
payments. Users with revolving balances roll over some of
their debt from month to month, typically paying a relatively
high interest rate (around 20%) on this borrowing.
Some revolving credit card balances incur no interest. Under a
typical ‘balance transfer’ offer, borrowers move existing debt
onto a new card, in some cases paying an upfront fee. For a set
period they then pay 0% interest on this balance, while paying
down a proportion of the outstanding balance (typically a
minimum of 1% per month, plus nominal fees). At the end of
the 0% offer period, interest is charged on any outstanding
debt at a standard rate. Other interest-free offers allow
customers to accrue new debt at 0% interest for a period on
new purchases or money transfers.
The major UK banks have around £42 billion of outstanding
credit card lending — around 19% of their common equity
Tier 1 (CET1) capital.
Other consumer credit
The bulk of this category is made up of unsecured personal
loans, which typically have a three to five-year maturity at a
fixed interest rate. Just over a quarter of personal loans are
taken out for the purposes of debt consolidation, with most of
the remainder funding large purchases such as home
improvements.
The major UK banks have around £41 billion of ‘other’
consumer credit lending — around 19% of their CET1.
Dealership car finance
‘Dealership car finance’ refers to loans offered to car buyers at
the point of sale. Traditionally, this takes the form of a hire
purchase agreement. Under a standard hire purchase, the
customer pays a fraction of the car’s purchase price as a
deposit, and takes out a loan to cover the rest, which is paid off
with interest in regular monthly instalments, amortising fully.
(1) The Bank’s headline measure of consumer credit excludes income-contingent student
loans.
Banks £24 billion
Non-banks
£34 billion
Banks £52 billion
Banks £62 billion
Personal loans,
o
verdrafts and
other lending
£72 billion
Dealership car
finance
£58 billion
C
redit cards
£
67 billion
N
on-banks
£
14 billion
Non-
banks
£
10
billion
C
hart A
C
omposition of the stock of consumer credit,
e
nd-March 2017
(a)(b)(c)
Sources: Bank of England, Finance & Leasing Association, published accounts and
Bank calculations.
(a) Banks include monetary financial institutions (MFIs) and, where identified, non-bank
subsidiaries of UK MFIs.
(b) Excludes income-contingent student loans.
(
c) Numbers may not sum to totals because of rounding.
P
art A
U
K consumer credit 19
‘Personal contract purchase’ (PCP) agreements are a type of
hire purchase agreement with lower monthly payments. At
the end of the loan period customers have the option to make
a pre-agreed ‘balloon payment’ or return the vehicle to the
d
ealer. The size of this balloon payment is set when the loan
is issued, based on the estimated value of the vehicle in the
used car market at the end of the loan period.
Dealership car finance has been growing rapidly in recent
years, with an average annual growth rate of around 20%
since 2012. The total stock of dealership car finance increased
by more than £30 billion over that period, representing three
quarters of total growth in the stock of consumer credit.
This growth partly reflects a recovery in the car market: total
new car registrations in 2016 were 30% higher than in 2012.
It also reflects a structural shift in how cars are purchased.
Around 85% of new car purchases used dealership car finance
in 2016, compared with about half in 2009. In particular, use
of PCP agreements has grown rapidly (Chart B).
A significant share of dealership car finance is provided by
subsidiaries of global car manufacturers. Around half of the
debt funding for these subsidiaries comes from their parent
companies, around a quarter from securitisation, with the
remainder from bank lending.
Bank staff estimate major UK banks’ total exposures to UK car
finance to be around £20 billion, or 9% of CET1, comprising:
• Direct exposures — around £17 billion, or 8% of CET1. This
l
argely represents lending by banks’ asset finance
subsidiaries, which make loans to car buyers, arranged
through dealerships.
• Indirect exposures via lending to UK finance subsidiaries
of car manufacturers — around £2 billion, less than 1% of
CET1.
(
1)
• Indirect exposures via holdings of asset-backed securities
in banks’ liquid asset buffers — around £1 billion, less than
0.5% of CET1.
Arrears rates on dealership car finance tend to be lower than
for other forms of consumer credit. Unlike most other
consumer credit, this lending is secured, with the vehicle
acting as collateral. But the value of this collateral declines
over time, and is dependent on conditions in the used car
market.
Exposures to PCP lending may be particularly sensitive to
market conditions: if the borrower chooses to return the car
at the end of the loan, and the value of the used car is less
than the outstanding loan amount, the lender will make a loss.
Lenders can seek to mitigate these risks by making
conservative assumptions about the future value of used cars,
though these assumptions are inherently uncertain.
In a hypothetical scenario where car values at the end of
contracts turn out to be 20% lower than expected by lenders
at origination — and all PCP borrowers returned their vehicles
— Bank staff estimate that market-wide PCP losses could be
3%–6% of the total outstanding stock of car finance. Other
things equal, if these loss rates were experienced on the major
UK banks’ dealership car finance portfolios, it would imply a
reduction of 2–7 basis points in major UK banks’ aggregate
CET1 ratio.
(2)
That is, from a starting point of 13.92%, the
ratio would fall to 13.85%–13.90%. Market-wide losses would
rise to 7%–10% of the outstanding stock in a more severe
scenario where car values at the end of contracts turn out to
be 30% lower than originally expected. If these loss rates
were applied to major UK banks’ portfolios, they would imply
a reduction of 7–11 basis points in their aggregate CET1 ratio.
0
10
20
30
40
50
60
70
80
90
100
0
2
4
6
8
10
12
14
16
18
20
2008 09 10 11 12 13 14 15 16
Personal contract purchase
(a)
(left-hand scale)
O
ther dealership car finance lending (left-hand scale)
Proportion of new car sales funded with dealership car finance
(
b)
(right-hand scale)
£ billions
Per cent
Chart B Value of annual dealership car finance for new
car purchases, and proportion of private new car
purchases funded with dealership car finance
Sources: Finance & Leasing Association, Society of Motor Manufacturers and Traders (SMMT)
and Bank calculations.
(a) Annual sterling gross lending to individuals on dealership car finance for new car purchases
provided by Finance & Leasing Association members, attributed to personal contract
purchase (PCP).
(b) Annual transactions on dealership car finance for new car purchases provided by
Finance & Leasing Association members, as a proportion of SMMT new car registrations.
(1) Major UK banks also have over £10 billion of lending to global car manufacturers, but
these exposures are less directly linked to the UK car market.
(2) Given direct exposures of £17 billion, a 5% loss rate implies a loss of £0.85 billion.
2
0 Financial Stability Report
J
une 2017
Global environment
Near-term prospects for the global economy have improved
slightly…
Since the November 2016 Report, the near-term global
economic outlook has improved slightly. In April, for the first
time since 2011, the International Monetary Fund (IMF)
revised up its forecast for world GDP growth in 2017, to 3.5%
from 3.4% (Chart A.26). In May, the Monetary Policy
Committee’s projection for average annual global GDP
growth, weighted by countries’ share in UK exports, was 2½%
in 2017–19. This is up from 2¼% in November, partly due to
stronger growth prospects in the euro area.
The US Federal Open Market Committee raised its policy rate
by 25 basis points in December, March and again in June.
These increases were in line with market expectations and
yields on US ten-year Treasury bonds have fallen by close to
20 basis points since the November Report,
(1)
offsetting
around a third of the rise in US bond yields that followed
immediately after the US election. These developments have
helped to ease external financing pressures on emerging
market economies (EMEs), though vulnerabilities from high
levels of debt remain.
…though geopolitical and policy uncertainties remain high.
While uncertainty about the outlook for US fiscal policy
remains, some other aspects of the new US administration’s
economic policy have become clearer. In particular, the
agreement between the United States and China to liberalise
bilateral trade in certain food products and financial services,
and signs that the United States will seek to renegotiate,
rather than withdraw from, the North American Free Trade
Agreement have eased some immediate concerns about the
future of the global trading system. Nevertheless, political
and policy uncertainties, in particular about the prospects for a
longer-term shift towards protectionism or for a weakening in
global co-operation in financial regulation, remain high.
As home to the leading international financial centre, the
resilience of the UK financial system depends in part on
Near-term growth prospects for the global economy have improved slightly and some possible risks
have not crystallised. But financial vulnerabilities in China remain pronounced as, although capital
outflows have stabilised, economic growth continues to be accompanied by rapid credit expansion.
In the euro area, bank equity prices have risen, reflecting expectations of stronger economic growth
and further progress in strengthening banks’ capital positions.
(1) The data cut-off for the November Report was 18 November 2016.
0
2
4
6
8
10
1
2
1999–2007
17
18
17
18
17
18
17
18
17
18
1999–2007
1999–2007
1999–2007
1999–2007
World
Emerging
market and
developing
economies
China Euro area United States
October 2016
A
pril 2017
P
re-crisis average
Percentage changes on a year earlier
Chart A.26 Global growth projections have been revised
up slightly, though they are still below pre-crisis levels
International annual GDP growth projections
Sources: IMF World Economic Outlook (WEO) and Bank calculations.
P
art A
G
lobal environment 21
standards applied in other jurisdictions. Absent consistent
implementation of standards internationally and appropriate
supervisory co-operation, the FPC will need to assess how best
to protect the resilience of the UK financial system.
I
n China, foreign reserves have stabilised and capital
outflows have fallen…
Pressure from persistent capital outflows from China has
eased. The Institute of International Finance (IIF) estimates
that the net outflow of capital from China was around
US$42 billion in the first four months of 2017, down from
around US$250 billion in the previous four months. China’s
official foreign exchange reserves were broadly stable in the
first five months of 2017. This stabilisation appears to reflect a
combination of tighter controls on capital movements by the
Chinese authorities, higher policy rates in China and a
weakening in expectations of an appreciation in the US dollar.
…but credit has continued to grow rapidly.
However, underlying vulnerabilities remain pronounced. Total
social financing, a broad measure of domestic private sector
credit, rose by 15.4% in the year to May 2017;
(1)
total
non-financial sector debt was estimated to be just under
260% of GDP (Chart A.27). In addition to raising policy rates,
the Chinese authorities have tightened bank regulation in
order to reduce leverage in the financial sector. This will push
up market rates further, making it more difficult for borrowers
to service their debts. The authorities have also set a target
for economic growth of ‘around 6.5% or higher’ in 2017, and
continued credit growth is expected to be an important tool
for achieving this, making their task of reducing medium-term
risks to financial stability more challenging. If these risks were
to crystallise, economic growth in China would slow sharply,
leading to a sharp fall in Chinese imports and weaker growth
in its trading partners. This could have a larger effect globally
than previously, given that China is becoming a more
important trading partner for many economies (Chart A.28).
UK banks’ exposures to China and the closely linked economy
of Hong Kong are large (around US$535 billion, 183% of
common equity Tier 1 capital, Chart A.29).
Other EMEs have seen a resumption of portfolio inflows.
After a brief period of outflows in late 2016, there have been
substantial inflows of non-resident portfolio capital to other
EMEs. Credit-GDP gaps and current account deficits in many
EMEs have also continued to narrow. However, external debt,
often taken out by the private sector and denominated in
US dollars, remains high in many individual EMEs. This may
leave some countries like Turkey with weaker credit ratings
and high borrowing needs (the IMF projects that maturing
external debt in Turkey will be nearly 20% of GDP each year in
2017–19) exposed to a shift in risk appetite away from EMEs, a
0
1
0
20
3
0
4
0
5
0
0
50
100
150
2
00
2
50
3
00
2006 08 10 12 14 16
Non-financial sector
(a)
(right-hand scale)
Adjusted total social financing
(b)
(left-hand scale)
Headline total social financing
(c)
(left-hand scale)
Percentage changes on a year earlier Per cent of GDP
Chart A.27 Credit continues to grow rapidly in China
China non-financial sector debt and growth of total social
financing
Sources: BIS total credit statistics, CEIC and Bank calculations.
(a) Non-financial sector debt data are to 2016 Q4. Includes lending by all sectors at market
value as a percentage of GDP, adjusted for breaks.
(b) Total social financing adjusted for net issuance of local government bonds.
(c) The People’s Bank of China stock of total social financing used from December 2014
onwards. Prior to this the stock of total social financing is estimated using monthly ‘newly
increased’ total social financing flows.
0510 15 20
Taiwan
Hong Kong
Malaysia
Korea
Singapore
Australia
South Africa
Indonesia
Brazil
India
Mexico
Turkey
Japan
Germany
United States
United Kingdom
2005
2011
Per cent of GDP
Chart A.28 Exports to China account for an increasing
share of value added in many countries
Domestic value added in exports to China
Sources: OECD Trade in Value Added database and Bank calculations.
(1) After adjusting for the statistical effect of replacing local government borrowing
through financing vehicles with the issuance of municipal bonds.
2
2 Financial Stability Report
J
une 2017
faster-than-expected rise in US interest rates or a renewed
strengthening in the US dollar.
UK banks’ exposures to non-China EMEs have fallen in recent
years and broadly matched those to China and Hong Kong in
2
017 Q1 (Chart A.29).
Pressures on the European banking system have eased but
concerns about profitability remain.
Bank equity prices in continental Europe have continued to
recover since the November Report. The improved economic
outlook is expected to increase the demand for credit and also
to lead to lower provisions for non-performing loans; both
will tend to raise bank earnings. In addition, successful
capital-raising exercises by three major European banks,
progress towards recapitalising some smaller Italian banks and
the orderly resolution of a domestically significant Spanish
bank, have eased some immediate concerns about the
resilience of the banking sector.
Euro-area banks’ common equity Tier 1 capital ratios are
significantly higher than before the crisis. However,
non-performing loans remain at high levels in several
European countries (Chart A.30) and concerns about the
longer-term viability of some European banks’ business
models remain. In particular, the profitability of some
European banks appears to be lower than the cost of raising
new capital, which could make it difficult for them to rebuild
their capital positions if they were to be hit by adverse shocks.
Sovereign bond spreads in the euro area have fallen but
government debt is still high in several economies.
Sovereign bond spreads in several euro-area Member States
rose in early 2017, reflecting heightened uncertainty ahead of
elections in a range of countries. They have since fallen back
as the parliamentary election in the Netherlands (March 2017)
and the presidential election in France (May 2017) passed
without major shocks to markets. However, government debt
is still high in several euro-area economies. Italy, where
government debt was 133% of GDP in 2016, remains a
particular concern. Risks from the euro area may also emerge
as a consequence of the United Kingdom’s withdrawal from
the European Union (see Box 2).
The FPC continues to assess UK banks’ ongoing resilience to
the risks stemming from China and the euro area in its
annual stress test. The 2017 stress test includes an
extremely severe scenario for China in which Chinese
growth slows from just under 7% at the end of 2016
to -1.2% by the end of 2017, reflecting the scale of risks
created by rapid credit expansion.
0
100
200
300
400
5
00
6
00
7
00
800
900
2
005 07 09 11 13 15 17
Non-China emerging markets
C
hina and Hong Kong
US$ billions
C
hart A.29 UK banks’ claims on non-China EMEs have
f
allen back to broadly match those on China and
Hong Kong
UK-owned banking groups’ consolidated exposures to selected
countries and regions
S
ources: Bank of England and Bank calculations.
0510 15 20
Core excluding
Germany
(c)
Periphery
excluding Italy
(b)
Scandinavia
(a)
Germany
Italy
United Kingdom
Per cent
Chart A.30 Non-performing loan ratios are still high for
banks in some European countries
Ratio of non-performing loans and advances to total loans and
advances (2016 Q4)
Sources: European Banking Authority and Bank calculations.
(a) ‘Scandinavia’ refers to banks in Denmark, Norway and Sweden.
(b) ‘Periphery excluding Italy’ refers to banks in Ireland, Portugal and Spain.
(c) ‘Core excluding Germany’ refers to banks in Austria, Belgium, France and the Netherlands.
Asset valuations
P
art A
A
sset valuations 23
Long-term risk-free interest rates are at levels consistent
with pessimistic growth expectations and high perceived
tail risks…
In the United Kingdom, ten-year real government bond yields
are at around -2% (Chart A.31), and long-term real interest
rates remain low across the G7. While reflecting, in part,
monetary policy actions over most of the past decade, these
low interest rates — which are adjusted for compensation for
inflation — also appear to be consistent with pessimistic
growth expectations (despite recent improvements to the
near-term outlook), and high perceived tail risks. Those
perceptions may in part be associated with a high degree of
geopolitical uncertainty (see Global environment chapter). If
severe, but plausible, downside risks were to materialise, they
could lead to a sharp reduction in growth, increasing the
attractiveness of longer-term government bonds relative to
more risky assets, such as equities and corporate bonds.
(1)
…but measures of uncertainty implied by options prices are
low…
Market-based measures of perceived risks in the near term —
or implied volatilities — have been low by historical standards
across a number of markets (Chart A.32). These measures are
derived from the options used to insure against asset price
moves. The current low level of these measures may suggest
that there is a relatively low expectation that geopolitical risks
will actually materialise, at least over short horizons. In June,
the VIX measure of implied equity volatility, derived from
S&P 500 stock index option prices, reached its lowest level
since 1993.
Long-term risk-free interest rates are at levels consistent with pessimistic growth expectations and
high perceived tail risks. In contrast, measures of uncertainty implied by options prices are low.
Some asset valuations, particularly for some corporate bonds and UK commercial real estate,
appear to factor in a low level of long-term market interest rates but do not appear consistent with
the pessimistic and uncertain outlook embodied in these rates.
These asset prices are therefore vulnerable to a repricing, whether through an increase in long-term
interest rates or adjustment of growth expectations, or both. A sudden fall in asset prices could be
amplified given reduced liquidity in some markets, particularly if some types of investors increase
sales in response to price falls. The resilience of the UK banking system to these repricing risks is
being assessed through the 2017 stress test.
(1) See Broadbent, B (2014), ‘Monetary policy, asset prices and distribution’;
www.bankofengland.co.uk/publications/Pages/speeches/2014/770.aspx.
3
2
1
0
1
2
3
4
2006 07 08 09 10 11 12 13 14 15 16 17
United Kingdom
United States
Euro area
Per cent
+
–
Chart A.31 Advanced-economy risk-free real interest
rates remain close to historically low levels
International ten-year real government bond yields
(
a)
Sources: Bloomberg and Bank calculations.
(a) Zero-coupon bond yields derived using inflation swap rates. UK real rates are defined
relative to RPI inflation, whereas US and euro-area real rates are defined relative to CPI and
HICP inflation respectively.
2
4 Financial Stability Report
J
une 2017
These very low levels of market volatility could be supported
by improvements in the near-term economic outlook, and
expectations that the pace of monetary policy normalisation,
which is already under way in the United States, will be
g
radual. Increased use by investors of strategies that sell
insurance against a rise in volatility, for which they get paid a
premium, may have further contributed to the most recent
reduction in measures of volatility.
The low volatility environment could incentivise behaviour
that leads to risk-taking and higher leverage, which would
build up risks and fragilities in the financial system.
…and some asset valuations appear inconsistent with the
pessimistic and uncertain outlook embodied in long-term
market interest rates.
In corporate bond markets, returns appear low relative to the
risk borne by investors, offering little compensation for
downside risks. For example, corporate bond spreads are
compressed, most notably in the riskier high-yield sector
(Chart A.33). In the United States, this compression in
spreads has been accompanied by elevated levels of corporate
leverage. Non-price terms for corporate borrowing have also
eased: in both Europe and the United States, the share of
leveraged lending deals with weaker covenants — where
investors accept fewer safeguards in the event of a
deterioration in the debtor company’s finances — has
increased to over 70% in the first five months of 2017, from
less than 5% in 2010.
In the United Kingdom, real interest rates have fallen further,
which does not appear to be reflected in some risky asset
prices…
In the United Kingdom, long-term risk-free real interest rates
have fallen by 26 basis points since the previous Report,
compared with a smaller fall in euro-area rates, and broadly
unchanged US real rates (Chart A.34). These developments
suggest investors may be factoring in a higher probability of an
adverse outcome for the UK economy.
That probability of an adverse outcome does not, however,
appear to be reflected in some risky asset prices. Spreads on
sterling investment-grade corporate bonds, for example, are
compressed relative to historical averages and high-yield
spreads are close to post-crisis lows.
…including in the UK commercial real estate market…
Valuations in some segments of the UK commercial real estate
(CRE) sector also continue to appear stretched. Aggregate
prices rose by 0.9% on the quarter in 2017 Q1, leaving them
broadly flat on a year earlier, despite a sharp fall in the quarter
following the UK referendum on EU membership. CRE rental
yields remain low by historical standards, suggesting that
future rental income expectations embody growth prospects
0
500
1,000
1,500
2,000
2,500
3
,000
2000 02 04 06 08 10 12 14 16
US dollar
Euro
Sterling
Basis points
D
ashed lines: averages since 2000
Chart A.33 Spreads on high-yield corporate bonds are
compressed relative to past averages
High-yield corporate bond spreads
(a)
Sources: Bank of America Merrill Lynch Global Research and Bank calculations.
(a) Option-adjusted spreads. The US dollar series refers to US dollar-denominated bonds issued
in the US domestic market, while the sterling and euro series refer to bonds issued in
domestic or eurobond markets in the respective currencies.
3
2
1
0
1
2
3
4
5
6
7
8
U
S$/€ FTSE 100 S&P 500 UK rates US rates
Interquartile range since 2000
D
ata for 16 June 2017
S
tandard deviations from averages since 2000
H
istorical range since 2000
U
S$/£
+
–
C
hart A.32 Implied volatilities are low across a range of
f
inancial markets
Dispersion in implied volatilities in foreign exchange, interest rate
and equity markets
(a)
Sources: Barclays Live, BBA, Bloomberg, Chicago Mercantile Exchange, NYSE ICE and
Bank calculations.
(a) Three-month implied volatilities for exchange rates, equities and ten-year interest rates.
P
art A
A
sset valuations 25
that are inconsistent with those embodied in the low risk-free
rates by which they are being discounted.
…which is therefore vulnerable to repricing.
A
range of sustainable valuations in the CRE market can be
generated using a valuation model based on a number of
assumptions, including about CRE rental yields.
Based on this approach, current prices lie at the top end of the
range of sustainable valuations (the blue range in Chart A.35),
which is consistent with persistently low rental yields.
Were rental yields to return to their historical averages, this
would suggest that current prices are above estimated
sustainable valuation levels (the lower bound of the blue range
in Chart A.35). This would be consistent with either an
increase in long-term risk-free interest rates or an adjustment
of risk premia and medium-term rental growth expectations,
or both.
Some segments of the CRE market appear more stretched
than the aggregate picture. For example, current London
West End office prices are well above the range of estimated
sustainable valuation levels (the orange range in Chart A.35).
Consensus forecasts from the Investment Property Forum,
published in March, point to average price falls of around
0.7% for aggregate UK CRE and 6.0% for London West End
offices by end-2018.
An adjustment in asset prices could be amplified by the
behaviour of some investors, which would affect the supply
of credit to the real economy.
Any adjustment in asset prices could be amplified, given
reduced liquidity in some markets, particularly if some
investors behave procyclically — that is, if they sell risky assets
in large quantities purely in response to a reduction in the
performance of their portfolios. A range of assets, including
UK CRE, are held in open-ended investment funds, some of
which allow investors to redeem their investments on a daily
basis. The structure of these funds could create incentives for
investors to redeem ahead of others, for example if investors
remaining in a fund were to bear some of the costs of meeting
redemptions.
(1)
This could test market liquidity (see
Market-based finance chapter).
In corporate bond markets, the amount held by open-ended
investment funds has increased substantially in recent years,
outpacing growth in the global market. Firms have also been
issuing longer-term bonds. While this locks in financing for a
longer period of time, it can mean that bondholders are
exposed to larger movements in prices in the event of a sharp
rise in interest rates. In addition, at low interest rates, the
(1) For further discussion see pages 23–25 of the December 2015 Financial Stability
Report; www.bankofengland.co.uk/publications/Documents/fsr/2015/dec.pdf.
40
50
60
70
80
90
100
110
120
130
140
150
03 05 07 09 11 13 15 17
Indices: 2007 Q2 = 100
2001
London West End
office prices
Aggregate
CRE prices
Ranges of sustainable valuations
(a)
U
pper part of ranges: low rental yields persist.
Lower part of ranges: rental yields rise,
consistent with a fall in rental growth
expectations or a rise in risk premia.
Chart A.35 UK commercial real estate prices look
stretched based on ranges of sustainable valuations
Commercial real estate prices in the United Kingdom and ranges
of sustainable valuations
Sources: Bloomberg, Investment Property Forum, MSCI Inc. and Bank calculations.
(a) Sustainable valuations are estimated using an investment valuation approach and are based
on an assumption that property is held for five years. The sustainable value of a property is
the sum of discounted rental and sale proceeds. The rental proceeds are discounted using a
5-year gilt yield plus a risk premium, and the sale proceeds are discounted using a 20-year,
5-year forward gilt yield plus a risk premium. Expected rental value at the time of sale is
based on Investment Property Forum Consensus forecasts. The range of sustainable
valuations represents varying assumptions about the rental yield at the time of sale: either
rental yields remain at their current levels (at the upper end), or rental yields revert to their
15-year historical average (at the lower end). For more details, see Crosby, N and Hughes, C
(2011), ‘The basis of valuations for secured commercial property lending in the UK’, Journal of
European Real Estate Research, Vol. 4, No. 3, pages 225–42.
0.5
0
.4
0.3
0
.2
0.1
0.0
0
.1
United Kingdom United States Euro area
P
ercentage points
Real
I
nflation
Nominal
+
–
C
hart A.34 The causes of changes in nominal
g
overnment bond yields differ across economies
Changes in nominal ten-year interest rates since the
November Report
(a)(b)
Sources: Bloomberg and Bank calculations.
(a) Zero-coupon rates derived from government bonds. The contribution of real rates and
implied inflation to the change in nominal rates is calculated using inflation swaps, which
r
eference RPI for the United Kingdom, CPI for the United States and HICP for the euro area.
(b) Shows changes in interest rates between 18 November 2016 and 16 June 2017.
2
6 Financial Stability Report
J
une 2017
0
100
200
3
00
400
500
600
7
00
2000 02 04 06 08 10 12 14 16
US$ billions
C
hart A.36 Interest rate risk related to corporate bond
m
arkets has increased
Estimated losses in global corporate bond markets following a
100 basis point increase in interest rates
(a)
Sources: Barclays Live, Thomson Reuters Datastream and Bank calculations.
(a) The chart is a measure of how aggregate risk exposures to corporate bond markets globally
have risen over time. It is calculated based on the Barclays Global Aggregate Corporate
Bond Index, which can be used as a representative measure of global investment-grade
corporate bond markets. However, the index does not capture all investment-grade
corporate bonds. The measure has been inflation-adjusted, with prices indexed to
J
anuary 2017.
responsiveness of corporate bond prices to shocks will tend to
be higher. Together, these factors mean that the losses that
would be incurred given a 100 basis point rise in interest rates
in global corporate bond markets, all else equal, have
i
ncreased markedly over the past few years (Chart A.36).
Such higher losses could increase the likelihood of investors
exhibiting procyclical behaviour and therefore magnifying
price falls.
In the limit, the supply of credit to the real economy, and
transfer of risk to those who are best placed to manage it,
could be impaired. CRE is widely used as collateral for
corporate borrowing. An amplified downturn in the CRE
market could be transmitted to the real economy by reducing
companies’ access to bank loans and their ability to undertake
new investment. Research by Bank staff suggests that every
10% fall in UK CRE prices is associated with a 1% decline in
economy-wide investment.
(
1)
A sharp fall in asset prices could further adversely impact the
balance sheets of banks and other financial institutions at the
core of the financial system.
UK banks have more than halved their stock of CRE lending
since the crisis (Chart A.37). The total stock of UK banks’ CRE
lending fell from around £160 billion at end-2008 to around
£77 billion at end-2016. For large UK banks involved in the
2017 stress test,
(2)
their exposures to the CRE sector averaged
around 50% of common equity Tier 1 capital at end-2016.
And the 2016 stress test demonstrated that they have become
more resilient to stresses in the CRE market.
The Bank’s 2017 annual cyclical stress-test scenario will
assess the resilience of the banking system to an increase in
volatility, a reduction in market participants’ appetite for
risk, and falls in asset prices, including a 40% fall in UK CRE
prices.
The FPC continues to emphasise the importance of market
participants recognising the distribution of risks in different
asset classes, managing them prudently, and pricing them
accordingly.
(1) Bahaj, S, Foulis, A and Pinter, G (2016), ‘The residential collateral channel’, Centre for
Macroeconomics Discussion Paper CFM-DP2016-07.
(2) The figure includes gross on balance sheet exposures as well as committed credit
lines, and exposures booked in Jersey and Guernsey. Standard Chartered Bank is
excluded, as it has minimal UK CRE exposures.
UK banks and building societies
I
nternational banks
Insurance companies
Other non-bank lenders
0
50
100
150
200
250
300
2004 05 06 07 08 09 10 11 12 13 14 15 16
£
billions
Chart A.37 UK banks’ stock of CRE lending has more
than halved since the crisis
UK CRE debt reported to De Montfort University survey
(a)
Sources: De Montfort University and Bank calculations.
(a) The composition of the survey sample was altered as follows: a category for insurance
companies was created in 2007, and another one for non-bank lenders in 2012. The
category of insurance companies includes only UK insurers from 2007 to 2011, and all
insurers from 2012 onwards. Data exclude commercial mortgage-backed securities.
Banking sector resilience
P
art B
B
anking sector resilience 27
The UK banking system remains resilient. The aggregate Tier 1 capital position of major UK banks
was 15.7% of risk-weighted assets in March 2017. The FPC intends to set the minimum leverage
requirement for major UK banks at 3.25% of non-reserve exposures, subject to consultation. This is
intended to offset the impact of removing central bank reserves from the leverage ratio exposure
measure, and restore the original level of resilience delivered by the leverage ratio standard. The
United Kingdom is on course to ensure that, by 2022, any remaining barriers to the resolvability of
the largest UK banks have been removed, and to implement ring-fencing requirements by 2019.
Work is also ongoing to build resilience to cyber risks.
UK banks’ funding costs remain low, and equity prices have recovered over the past year. However,
price to book ratios remain low, reflecting continued headwinds from misconduct costs and low
investment banking returns. Weak profitability diminishes banks’ future ability to rebuild capital
following a shock that incurs losses. The exploratory scenario in the 2017 stress test will consider
how the resilience of the UK banking system might evolve if recent headwinds to bank profitability
persist and intensify.
The UK banking system remains resilient.
UK banks have continued to strengthen their capital positions.
In aggregate, the major UK banks had a common equity
Tier 1 (CET1) ratio of 13.9% of risk-weighted assets in
March 2017 (Chart B.1), and a total Tier 1 capital ratio of
15.7%. This CET1 ratio is 40 basis points higher than at the
time of the November 2016 Report, when the FPC judged that,
as a consequence of the 2016 stress test, the UK banking
system was, in aggregate, capitalised to support the real
economy in a severe macroeconomic stress. On a non
risk-weighted basis, the major UK banks’ aggregate leverage
ratio was 5.3% of total exposures in March 2017, compared
with total requirements and buffers of 3.3% (Chart B.2).
UK banks’ liquidity and funding positions are also robust. For
major UK banks, the aggregate ratio of liquid assets to
potential net outflows under stressed conditions (known as
the Liquidity Coverage Ratio) was 128% in March 2017. And
all major UK banks have sufficient stable funding to meet the
proposed Net Stable Funding Ratio (NSFR) requirement.
(1)
Since 2014, the FPC has been contributing to the Bank’s
annual review of the Sterling Monetary Framework (SMF),
with members giving views on whether the SMF’s liquidity
insurance facilities remained fit for purpose from a
(1) The implementation date for the NSFR in the European Union is still to be confirmed.
Chart B.1 Major UK banks have continued to strengthen
their capital positions
Major UK banks’ capital ratios
Sources: PRA regulatory returns, published accounts and Bank calculations.
(a) Major UK banks’ core Tier 1 capital as a percentage of their risk-weighted assets. Major UK
banks are Banco Santander, Bank of Ireland, Barclays, Co-operative Banking Group, HSBC,
Lloyds Banking Group, National Australia Bank, Nationwide, RBS and Virgin Money. Data
exclude Northern Rock/Virgin Money from 2008.
(b) Between 2008 and 2011, the chart shows core Tier 1 ratios as published by banks, excluding
hybrid capital instruments and making deductions from capital based on FSA definitions.
Prior to 2008 that measure was not typically disclosed; the chart shows Bank calculations
approximating it as previously published in the Report.
(c) Weighted by risk-weighted assets.
(d) From 2012, the ‘Basel III common equity Tier 1 capital ratio’ is calculated as common equity
Tier 1 capital over risk-weighted assets, according to the CRD IV definition as implemented in
the United Kingdom. The Basel III peer group includes Barclays, Co-operative Banking Group,
HSBC, Lloyds Banking Group, Nationwide, RBS and Santander UK.
(e) CET1 ratio less the aggregate percentage point fall projected under the Bank of England’s
2016 annual cyclical stress scenario for the six largest UK banks.
2
8 Financial Stability Report
J
une 2017
macroprudential perspective. As part of the Bank’s 2017
review, at its March meeting the FPC reviewed developments
in the SMF over the previous year. These included the
additional indexed long-term repo operations that the Bank
h
ad announced prior to the EU referendum, and the increased
number of SMF participants over the course of the year. In the
FPC’s view, and given developments in the SMF since the
Bank’s 2016 review, the SMF remained fit for purpose from a
macroprudential perspective. The Bank’s 2017 annual review
of the SMF will be published alongside the Bank’s forthcoming
Annual Report.
(
1)
Consistent with its previous commitment, the FPC is
restoring the level of resilience delivered by its leverage ratio
standard…
In July 2016, the FPC excluded central bank reserves from the
measure of banks’ exposures used to assess their leverage.
This change reflected the special nature of central bank
reserves and was designed to avoid a situation in which the
Committee’s leverage standards impeded the transmission of
monetary policy. The FPC committed last year that it would
make an offsetting adjustment to ensure that the amount of
capital needed to meet the UK leverage ratio standard would
not decline. The FPC did not intend for there to be a
permanent loosening of the standard. At its June 2017
meeting, the FPC therefore agreed to consult on a draft
Recommendation to the PRA that it amend its rules on the
leverage ratio to:
(i) exclude from the calculation of the total exposure measure
those assets constituting claims on central banks, where
they are matched by deposits accepted by the firm that are
denominated in the same currency and of identical or
longer maturity; and
(ii) require a minimum leverage ratio of 3.25%.
By raising the minimum leverage standard from 3% to 3.25%
on the revised exposure measure, the FPC intends to ensure
that the original level of resilience is restored, while also
preserving the benefits of excluding central bank reserves from
the exposure measure. The FPC and PRA consultations on
these proposals will run from 27 June to 12 September 2017.
The Committee will further consider the impact of changes to
international standards on the calibration of the UK capital
and leverage framework. These changes include:
• the finalisation of Basel III; and
• the implementation of IFRS 9, due to come into effect in
2018.
(1) A ‘Concordat’ between the FPC and the Bank’s Executive, describing the role of the
FPC in relation to the SMF, was first published in 2013. At its 2017 Q2 meeting, the
FPC agreed a small number of updates to the Concordat, designed to clarify the FPC’s
involvement in SMF decision-making. The MPC SMF Concordat is being updated in
parallel. The updated concordats will replace the 2013 versions published on the
Bank’s website.
1
2
3
4
5
6
7
0
1
2
3
4
5
6
P
er cent
Per cent
2001 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17
Weighted average
Basel III definition
(b)(c)
(right-hand scale)
Weighted average
(a)(b)
(left-hand scale)
Adjusted leverage ratio
(d)
(right-hand scale)
Basel III definition of capital
C
hart B.2 Leverage ratios have strengthened
Major UK banks’ leverage ratios
Sources: PRA regulatory returns, published accounts and Bank calculations.
(a) Prior to 2012, data are based on the simple leverage ratio defined as the ratio of
shareholders’ claims to total assets based on banks’ published accounts (note a discontinuity
due to introduction of IFRS accounting standards in 2005, which tends to reduce leverage
ratios thereafter). The peer group used in Chart B.1 also applies here.
(
b) Weighted by total exposures.
(c) The Basel III leverage ratio corresponds to aggregate peer group Tier 1 capital over aggregate
leverage ratio exposure. Up to 2013, Tier 1 capital includes grandfathered capital
instruments and the exposure measure is based on the Basel 2010 definition. From 2014 H1,
Tier 1 capital excludes grandfathered capital instruments and the exposure measure is based
on the Basel 2014 definition. The Basel III peer group used in Chart B.1 also applies here.
(d) Estimated leverage ratio for the major UK banks excluding central bank reserves from the
e
xposure measure. Data are 2017 Q1 for all firms except Co-operative Banking Group
(2016 Q4).
P
art B
B
anking sector resilience 29
…and the United Kingdom is on course to ensure that, by
2022, any remaining barriers to the resolvability of the
largest UK banks have been removed…
A core element of global regulatory reform has been ensuring
t
hat banks can fail safely, without interrupting the provision of
financial services or requiring publicly funded bailouts.
Significant progress has been made to remove barriers to
resolvability but there is still more to do. The United Kingdom
now has a comprehensive and effective bank resolution
regime, under which the Bank has a wide toolkit, including the
power to ‘bail in’ the shareholders and creditors of failed
banks. This requires banks to maintain a minimum amount of
loss-absorbing resources known as ‘minimum requirements for
own funds and eligible liabilities’ (MREL). In May 2017, the
Bank published estimates of the amount of MREL that the
largest UK banks and building societies will be required to
maintain when requirements are implemented in full in
2022.
(
1)
Based on these indicative estimates, the largest UK banks will
be required by 2022 to have aggregate loss-absorbing
resources of 28% of risk-weighted assets on average.
(
2)
Around half of this will be in the form of Tier 1 capital that can
absorb losses before resolution. Large UK banks have issued
around £70 billion of senior unsecured holding company debt
over the past two years, which can readily be bailed in. The
Bank estimates that these banks will need to issue around an
additional £150 billion of MREL-eligible instruments to meet
the indicative 2022 requirements.
(3)
With a view to further
increasing transparency about the resolvability of firms, the
Bank intends to provide summaries of major UK banks’
resolution plans and its assessment of their effectiveness,
including any further steps that need to be taken, by 2019.
(4)
…while firms are on track to implement their ring-fencing
plans by 2019.
The largest UK banks are in the process of separating their core
retail banking activities into ‘ring-fenced banks’ (RFBs), with
investment and international banking activities situated
outside the ring-fence. Ring-fencing will deliver significant
financial stability benefits, by protecting core retail banking
activities from risks associated with activities such as
investment banking, and by enhancing the resolvability of
large banking groups.
(1) Refers to the four UK global systemically important banks (G-SIBs) — Barclays, HSBC,
RBS and Standard Chartered — plus Lloyds Banking Group, Nationwide and
Santander UK.
(2) This includes Basel III capital buffers, and assumes a 1% countercyclical capital buffer.
(3) As a firm’s MREL will depend upon its going concern requirements in a particular year,
the 2022 MRELs are simply indicative and are based on the calibration methodology
set out in the Bank’s Statement of Policy, with reference to the firms’ minimum
capital requirements and balance sheets as at December 2016.
(4) For more details on the UK bank resolution framework, see the Bank’s response to the
Treasury Committee’s inquiry into capital, available at
data.parliament.uk/writtenevidence/committeeevidence.svc/evidencedocument/treas
ury-committee/capital-and-resolution/written/69208.pdf.
3
0 Financial Stability Report
J
une 2017
All relevant firms are already implementing their plans to
meet the statutory deadline of 1 January 2019, though
significant further work remains to be done.
Implementation of ring-fencing carries operational risks.
Firms will need to ensure customers are served, and products
and services provided, from separate legal entities. This will
involve some major reorganisations and transfers, including
m
oving a large number of customers to new sort codes over
the next year. Some IT systems will be separated and direct
access will need to be established for RFBs to financial
infrastructure, such as clearing and payment systems. Firms
have plans in place to mitigate the risk of disruption to
essential financial services during this process, and the PRA is
monitoring the implementation of ring-fencing closely.
Work is also ongoing to build cyber resilience.
Cyber attacks pose a serious threat to the resilience of the
UK financial system. At its June 2017 meeting, the FPC
reviewed progress by the UK authorities and financial sector
in building cyber resilience, and decided to withdraw its
June 2015 Recommendation on CBEST vulnerability testing.
The FPC has also set out its approach to assessing cyber
resilience in the future. These assessments are set out in
Box 7.
Market indicators support a view that UK banks are
resilient…
Reflecting the overall resilience of the UK banking sector, bank
funding costs have remained low since the November Report
(Table B.1). Credit default swap (CDS) premia, which measure
the cost of insuring against bank default, are now close to
post-crisis lows, at just under 60 basis points. And spreads on
additional Tier 1 (AT1) instruments — bonds that convert to
equity if a bank’s capital ratio falls below a certain level — are
at their lowest level on record. The market for European
(including UK) banks’ AT1 instruments has made a strong
recovery following turbulence in February 2016; market
contacts attribute this to an improving economic
environment, a perceived fall in European political risks and
changes in the regulatory treatment of AT1 in Europe.
…and the Bank's Term Funding Scheme has helped to
ensure lending rates have fallen in line with Bank Rate
since July 2016…
In August 2016, the Monetary Policy Committee (MPC) cut
Bank Rate to 0.25%, and launched a Term Funding Scheme
(TFS) to provide funding for banks at interest rates close to
Bank Rate. The TFS was intended to reinforce the
transmission of the cut in Bank Rate to the real economy.
In addition, the TFS provides participants with a cost-effective
source of funding to support additional lending to the real
economy, insuring against the risk that conditions tighten
in bank funding markets.
Domestically focused banks
(c)
Internationally
focused banks
(b)
FTSE All-Share
40
60
80
100
120
140
160
Jan. Apr. July Oct. Jan. Apr.
2016
17
EU referendum
Indices: 1 January 2016 = 100
Chart B.4 Share prices have recovered since their
mid-2016 lows
UK banks’ share prices and FTSE All-Share index since
1 January 2016
(a)
Sources: Thomson Reuters Datastream and Bank calculations.
(a) Bank indices use weighted averages by market capitalisation.
(b) International banks: HSBC and Standard Chartered.
(c) UK domestic banks: Barclays, Lloyds Banking Group and RBS.
£10,000 unsecured loan (household)
(a)
Bank Rate
Two-year fixed-rate mortgage
(a)(b)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4
.5
5.0
Jan. Apr. July Oct. Jan. Apr.
2016 17
P
er cent
Chart B.3 Quoted rates on new real-economy lending
have fallen broadly in line with Bank Rate
Average interest rates on new lending and Bank Rate
Sources: Bank of England and Bank calculations.
(a) Sterling-only end-month quoted rates.
(b) On mortgages with a loan to value ratio of 75%.
T
able B.1 Funding costs have fallen
Selected measures of UK banks’ funding costs
(a)(b)
Global
Pre-crisis financial July 2016 Nov. 2016
(1 Jan. 2007) crisis Report Report Latest
A
dditional Tier 1
(c)
– – 737 660 350
Covered bond
(
d)
-24 218 11 3 -1
S
enior unsecured bond
(e)
– 368 96 59 42
S
enior CDS
(f)
5 222 134 97 56
Sources: Bloomberg, Markit Group Limited and Bank calculations.
(a) UK banks are Barclays, HSBC, Lloyds Banking Group and RBS.
(b) Funding spreads are measured in basis points.
(c) Simple average of secondary market spreads over government bonds.
(d) Constant-maturity simple average of secondary market spreads to swaps for five-year euro-denominated
covered bonds or a suitable proxy.
(e) Constant-maturity simple average of secondary market spreads to mid-swaps for five-year
euro-denominated senior unsecured bonds, or a suitable proxy when unavailable.
(f) Simple average of five-year senior CDS premia.
P
art B
B
anking sector resilience 31
When the TFS was launched, the MPC stated that the value
of funding provided by the TFS would be determined by usage
of the scheme. As of June 2017, aggregate outstanding TFS
drawings by UK banks and building societies were £65 billion.
A
s set out in the Bank’s 2017 Q1 Credit Conditions Review,
most major UK lenders reported that the TFS had contributed
to recent falls in bank funding spreads. As a result, since the
TFS was introduced, quoted rates on new real-economy
lending have, overall, fallen broadly in line with the reduction
in Bank Rate as intended (Chart B.3).
…but low profitability may pose a risk to resilience in future.
UK banks’ equity prices have recovered from their mid-2016
lows (Chart B.4). While aggregate return on equity for the
largest UK banks was just below zero in 2016, ‘underlying’
returns, which strip out misconduct costs and one-time
charges such as restructuring costs, remained positive, at
around 6%. 2017 Q1 results showed material improvement in
profits, driven by a partial recovery in investment banking
returns (Chart B.5).
Price to book ratios, which measure the market value of equity
relative to the value of equity recorded on banks’ balance
sheets, remain persistently below one (Chart B.6). The FPC
continues to judge that the low equity prices of UK banks can
likely be explained by anticipated misconduct redress costs
and weak expected operating profitability of investment
banking services in particular, rather than by market concerns
about asset quality. For major global banks generally, the
correlation between price to book ratios and expected returns
on equity is currently above 90% (Chart B.7), and much
higher than the correlation between price to book ratios and
measures of asset quality. This means that low price to book
ratios for UK banks do not signify that their ability to absorb
losses is diminished.
Weak profitability does affect banks’ future ability to rebuild
capital following any shock that results in losses while also
maintaining credit supply. The exploratory scenario in the
2017 stress test will consider how the resilience of the UK
banking system might evolve if recent headwinds to bank
profitability persist and intensify. The purpose of this scenario
is to explore the impact of banks’ actions on both the real
economy and the future resilience of the system to shocks.
The aggregate results of the test will be published in
November 2017.
(1)
Statutory RoE (year-end)
Underlying RoE (year-end)
Statutory RoE (2017 Q1)
Underlying RoE (2017 Q1)
+
–
1
5
10
5
0
5
1
0
15
20
2005 06 07 08 09 10 11 12 13 14 15 16 2017
Q1
Per cent
C
hart B.5 Bank profitability remains weak
UK banks’ statutory and underlying return on equity
(a)(b)(c)(d)
Sources: Published accounts and Bank calculations.
(a) Weighted average by shareholders’ equity.
(b) Statutory return on equity (RoE) is defined as net income attributable to shareholders divided
by average shareholders’ equity. Underlying RoE strips out misconduct costs as well as
one-time charges such as restructuring costs.
(c) UK banks are Barclays, HSBC, Lloyds Banking Group and RBS.
(d) 2017 Q1 results are annualised and may display seasonality. They are not directly
comparable to full-year results.
0.0
0.2
0.4
0.6
0.8
1.0
1
.2
1.4
1.6
1
.8
2
.0
2.2
2
.4
2007 08 09 10 11 12 13 14 15 16 17
Price to book ratio
Chart B.6 Price to book ratios have improved but
remain below one
UK banks’ average price to book ratio
(
a)(b)(c)
So
urc
es
:
Tho
m
s
o
n
R
euters
Da
ta
s
trea
m
a
n
d
B
a
n
k
c
a
l
c
ul
a
tio
n
s
.
(a
)
U
K
b
a
n
k
s
a
re
B
a
rc
l
a
y
s
,
HSB
C
,
Ll
o
y
d
s
B
a
n
k
in
g
G
ro
up
a
n
d
R
B
S.
(b
)
R
el
a
tes
the
s
ha
re
p
ric
e
w
ith
the
b
o
o
k
,
o
r
a
c
c
o
un
tin
g
,
v
a
l
ue
o
f
s
ha
reho
l
d
ers
’
equity
p
er
s
ha
re.
(c
)
HSB
C
p
ric
e
to
b
o
o
k
ra
tio
a
d
j
us
ted
f
o
r
c
urren
c
y
m
o
v
em
en
ts
.
UK banks
Other global banks
0
2
4
6
8
10
12
14
16
18
0.0 0.5 1.0 1.5 2.0 2.5
Expected 2018 RoE (per cent)
Price to book ratio
Chart B.7 Price to book ratios are broadly in line with
expected return on equity
Price to book ratios for major global banks compared with
expected returns on equity
Sources: Bloomberg, Thomson Reuters Datastream and Bank calculations.
(1) The Bank does not intend to publish individual bank results under the exploratory
scenario, based on considerations around the possible commercial sensitivity of the
projections banks will provide.
3
2 Financial Stability Report
J
une 2017
Box 7
Building cyber resilience in the UK financial
system
Cyber attacks pose a threat to the stability of the provision of
U
K financial services. There has been a growing number of
high-profile cyber incidents in recent years, some of which
have involved financial sector firms. These incidents highlight
the importance of firms developing capabilities to protect
against, respond to and recover from cyber attacks.
Progress made towards building cyber resilience in the
financial system
Since 2013, the FPC has made two Recommendations to
regulators regarding cyber resilience (Table 1). This box
describes how these Recommendations have been
implemented, and sets out the FPC’s future approach.
Since the 2013 Recommendation, the organisation of
authorities who have a role in strengthening the resilience
of the financial system to cyber risks has been improved.
The National Cyber Security Centre (NCSC) has been
established as the main technical authority for UK cyber
security in order to share knowledge and address systemic
vulnerabilities. UK financial authorities, together with the
NCSC and the industry, have since improved information
sharing across the financial sector.
A response framework allows the UK authorities to
co-ordinate their response to cyber attacks that have affected,
or have the potential to affect, the financial sector. It now
includes the NCSC and, if appropriate, the National Crime
Agency.
At the international level, the G7 Cyber Expert Group has
published guidance on cyber security for the financial sector.
(1)
The group is now developing guidance for effective cyber
security assessment; identifying and treating cyber risk arising
from third parties; and co-ordinating with other critical
sectors such as telecommunications and energy.
The FPC’s Recommendations have catalysed testing of the
resilience of core financial companies to cyber attack.
The 2015 FPC Recommendation called for financial services
firms and financial market infrastructures at the core of the
UK financial system (‘core firms’) to complete so-called CBEST
tests. These tests subject firms to simulated cyber attacks,
and are tailored to each firm, drawing on government,
intelligence agency and private sector expertise. Since 2014,
31 out of 34 firms have finished testing, and two further firms
are close to completion. This includes banks representing
more than 80% of the outstanding stock of PRA-regulated
banks’ lending to the UK real economy. The first round of
CBEST testing is therefore materially complete.
The results of testing, together with the actions taken to
address weaknesses, demonstrate that core firms have made
significant progress in building cyber resilience.
The first round of tests served its intended purpose by
identifying weaknesses in core firms’ cyber resilience. For
obvious security reasons, these results will not be detailed.
But where shortcomings were identified, firms are
implementing action plans to remedy the issues, overseen by
supervisors. In some cases, controls on the integrity of
systems and confidentiality of data needed to be
strengthened. In others, the tests identified the need for
further investment in capabilities to detect, mitigate and
respond to attacks. And in general, the tests highlighted the
importance of firms continuing to invest in their people,
processes and technology in order to counter the risks of
cyber attack.
Consistent with the FPC’s Recommendation, CBEST will
become a regular component of supervisory assessment of
firms. Core firms will be expected to conduct their own
regular tests of cyber resilience. They will also be subject to
supervisor-led CBEST testing at regular intervals. The
frequency of these tests will be proportionate to firms’
importance for financial stability. CBEST is also being adopted
in other jurisdictions and sectors.
Based on the progress made, the FPC judges that its
2015 Recommendation has been implemented, and that
it is appropriate to withdraw its Recommendation.
Table 1 The FPC’s Recommendations on cyber resilience
In June 2013, the FPC recommended that:
‘HM Treasury, working with the relevant government agencies, the PRA, the
Bank’s financial market infrastructure supervisors and the FCA should work
with the core UK financial system and its infrastructure to put in place a
programme of work to improve and test resilience to cyber attack.’
In July 2015, the June 2013 Recommendation was replaced with the
following Recommendation:
‘The FPC recommends that the Bank, the PRA and the FCA work with firms at
the core of the UK financial system to ensure that they complete CBEST tests
and adopt individual cyber resilience action plans. The Bank, the PRA and the
FCA should also establish arrangements for CBEST tests to become one
component of regular cyber resilience assessment within the UK financial
system.’
Source: Bank of England.
(1) ‘G7 fundamental elements for cyber security’, October 2016,
www.gov.uk/government/publications/g7-fundamental-elements-for-cyber-security.
P
art B
B
anking sector resilience 33
The next phase in building and maintaining resilience
Given the progress made and the lessons from work to date,
the FPC is now moving to the next stage.
Its focus is on systemic risk, rather than risk to individual
companies or consumers. That is, the risk of cyber attack
causing disruption to critical financial services on a scale that
causes material disruption to the UK economy.
The FPC is setting out the elements of the framework of
regulation for the UK financial system’s cyber resilience that
are necessary to mitigate systemic risk. The FPC’s future role
will be to check that each element is being fulfilled by the
relevant authorities.
The FPC judges that effective regulation requires:
(1) Clear baseline expectations for firms’ resilience that
reflect their importance for the financial system.
Working with the Bank, PRA and FCA, the FPC will consider
its tolerance for the disruption to important economic
functions provided by the financial system. Supervisors
will set clear expectations for firms consistent with this
tolerance.
(2) Regular testing of resilience by firms and supervisors.
This will build on the first round of CBEST testing and
ensure the most systemic firms are subject to regular
checks, with the frequency and scope of the tests set in
line with supervisory strategy. This will allow supervisors
to keep pace with the evolving nature of the risk.
Assessment of the system’s resilience will also be informed
by sector-wide simulation exercises, which test the
industry’s response and recovery capabilities. The FPC has
asked for regular reports from supervisors on the systemic
risks exposed by these regular tests and exercises.
(3) Identification of firms that are outside the financial
regulatory perimeter, but which may be important for
regulated firms.
Cyber attacks that result in disruption to third-party
p
roviders outside the regulatory scope could still have
effects on the financial system. The FPC has therefore
requested annual updates from the financial authorities,
which include HM Treasury, the Bank, PRA and FCA, on the
cyber resilience of firms that are outside the regulatory
perimeter, but which are important for the UK financial
sector.
(4) Clear and tested arrangements to respond to cyber
attacks when they occur.
UK authorities co-ordinate their response to cyber attacks
that may potentially affect the financial system. The
financial authorities regularly test and review these
arrangements, in line with evolving threats and experience
gained from managing incidents. Co-operation with the
industry can also enhance the sector’s capacity to respond
to threats and share information during an incident.
The FPC has requested an annual update on the
effectiveness of the response framework from financial
authorities, to check that the system has the capacity to
respond to and recover from a cyber attack.
Where the FPC receives reports that demonstrate
shortcomings in the regulatory framework, it will consider
using its statutory powers to make Recommendations to
the relevant authorities that these be remedied.
Market-based finance
3
4 Financial Stability Report
J
une 2017
Market-based finance accounts for almost half of the UK financial system’s total assets, and
supports the provision of financial services to the real economy. The provision of market-based
finance relies on core financial intermediaries, such as dealers, which remain resilient. But there is
evidence that conditions in some markets, such as the repo market, have declined in recent years,
as dealers are less willing to act as intermediaries. High demand for market liquidity in stresses,
including from open-ended funds, also remains a risk. Life insurance companies, meanwhile,
continue to support market-based finance through increased holdings of illiquid assets.
Market-based finance is an increasingly important source of
financing for the UK real economy.
The importance of market-based finance as a means of
providing finance to the real economy has grown over the
past few years. Non-bank financial institutions represent key
sources of market-based finance and account for almost
half of the UK financial system’s total assets, up by
10 percentage points since 2009. These institutions provide
financial services to the real economy, including by investing
in capital markets, such as equity and corporate bond markets
(Chart B.8).
On a cumulative basis, capital markets account for almost all
net finance raised by UK private non-financial corporations
(PNFCs) since the global financial crisis, primarily in the form
of bond issuance. In 2017 to date, sterling investment-grade
issuance by UK companies has been broadly in line with its
2010–16 average.
In April, the Bank completed its purchase of £10 billion of
investment-grade corporate bonds under the Corporate Bond
Purchase Scheme (CBPS), announced as part of the Monetary
Policy Committee’s policy package in August 2016.
Due to its importance for the UK real economy, the FPC has
established a medium-term priority of completing post-crisis
reforms to market-based finance in the United Kingdom, and
improving the assessment of systemic risks across the financial
system (see The FPC’s medium-term priorities chapter).
The provision of market-based finance relies on core
financial intermediaries, such as dealers, which remain
resilient…
Some financial markets, such as cash fixed-income markets,
rely on dealers to intermediate between clients, including by
building and releasing inventories as part of their
market-making activity. Measures of dealer resilience remain
15
10
5
0
5
10
15
20
2010 11 12 13 14 15 16 17
Loans
M
arket-based finance
(b)
T
otal
(c)
£ billions
+
–
Chart B.8 Market-based finance is an important source
of financing for UK companies
Net finance raised by UK private non-financial corporations
(PNFCs)
(a)
Sources: Bank of England and Bank calculations.
(a) Finance raised by PNFCs from UK MFIs and from capital markets. Data cover funds raised in
both sterling and foreign currency, converted to sterling. Seasonally adjusted. Bonds and
commercial paper are not seasonally adjusted.
(b) Market-based finance is composed of bonds, equities and commercial paper.
(c) Owing to the seasonal adjustment methodology, the total series may not equal the sum of
its components.
P
art B
M
arket-based finance 35
robust. The aggregate leverage ratio of the world’s largest
dealers was 5.2% at end-December (Chart B.9). Since the
November 2016 Report, market perceptions of UK dealers’
credit risk, as measured by the cost of default protection
(
CDS premia), have decreased.
…while reforms to over-the-counter (OTC) derivatives
markets are ongoing.
Through OTC derivatives markets, dealers and other financial
institutions are exposed to counterparty credit risk related to
clients, one another and — for those markets that are cleared
— central counterparties (CCPs). In March 2017 new rules
were introduced to implement international requirements for
the margining of non-cleared OTC derivatives, which aim to
improve the mitigation of counterparty credit risk in those
transactions. These rules are one element of the G20 reforms
to derivatives markets agreed in 2009. In November 2016, the
FPC asked the Bank to conduct an in-depth assessment to
examine progress towards the implementation of the
post-crisis reforms in derivatives markets and the implications
for the resilience of the financial system. This assessment
continues to feed into broader work by the Financial Stability
Board (FSB).
But there is evidence that conditions in some markets, such
as the repo market, have declined over recent years, where
dealers are less willing to act as intermediaries.
Market liquidity refers to the ability of investors to buy and
sell assets in reasonable size, and within a reasonable time
frame, without having a large impact on prevailing prices.
When market liquidity is reliable, it encourages participation in
financial markets, by providing confidence both for issuers
(who want to be able to borrow when required at competitive
terms) and for investors (who want to be able to move
smoothly in and out of positions).
(1)
The provision of
market-based finance is more likely to be stable when financial
markets are liquid and function smoothly. The resilience of
market liquidity remains uneven across markets.
Repo markets contribute to effective market functioning by
enabling market makers, such as dealers, to finance their
inventories, and leveraged investors, such as hedge funds, to
transact in securities, thus supporting market liquidity. The
July 2016 Report noted that there has been a decline in the
availability of, and increase in the cost of, repo financing in
some jurisdictions. This is consistent with findings from a
cross-jurisdiction study of developments in repo markets
conducted by the Committee on the Global Financial System
(CGFS), published in April 2017.
(2)
(1) Fair and Effective Markets Review, Final Report, June 2015;
www.bankofengland.co.uk/publications/Pages/news/2015/055.aspx.
(2) www.bis.org/publ/cgfs59.htm.
0
1
2
3
4
5
6
7
2000 02 04 06 08 10 12 14 16
Interquartile range
W
eighted average
P
er cent
C
hart B.9 Aggregate dealer leverage ratios have
r
emained high in the second half of 2016
Dealers’ leverage ratios
(a)(b)
Sources: Banks’ published accounts, SNL Financial, The Banker Database and Bank calculations.
(a) Leverage ratio defined as reported Tier 1 capital (or common equity where not available)
divided by total assets, adjusted for accounting differences on a best-endeavours basis. This
accounting measure differs from regulatory leverage ratios.
(b) Dealers included are Bank of America Merrill Lynch, Barclays, BNP Paribas, Citigroup,
C
redit Agricole, Credit Suisse, Deutsche Bank, Goldman Sachs, HSBC, JP Morgan,
Mitsubishi UFJ, Morgan Stanley, RBS, Société Générale and UBS. Pre-crisis data also include
Bear Stearns, Lehman Brothers and Merrill Lynch.
3
6 Financial Stability Report
J
une 2017
The CGFS report noted that in some jurisdictions there is
evidence of dealers being less willing or able to undertake repo
market intermediation in recent years. In some cases, this has
led to other financial institutions finding it difficult to place
c
ash. One example cited is that only four of the eleven
counterparties that had offered repo facilities to a major
European asset manager in 2014 were still willing to do so in
2016 (Chart B.10). The CGFS report also provides evidence of
increased costs faced by end-users borrowing in the repo
market; for example, repo rates (relative to expectations of
policy interest rates) paid by pension funds to borrow cash in
the UK gilt repo market increased around fourfold between
2014 and 2016 (Chart B.11). At the same time, the report
notes that the rate at which end-users place cash in repo has
remained broadly constant since late 2014.
One indicator of reduced repo market liquidity is the volatility
in repo rates that has started to occur around reporting dates
as dealers reduce their repo activity (Chart B.12). Dislocations
in European government bond repo markets at end-2016 were
particularly acute, with gilt repo rates falling to -0.5%, the
lowest level reached in the past decade. The fall in rates at the
end of 2017 Q1 was more muted, which market contacts
attribute in part to participants improving their preparations
following the challenging year-end.
The CGFS report identifies several drivers behind these
changes in repo market conditions. One driver has been
changes in the economic environment, such as those driven by
central banks. Exceptional monetary policy measures reduce
the demand for central bank reserves to meet daily liquidity
needs while increasing the supply, including through asset
purchases, of those reserves. Changes in the regulatory
environment have also played a role — particularly those that
act on the size, as well as the composition, of banks’ balance
sheets, such as the introduction of leverage ratio
requirements. In this context, the FPC will consider the
impact of changes to international standards on the
calibration of the UK capital and leverage framework (see
Banking sector resilience chapter).
High demand for market liquidity, including from
open-ended funds invested in less liquid assets, remains
a risk.
The functioning of some markets could be tested by high
demand for liquidity, including from open-ended investment
funds. These funds offer short-term redemptions to investors
while investing in some cases in longer-dated and potentially
illiquid assets. Assets under management in such funds
increased by 80% between 2008 and 2015.
Large-scale redemptions could result in sales of assets by
funds that exceed the ability of dealers and other investors to
absorb them, potentially impairing market liquidity. These
effects could be amplified if resulting falls in prices lead to
0
10
20
30
4
0
50
60
7
0
J
an.
July
Per cent
2014 15 16 17
J
an.
July
Jan.
J
uly
J
an.
C
hart B.10 Dealers, who act as counterparties in repo
t
rades, appear to be less willing to offer repo facilities
Proportion of cash balance placed in repo of a major European
asset manager, split by counterparty
(a)
Source: A European asset manager.
(a) Each colour represents one repo counterparty, with eleven in total.
0.0
0.1
0.2
0.3
0
.4
0
.5
0.6
2014 15 16
Per cent
One-month
T
hree-month or four-month
F
ive-month
Six-month
Chart B.11 For UK pension funds, the cost of borrowing
cash through repo has increased fourfold since 2014
Gilt repo rates paid by a group of pension fund asset managers in
excess of expectations of policy interest rates
(a)
Source: Bloomberg, data collected from a number of asset managers and Bank calculations.
(a) As implied by overnight index swap for the relevant term.
1000
800
600
400
200
0
200
Oct. Jan. Apr.
United Kingdom
France
Germany
Indices: 1 Oct. 2016 = 100
+
–
2016 17
Chart B.12 Overnight repo rates in Europe showed
year-end volatility
UK, French and German repo rates
Sources: Bloomberg and Bank calculations.
P
art B
M
arket-based finance 37
further redemptions by investors. This potential imbalance
between the demand for, and supply of, market liquidity could
be particularly acute if investor redemptions were
accompanied by volatility rising from its current low level,
w
hich could lead to a reduction in dealers’ market-making
capacity.
Following the UK referendum on EU membership, funds
invested in UK commercial real estate suffered significant
redemptions and some suspensions, with implications for
market liquidity (see the November 2016 Report). The
Financial Conduct Authority (FCA) subsequently published a
discussion paper in February 2017 covering all open-ended
funds investing in illiquid assets.
(
1)
The paper aims to gather
evidence to decide whether changes to the FCA’s regulatory
approach are needed to enhance market stability and promote
competition in the sector, while protecting consumers.
The FSB also finalised in January its policy recommendations
to address related structural vulnerabilities from asset
management activities.
(2)
The key recommendations are:
(i) funds’ investment strategies should be consistent with their
redemption terms; (ii) authorities should give consideration to
system-wide stress testing; and (iii) globally consistent
measures of leverage should be developed. The International
Organization of Securities Commissions (IOSCO) is currently
operationalising the FSB recommendations, and will consult
on draft updated liquidity risk management principles.
Life insurance companies, meanwhile, continue to support
market-based finance through increased holdings of illiquid
assets.
Life insurance companies are important investors in financial
assets and hence in supporting market-based finance. In the
United Kingdom, life insurers hold £1.8 trillion of investment
and cash assets and account for a significant proportion of the
total assets outstanding in several UK securities markets.
Low market interest rates continue to provide an incentive for
life insurers to invest in more illiquid assets such as
lower-rated fixed-income securities and real-economy assets,
such as equity release mortgages, commercial real estate and
infrastructure, in order to earn higher returns. For example,
property-related non-linked exposures (ie exposures where
insurance firms bear all or part of the market risk on asset
holdings) increased from £93 billion in 2016 Q3 to £99 billion
in 2016 Q4 (Chart B.13), to stand at 11% of total non-linked
assets of UK insurance companies.
(1) www.fca.org.uk/publications/discussion-papers/illiquid-assets-open-ended-
investment-funds.
(2) www.fsb.org/2017/01/policy-recommendations-to-address-structural-vulnerabilities-
from-asset-management-activities/.
0
20
40
60
80
100
120
Mortgages and loans
Direct property
Real estate collective investment schemes (CIS)
£ billions
Q1
Q2
Q3
Q4
2016
Chart B.13 UK life insurers are increasingly investing in
illiquid assets
(a)
Stock of property-related non-linked illiquid assets during 2016
(b)
Sources: Solvency II submissions and Bank calculations.
(a) Trend is also driven in part by insurers reclassifying illiquid assets.
(b) Illiquid assets cover: direct property; mortgages and loans; and CIS real estate funds.
3
8 Financial Stability Report
J
une 2017
Insurers are further incentivised to invest in illiquid assets to
match long-dated stable liabilities, such as annuities; under
Solvency II, the ‘matching adjustment’ allows insurers to look
through the impact of short-term market movements on
a
ssets when valuing such liabilities.
(
1)
I
n the November 2016
Report, the FPC concluded that the matching adjustment is
beneficial from a macroprudential perspective by reducing
potential instability across the financial system. UK annuity
writers are well-matched between their assets and liabilities
versus their European counterparties, as evidenced in the
European Insurance and Occupational Pensions Authority
(EIOPA) 2016 stress test.
Matching long-dated illiquid liabilities with suitable
long-dated illiquid assets can help ensure availability of
finance to the wider economy, but can also present risks
where firms may not have adequate systems and controls to
manage the risks associated with these assets. The PRA issued
a consultation on this topic in December 2016.
(2)
The FPC supports ongoing work by the Bank and other
authorities to strengthen the regulatory framework for, and
oversight of, non-bank financial institutions.
Following efforts by the G20 and the FSB to strengthen the
oversight and regulation of the ‘shadow banking’ system, in
April the European Parliament agreed reforms to money
market funds, likely to come into force in 2018. The reforms
are aimed at making these funds more robust, in order to
maintain the essential role that money market funds play in
financing the real economy.
The FCA has continued to review the regulatory framework for
peer-to-peer platforms in light of the sector’s rapid growth
and developments in firms’ business models. In their interim
feedback report on this work, the FCA raised a number of
concerns including around information quality, inconsistent
disclosures and insufficient wind-down procedures for some
firms. The FCA has further set out expectations of firms that
operate loan-based crowdfunding platforms that provide loans
to businesses. The Bank has also contributed to a CGFS and
FSB report on FinTech credit.
(3)
The report concludes that in
future FinTech could improve financial stability by providing
an alternative source of finance. But it may also give rise to
systemic concerns, for example by lowering lending standards,
procyclical credit provision, and by posing challenges to the
regulatory perimeter. The FPC supports this work and will
continue to monitor the sector.
(1) The matching adjustment replaces the so-called ‘liquidity premium’ under the former
Individual Capital Adequacy Standards regime.
(2) ‘Solvency II: matching adjustment — illiquid unrated assets and equity release
mortgages’, Prudential Regulation Authority Consultation Paper CP48/16;
www.bankofengland.co.uk/pra/Pages/publications/cp/2016/cp4816.aspx.
(3) www.bis.org/publ/cgfs_fsb1.pdf.
P
art B
M
arket-based finance 39
Box 8
The UK High-Value Payment System
Electronic payments are essential to the functioning of
modern economies. Most electronic payments are made via
‘payment systems’ in which financial institutions participate to
exchange money on behalf of their customers. Each payment
s
ystem maintains: (a) a set of rules and procedures that
govern the transfer of funds between these institutions; and
(b) an infrastructure to facilitate those transfers. The largest
and most critical payments in the United Kingdom are made
over CHAPS, the country’s High-Value Payment System
(HVPS).
The scale, type and sophistication of potential threats to the
stability of payment systems are rising. Risks such as
cyber attack are increasingly capable of striking at any part of
the payments chain. It is essential that payment systems are
able to take an end-to-end view of these risks. Having
identified constraints in the existing structure, the Bank
decided in April, with the FPC’s endorsement, to move to
‘direct delivery’ of the HVPS. The Bank is now working closely
with CHAPS Co, the current operator of the HVPS and with
industry to ensure a smooth and timely transition to the new
structure this calendar year.
This box explains the rationale for the decision to move to
direct delivery. It then places it in the broader context of the
Bank’s programme for renewal of its Real-Time Gross
Settlement (RTGS) infrastructure. The Bank’s aim is for this
programme to deliver a materially stronger, more resilient,
flexible and highly innovative sterling settlement system for
the United Kingdom’s high-value payments. Moving to direct
delivery of the HVPS is a vital part of that vision.
Current delivery of HVPS
The current delivery model for the UK HVPS involves a split in
responsibilities across two institutions. The core infrastructure
is provided by the Bank, as part of its RTGS system.
CHAPS Co, a small private sector company owned by its
members is responsible for operating the system’s governance
and rulebook and managing risks across the HVPS as a whole;
CHAPS Co is supervised by the Bank of England. This split is
unusual internationally. The norm in most countries is for the
central bank to deliver all aspects of the HVPS.
There are structural deficiencies in this model that pose risks
to financial stability. The RTGS system has several policy
functions, including the implementation of monetary policy.
For security reasons therefore CHAPS Co cannot be given the
full range of information on, or contractual control over, RTGS
that it needs to identify and manage risks across the whole
payment system. This includes in important areas of
operational risk such as cyber risk and fraud. The IMF
highlighted these constraints in its past two Financial Stability
Assessment Programmes on the United Kingdom, most
recently in 2016. Such risks are increasingly important and
c
omplex. Their potential impact is further amplified as new
types of users have sought access to payment systems. At the
same time, the tolerance for even short periods of system
outage has fallen.
CHAPS Co had made a number of important enhancements to
the existing model in recent months to reflect some of these
concerns. But these enhancements could not fully address the
structural constraints described above. As such, CHAPS Co
was unable to fully meet regulatory expectations.
On this basis, the FPC concluded at its 27 April meeting that
there were financial stability risks arising from the current
structure for delivery of the HVPS, including from the growing
threat from cyber attack.
Direct delivery model for HVPS
Under a direct delivery model, the Bank would be responsible
for both the operation of the HVPS scheme and the RTGS
infrastructure. This would allow the Bank to carry out
end-to-end risk management, identifying and responding to
emerging risks in the HVPS in a holistic way, using the full
range of tools at its disposal. The Bank would maintain its role
as the supervisor of the HVPS.
The FPC welcomed the Bank’s plan to mitigate the financial
stability risks through a move to the proposed direct delivery
model for operating the HVPS.
The move towards a direct delivery model is an opportunity
for the Bank to provide the FPC with more comprehensive
information on the HVPS. This is particularly important given
the growing risk environment. The FPC will be given regular
reports on the extent of systemic risks arising from the
operation of the HVPS. In addition, where the Bank proposes
material changes to the delivery of the HVPS in future that
affect risks to financial stability, it will consult the FPC.
The Bank has designed a model for the operation and
supervision of the HVPS that will involve transparency, user
voice and independent challenge. The Bank, CHAPS Co and
their stakeholders expect to complete the transition to direct
delivery by the end of the calendar year.
The RTGS renewal programme
The move to direct delivery of the HVPS is part of a wider
programme to reshape both the infrastructure and governance
of the RTGS system. The Bank published a blueprint for the
new generation of its RTGS system in May 2017 (Table 1).
On 27 April, the FPC met to discuss the current structure for
delivering the UK High-Value Payment System (HVPS), in the
context of evolving risks to systemically important financial
market infrastructure. It agreed that there were financial
stability risks arising from the current structure for delivery of
the UK HVPS and welcomed the Bank’s plan to mitigate these
risks through a move to the proposed direct delivery model
(see Box 8).
4
0 Financial Stability Report
J
une 2017
The renewed RTGS service will be delivered through a
multi-year programme of work. This includes the
procurement of a new central infrastructure.
R
TGS is vital UK infrastructure. It supports both the
UK economy and the Bank’s own balance sheet. At the same
time the renewal programme is complex, unique and large. It
will require co-ordinated changes across the payments
industry. It is critical that there are no disruptions to the RTGS
service. The Bank has planned accordingly; new features will
be introduced in phases. The Bank’s current intention is for
the majority of new functionality to be live by end-2020.
T
able 1
S
ervice characteristics of the renewed RTGS service
Service characteristic Objective
Resilience Strengthen resilience of RTGS and flexibility to
respond to emerging threats.
Access Facilitate greater direct access to central bank
money settlement for financial institutions and
infrastructures.
Interoperability Promote harmonisation and convergence with
critical domestic and international payment
systems.
User functionality Support emerging user needs in a changing
payment environment.
End-to-end risk Strengthen capacity to respond to evolving
management system-wide risks.
Source: ‘A blueprint for a new RTGS service for the United Kingdom’; www.bankofengland.co.uk/markets/
Documents/paymentsystem/rtgsblueprint.pdf.
P
art B
T
he FPC’s medium-term priorities 41
To help to meet its objectives, alongside its ongoing
assessment of the risk environment, the FPC is prioritising
three initiatives over the next two to three years:
(1) Finalising, and refining if necessary, post-crisis bank
capital and liquidity reforms
Over the past three years, in line with international reforms
following the global financial crisis, the FPC has focused
on establishing the medium-term capital framework for
UK banks and has reviewed the Bank’s work to stop banks
being ‘too big to fail’.
The FPC will now take stock of these reforms and will develop
more ways to assess how resilient the banking system would
be to shocks. It will:
• Review the judgements underlying its overall calibration of
the risk-weighted capital framework for UK banks, including
taking into account progress towards ensuring banks are
resolvable, changes in accounting standards and reforms
to the measurement of risk-weighted assets. The natural
point for a full review will be in 2019, in light of the
outcome of negotiations to finalise Basel III standards;
the FPC has already started reviewing some of these
judgements.
• Review the UK leverage ratio framework, in light of
progress towards an international standard for a leverage
requirement, and its scope of application.
• Take stock of overall liquidity and funding standards for
UK banks and consider the case for a time-varying
macroprudential liquidity standard.
• Input to the Bank’s work to develop its stress tests of the
banking system, to capture better the interactions between
banks in a stress.
(2) Completing post-crisis reforms to market-based finance
in the United Kingdom, and improving the assessment of
systemic risks across the financial system
Firms other than banks play an important role in providing
finance to the economy and a means of sharing risk. Their
activities can also be a source of systemic risk. Since 2014,
the FPC has completed an annual review of risks from, and
regulation of, market-based finance, with an in-depth look so
far at the activities of open-ended investment funds and
insurers, and on changes in market liquidity. The FPC will now:
• Continue its annual review and programme of in-depth
reviews on specific market-based finance activities. This
will include an assessment of the financial stability risks
associated with derivatives transactions, including progress
towards implementation of the mandated move to the use
of central clearing and other post-crisis reforms.
• Support the Bank’s work to develop a system-wide stress
simulation, to help understand how the financial system as
a whole is likely to respond to shocks.
• Consider whether macroprudential tools for market-based
finance might be needed to address systemic risks
originating from outside the banking system.
(3) Preparing for the United Kingdom’s withdrawal from the
European Union
Exit negotiations between the United Kingdom and the
European Union have begun. There are a range of possible
outcomes for, and paths to, the United Kingdom’s withdrawal
from the European Union.
The FPC will oversee contingency planning to mitigate risks to
financial stability as the withdrawal process evolves.
Irrespective of the particular form of the United Kingdom’s
future relationship with the European Union, and consistent
with its statutory responsibility, the FPC will remain
committed to the implementation of robust prudential
standards in the UK financial system. This will require a level
of resilience to be maintained that is at least as great as that
currently planned, which itself exceeds that required by
international baseline standards.
Across its priorities, the FPC remains committed to working
with relevant authorities domestically and internationally, to
protect and enhance the resilience of the UK financial system.
The FPC’s medium-term priorities
42 Financial Stability Report June 2017
Recommendations implemented or withdrawn since the previous Report
14/Q2/1 FPC Recommendation on mortgage affordability tests Superseded by 17/Q2/1 — see below
When assessing affordability, mortgage lenders should apply an interest rate stress test that assesses whether borrowers
could still afford their mortgages if, at any point over the first five years of the loan, Bank Rate were to be 3 percentage
points higher than the prevailing rate at origination. This Recommendation is intended to be read together with the FCA
requirements around considering the effect of future interest rate rises as set out in MCOB 11.6.18(2).
This Recommendation has been superseded by 17/Q2/1, to clarify the rate to which the 3 percentage points stress should be
applied, following a review by the FPC. Details of the review and the reasons for the clarification are set out in a chapter of this
Report: ‘The FPC’s approach to addressing risks from the UK mortgage market’.
This annex lists FPC Recommendations from previous periods that have been implemented since
the previous Report, as well as Recommendations and Directions that are currently outstanding.
It also includes those FPC Policy decisions that have been implemented by rule changes and are
therefore still in force.
Each Recommendation or Direction has been given an identifier to ensure consistent referencing over time. For example, the
identifier 14/Q2/1 refers to the first Recommendation made at the 2014 Q2 Committee meeting.
Annex 1: Previous macroprudential policy decisions
14/Q3/1 Powers of Direction over housing instruments Implemented
The FPC recommends that HM Treasury exercise its statutory power to enable the FPC to direct, if necessary to protect and
enhance financial stability, the PRA and FCA to require regulated lenders to place limits on residential mortgage lending,
both owner-occupied and buy-to-let, by reference to: (a) loan to value ratios; and (b) debt to income ratios, including
interest coverage ratios in respect of buy-to-let lending.
Legislation granting the FPC powers of Direction over loan to value and debt to income limits in respect of mortgages on
owner-occupied properties came into force in April 2015.
Over 2015 and 2016, HM Treasury consulted on granting the FPC powers of Direction over buy-to-let lending. It published a
consultation response document and laid the final legislation before Parliament on 16 November. These powers were approved
in December and are now in place.
The FPC therefore decided to consider this Recommendation as implemented at its meeting on 22 March 2017.
Annex 1 Previous macroprudential policy decisions 43
Recommendations and Directions currently outstanding
17/Q2/1 FPC Recommendation on mortgage affordability tests Action under way
When assessing affordability, mortgage lenders should apply an interest rate stress test that assesses whether borrowers
could still afford their mortgages if, at any point over the first five years of the loan, their mortgage rate were to be
3 percentage points higher than the reversion rate specified in the mortgage contract at the time of origination (or, if the
mortgage contract does not specify a reversion rate, 3 percentage points higher than the product rate at origination). This
Recommendation is intended to be read together with the FCA requirements around considering the effect of future
interest rate rises as set out in MCOB 11.6.18(2). This Recommendation applies to all lenders which extend residential
mortgage lending in excess of £100 million per annum.
This Recommendation was made at the FPC’s meeting on 21 June 2017. Relative to the previous Recommendation on mortgage
affordability tests (14/Q2/1), it clarified the rate to which the 3 percentage points stress should be applied and introduced a
de minimis threshold. The explanation for this Recommendation is set out in a chapter of this Report: ‘The FPC’s approach to
addressing risks from the UK mortgage market’.
16/Q2/1 Distribution of capital to meet ‘fair shares’ of systemic buffers Implemented
The FPC recommends to the PRA that it should seek to ensure that, where systemic buffers apply at different levels of
consolidation, there is sufficient capital within the consolidated group, and distributed appropriately across it, to address
both global systemic risks and domestic systemic risks.
This Recommendation was made at the FPC’s May 2016 meeting to agree the final systemic risk buffer (SRB) framework.
Following a consultation in 2016 H2, the PRA published on 1 February 2017 its Policy Statement on the implementation of
ringfencing.
(1)
This outlined how it would implement the FPC’s Recommendation. The FPC therefore decided at its meeting on
22 March 2017 to consider this Recommendation as implemented, given that the PRA’s Policy Statement was in place.
(1) www.bankofengland.co.uk/pra/Pages/publications/ps/2017/ps317.aspx.
15/Q2/3 CBEST vulnerability testing Implemented
The FPC recommends that the Bank, the PRA and the FCA work with firms at the core of the UK financial system to ensure
that they complete CBEST tests and adopt individual cyber resilience action plans. The Bank, the PRA and the FCA should
also establish arrangements for CBEST tests to become one component of regular cyber resilience assessment within the
UK financial system.
Since 2014, 31 out of 34 firms have finished testing, and two further firms are close to completion. The first round of CBEST
testing is therefore materially complete. Consistent with the FPC’s Recommendation, CBEST will become a regular component
of supervisory assessment of firms. Financial services firms and market infrastructures at the core of the UK financial system will
be expected to conduct their own regular tests of cyber resilience. They will also be subject to supervisor-led CBEST testing at
regular intervals. The frequency of these tests will be proportionate to firms’ importance for financial stability. CBEST is also
being adopted in other jurisdictions and sectors.
Based on the progress made, the FPC judged at its meeting on 21 June 2017 that its 2015 Recommendation has been
implemented. The summary of the review is set out in a box in this Report: ‘Building cyber resilience in the UK financial system’.
44 Financial Stability Report June 2017
Other FPC policy decisions which remain in place
Set out below are previous FPC decisions, which remain in force, on the setting of its policy tools. The calibration of these tools
is kept under review.
Countercyclical capital buffer (CCyB)
The FPC is increasing the UK CCyB rate to 0.5%, from 0%, with binding effect from 27 June 2018. Absent a material change in
the outlook, and consistent with its stated policy for a standard risk environment and of moving gradually, the FPC expects to
increase the rate to 1% at its November meeting, with binding effect a year after that. This rate is reviewed on a quarterly basis.
The United Kingdom has also previously reciprocated a number of foreign CCyB decisions — for more details see the Bank of
England website.
(1)
Under PRA rules, foreign CCyB rates applying from 2016 onwards will be automatically reciprocated up to
and including 2.5%.
Recommendation on loan to income ratios
In June 2014, the FPC made the following Recommendation (14/Q2/2):
The Prudential Regulation Authority (PRA) and the Financial Conduct Authority (FCA) should ensure that mortgage lenders
do not extend more than 15% of their total number of new residential mortgages at loan to income ratios at or greater
than 4.5. This Recommendation applies to all lenders which extend residential mortgage lending in excess of £100 million
per annum. The Recommendation should be implemented as soon as practicable.
The PRA and the FCA have published their respective approaches to implementing this Recommendation: the PRA has issued a
Policy Statement, including rules,
(2)
and the FCA has issued general guidance.
The FPC reviewed this Recommendation in November 2016 and decided not to amend the calibration. The explanation for this
is set out in the November 2016 Financial Stability Report.
(1) www.bankofengland.co.uk/financialstability/Pages/fpc/ccbrates.aspx.
(2) www.bankofengland.co.uk/pra/Pages/publications/ps/2014/ps914.aspx.
Other FPC activities since the previous Report
The Bank of England’s financial stability objective is to ‘protect and enhance the stability of the financial system of the
United Kingdom’. This is defined by Parliament, through the Bank of England Act 1998. The Court of Directors of the Bank has a
statutory responsibility to determine the Bank’s strategy in relation to this objective and to review it at least every three years.
Court has delegated the review of the strategy to the FPC, as permitted by the Act; but Court retains ultimate responsibility for
the strategy.
At its 2017 Q2 meeting, the FPC agreed the Bank’s financial stability strategy, following its review of the strategy that was set by
Court in 2014, and following consultation, as required by statute, with HM Treasury. The strategy will be published in the Bank’s
forthcoming Annual Report.
Annex 2 Core indicators 45
Annex 2: Core indicators
Table A.1 Core indicator set for the countercyclical capital buffer
(a)
Indicator Average, Average Minimum Maximum Previous Latest value
1987–2006
(b)
2006
(c)
since 1987
(b)
since 1987
(b)
value (oya) (as of 16 June 2017)
Non-bank balance sheet stretch
(d)
1 Credit to GDP
(e)
Ratio 117.1% 152.1% 86.9% 172.2% 140.6% 145.2% (2016 Q4)
Gap 6.4% 6.2% -25.6% 20.8% -22.5% -15.6% (2016 Q4)
2 Private non-financial sector credit growth
(f)
10.1% 9.8% -3.1% 22.8% 2.5% 4.5% (2016 Q4)
3 Net foreign asset position to GDP
(g)
-2.3% -10.3% -22.4% 24.2% -4.6% 24.2% (2016 Q4)
4 Gross external debt to GDP
(h)
183.0% 309.9% 114.2% 398.0% 295.0% 313.7% (2016 Q4)
of which bank debt to GDP 120.9% 195.0% 78.6% 267.6% 161.0% 177.2% (2016 Q4)
5 Current account balance to GDP
(i)
-1.7% -2.2% -6.0% 0.8% -5.5% -2.4% (2016 Q4)
Conditions and terms in markets
6 Long-term real interest rate
(j)
1.45% 1.23% -2.05% 2.18% -1.17% -1.54% (16 June 2017)
7 VIX
(k)
19.1 12.8 10.5 65.5 15.5 10.6 (16 June 2017)
8 Global corporate bond spreads
(l)
84 bps 84 bps 74 bps 482 bps 150 bps 113 bps (16 June 2017)
9 Spreads on new UK lending
Household
(m)
480 bps 352 bps 285 bps 850 bps 656 bps 636 bps (Apr. 2017)
Corporate
(n)
104 bps 97 bps 82 bps 392 bps 234 bps 225 bps (Dec. 2016)
Bank balance sheet stretch
(o)
10 Capital ratio
Basel II core Tier 1
(p)
6.6% 6.3% 6.1% 12.3% n.a. n.a.
Basel III common equity Tier 1
(q)
n.a. n.a. n.a. n.a. 12.3% 13.9% (2017 Q1)
11 Leverage ratio
(r)
Simple 4.7% 4.1% 2.9% 6.6% 6.6% 6.6% (2016 H2)
Basel III (2014 proposal) n.a. n.a. n.a. n.a. 4.9% 4.9% (2016 H2)
12 Average risk weights
(s)
53.6% 46.4% 33.4% 65.4% 36.5% 33.4% (2016 H2)
13 Return on assets before tax
(t)
1.0% 1.1% -0.2% 1.5% 0.4% 0.3% (2016 H2)
14 Loan to deposit ratio
(u)
114.5% 132.4% 93.3% 133.3% 97.1% 93.3% (2016 H2)
15 Short-term wholesale funding ratio
(v)
n.a. 24.6% 10.1% 26.7% 10.5% 10.1% (end-2016)
of which excluding repo funding n.a. 15.8% 4.5% 15.8% 4.5% 4.9% (end-2016)
16 Overseas exposures indicator: countries to
which UK banks have ‘large’ and ‘rapidly growing’ In 2006 Q4: AU, BR, CA, CH, CN, DE, In 2016 Q1: KY In 2017 Q1: CH, DE, JP,
total exposures
(w)(x)
ES, FR, IE, IN, JP, KR, KY, LU, NL, US, ZA KY, NL, TW
17 CDS premia
(y)
12 bps 8 bps 6 bps 298 bps 132 bps 58 bps (June 2017)
18 Bank equity measures
Price to book ratio
(z)
2.13 1.94 0.49 2.86 0.57 0.84 (June 2017)
Market-based leverage ratio
(aa)
9.7% 7.8% 1.9% 15.7% 3.5% 5.5% (June 2017)
46 Financial Stability Report June 2017
Table A.2 Core indicator set for sectoral capital requirements
(a)
Indicator Average, Average Minimum Maximum Previous Latest value
1987–2006
(b)
2006
(c)
since 1987
(b)
since 1987
(b)
value (oya) (as of 16 June 2017)
Bank balance sheet stretch
(o)
1 Capital ratio
Basel II core Tier 1
(p)
6.6% 6.3% 6.1% 12.3% n.a. n.a.
Basel III common equity Tier 1
(q)
n.a. n.a. n.a. n.a. 12.3% 13.9% (2017 Q1)
2 Leverage ratio
(r)
Simple 4.7% 4.1% 2.9% 6.6% 6.6% 6.6% (2016 H2)
Basel III (2014 proposal) n.a. n.a. n.a. n.a. 4.9% 4.9% (2016 H2)
3 Average mortgage risk weights
(ab)
n.a. n.a. 12.6% 22.4% 14.2% 12.6% (2016 H2)
UK average mortgage risk weights
(ac)
n.a. n.a. 10.5% 15.8% 11.0% 10.5% (2016 H2)
4 Balance sheet interconnectedness
(ad)
Intra-financial lending growth
(ae)
12.0% 13.0% -18.4% 45.5% -18.4% 5.2% (2016 H2)
Intra-financial borrowing growth
(af)
14.1% 13.7% -21.5% 33.1% -16.9% 33.1% (2016 H2)
Derivatives growth (notional)
(ag)
37.7% 34.2% -25.9% 52.0% -19.1% 12.1% (2016 H2)
5 Overseas exposures indicator: countries to which
UK banks have ‘large’ and ‘rapidly growing’ non-bank In 2006 Q4: AU, CA, DE, In 2016 Q1: KY In 2017 Q1: KY
private sector exposures
(ah)(x)
ES, FR, IE, IT, JP, KR, KY, NL, US, ZA
Non-bank balance sheet stretch
(d)
6 Credit growth
Household
(ai)
10.3% 11.2% -0.6% 19.6% 3.7% 4.5% (2016 Q4)
Commercial real estate
(aj)
15.3% 18.5% -9.7% 59.8% 0.0% 0.7% (2017 Q1)
7 Household debt to income ratio
(ak)
100.1% 141.8% 78.2% 150.5% 132.2% 135.0% (2016 Q4)
8 PNFC debt to profit ratio
(al)
237.0% 297.0% 157.0% 407.4% 272.8% 294.3% (2016 Q4)
9 NBFI debt to GDP ratio (excluding insurance
companies and pension funds)
(am)
56.4% 122.0% 14.0% 176.8% 126.3% 125.6% (2016 Q4)
Conditions and terms in markets
10 Real estate valuations
Residential price to rent ratio
(an)
100.0 151.1 66.9 160.6 139.6 142.9 (2017 Q1)
Commercial prime market yields
(ao)
5.4% 4.1% 3.8% 7.1% 4.1% 4.0% (2017 Q1)
Commercial secondary market yields
(ao)
8.5% 5.6% 5.1% 10.2% 5.8% 6.0% (2017 Q1)
11 Real estate lending terms
Residential mortgage LTV ratio
(mean above the median)
(ap)
90.6% 90.6% 81.6% 90.8% 86.7% 87.3% (2017 Q1)
Residential mortgage LTI ratio
(mean above the median)
(ap)
3.8 3.8 3.6 4.2 4.1 4.2 (2017 Q1)
Commercial real estate mortgage LTV
(average maximum)
(aq)
77.6% 78.3% 57.5% 79.6% 62.6% 57.5% (2016 H2)
12 Spreads on new UK lending
Residential mortgage
(ar)
80 bps 51 bps 35 bps 379 bps 176 bps 161 bps (Apr. 2017)
Commercial real estate
(as)
137 bps 135 bps 119 bps 422 bps 264 bps 254 bps (2016 Q4)
Annex 2 Core indicators 47
(a) A spreadsheet of the series shown in this table is available at www.bankofengland.co.uk/financialstability/Pages/fpc/coreindicators.aspx.
(b) If the series starts after 1987, the average between the start date and 2006 end and the maximum/minimum since the start date are used.
(c) 2006 was the last year before the start of the global financial crisis.
(d) The current vintage of ONS data is not available prior to 1997. Data prior to this and beginning in 1987 have been assumed to remain unchanged since The Blue Book 2013.
(e) Credit is defined as debt claims on the UK private non-financial sector. This includes all liabilities of the household and not-for-profit sector except for the unfunded pension liabilities and financial derivatives of the not-for-profit
sector, and private non-financial corporations’ (PNFCs’) loans and debt securities excluding direct investment loans and loans secured on dwellings. The credit to GDP gap is calculated as the percentage point difference between
the credit to GDP ratio and its long-term trend, where the trend is based on a one-sided Hodrick-Prescott filter with a smoothing parameter of 400,000. See Countercyclical Capital Buffer Guide at
www.bankofengland.co.uk/financialstability/Pages/fpc/coreindicators.aspx for further explanation of how this series is calculated. Sources: BBA, ONS Revell, J and Roe, A (1971); ‘National balance sheets and national accounting
— a progress report’, Economic Trends, No. 211 and Bank calculations.
(f) Twelve-month growth rate of nominal credit (defined as the four-quarter cumulative net flow of credit as a proportion of the stock of credit twelve months ago). Credit is defined as above. Sources: ONS and Bank calculations.
(g) As per cent of annual GDP (four-quarter moving sum). Sources: ONS and Bank calculations.
(h) Ratios computed using a four-quarter moving sum of GDP. Monetary financial institutions (MFIs) cover banks and building societies resident in the United Kingdom. Sources: ONS and Bank calculations.
(i) As per cent of quarterly GDP. Sources: ONS and Bank calculations.
(j) Five-year real interest rates five years forward, implied from inflation swaps and nominal fitted yields. Data series runs from October 2004. Sources: Bloomberg and Bank calculations.
(k) Measure of market expectations of 30-day volatility. Conveyed by S&P 500 stock index option prices (one-month moving average). Sources: Bloomberg and Bank calculations.
(l) Global corporate bond spreads’ refers to a one-month moving average of the global aggregate market non-financial corporate bond spread. This tracks the performance of investment-grade corporate debt publicly issued in the
global and regional markets from both developed and emerging market issuers. Index constituents are weighted based on market value. Spreads are option-adjusted (ie they show the number of basis points the
matched-maturity government spot curve needs to be shifted in order to match a bond’s present value of discounted cash flows). Prior to 2016, published versions of this indicator showed the BofA Merrill Lynch Global
Industrial Index. Sources: Barclays and Bank calculations.
(m) The household lending spread is a weighted average of mortgage and unsecured lending spreads, with weights based on relative volumes of new lending. The mortgage spread is a weighted average of quoted mortgage rates over
risk-free rates, using 90% LTV two-year fixed-rate mortgages and 75% LTV tracker, two and five-year fixed-rate mortgages. Spreads are taken relative to gilt yields of matching maturity for fixed-rate products. Spreads are taken
relative to Bank Rate for the tracker product. The unsecured component is a weighted average of spreads on credit cards, overdrafts and personal loans. Spreads on unsecured lending are taken relative to Bank Rate. FCA
Product Sales Data includes regulated mortgage contracts only but is used to weight all mortgage products. Series starts in 1997. Sources: Bank of England, Bloomberg, Council of Mortgage Lenders, FCA Product Sales Data and
Bank calculations.
(n) The UK corporate lending spread is a weighted average of: SME lending rates over Bank Rate; CRE average senior loan margins over Bank Rate; and, as a proxy for the rate at which banks lend to large, non-CRE corporates,
UK investment-grade company bond spreads over maturity-matched government bond yields (adjusted for any embedded option features such as convertibility into equity). Weights are based on relative amounts outstanding
of loans. Series starts in October 2002. Sources: Bank of America Merrill Lynch Global Research, Bank of England, Bloomberg, British Bankers’ Association, De Montfort University, Department for Business, Energy and Industrial
Strategy and Bank calculations.
(o) Unless otherwise stated, indicators are based on the major UK bank peer group defined as: Abbey National (until 2003); Alliance & Leicester (until 2007); Bank of Ireland (from 2005); Bank of Scotland (until 2000); Barclays;
Bradford & Bingley (from 2001 until 2007); Britannia (from 2005 until 2008); Co-operative Banking Group (from 2005); Halifax (until 2000); HBOS (from 2001 until 2008); HSBC (from 1992); Lloyds TSB/Lloyds Banking
Group; Midland (until 1991); National Australia Bank (from 2005); National Westminster (until 1999); Nationwide; Northern Rock (until 2011); Royal Bank of Scotland; Santander (from 2004); TSB (until 1994); Virgin Money
(from 2012) and Woolwich (from 1990 until 1997). Accounting changes, eg the introduction of IFRS in 2005 result in discontinuities in some series. Restated figures are used where available.
(p) Major UK banks’ aggregate core Tier 1 capital as a percentage of their aggregate risk-weighted assets. The core Tier 1 capital ratio series starts in 2000 and uses the major UK banks peer group as at 2014 and their constituent
predecessors. Data exclude Northern Rock/Virgin Money from 2008. From 2008, core Tier 1 ratios are as published by banks, excluding hybrid capital instruments and making deductions from capital based on PRA definitions.
Prior to 2008, that measure was not typically disclosed and Bank calculations approximating it as previously published in the Financial Stability Report are used. The series are annual until end-2012, half-yearly until end-2013
and quarterly afterwards. Sources: PRA regulatory returns, published accounts and Bank calculations.
(q) The Basel II series was discontinued with CRD IV implementation on 1 January 2014. The ‘Basel III common equity Tier 1 capital ratio’ is calculated as aggregate peer group common equity Tier 1 levels over aggregate
risk-weighted assets, according to the CRD IV definition as implemented in the United Kingdom. The Basel III peer group includes Barclays, Co-operative Banking Group, HSBC, Lloyds Banking Group, Nationwide, RBS and
Santander UK. Sources: PRA regulatory returns and Bank calculations.
(r) A simple leverage ratio calculated as aggregate peer group equity (shareholders’ claims) over aggregate peer group assets over aggregate Basel 2010 leverage ratio exposure. The Basel III (2014) series corresponds to aggregate
peer group CRD IV end-point Tier 1 capital over aggregate Basel 2014 exposure measure. Note that the simple series excludes Northern Rock/Virgin Money from 2008. The Basel III series consists of Barclays, Co-operative
Banking Group, HSBC, Lloyds Banking Group, Nationwide, RBS and Santander UK. The latest value uses latest published figures, in the case of Nationwide these relate to 2016 H1. The series are annual until end-2012 and
half-yearly afterwards. On 25 July 2016, the FPC recommended to the PRA that, when applying its rules on the leverage ratio, it considers allowing firms to exclude from the calculation of the total exposure measure those assets
constituting claims on central banks where they are matched by deposits accepted by the firm that are denominated in the same currency and of identical or longer maturity. No adjustment has been made to the calculated
leverage ratio reported here for this policy. Sources: PRA regulatory returns, published accounts and Bank calculations.
(s) Aggregate end-year peer group risk-weighted assets divided by aggregate end-year peer group published balance sheet assets. Data for 2014 H1 onwards are on a CRD IV basis. Series begins in 1992 and is annual until end-2012
and half-yearly afterwards. Latest published figures have been used, in the case of Nationwide, these relate to 2016 H1. Sources: Published accounts and Bank calculations.
(t) Calculated as major UK banks’ annual profit before tax as a proportion of total assets, averaged over the current and previous year. When banks in the sample have merged, aggregate profits for the year are approximated by
those of the acquiring group. Series is annual until 2015 when it becomes semi-annual. The latest value uses latest published figures, in the case of Nationwide these relate to 2016 H1. Sources: Published accounts and Bank
calculations.
(u) Major UK banks’ loans and advances to customers as a percentage of customer deposits, where customer refers to all non-bank borrowers and depositors. Repurchase agreements are excluded from loans and deposits where
disclosed. One weakness of the current measure is that it is not possible to distinguish between retail deposits from households and deposits placed by non-bank financial corporations on a consolidated basis. Additional data
collections would be required to improve the data in this area. The series begins in 2000 and is annual until end-2012 and half-yearly afterwards. The latest value uses latest published figures, in the case of Nationwide these
relate to 2016 H1. Sources: Published accounts and Bank calculations.
(v) Share of total funding (including capital) accounted for by wholesale funding with residual maturity of under three months. Wholesale funding comprises deposits by banks, debt securities, subordinated liabilities and repo.
Funding is proxied by total liabilities excluding derivatives and liabilities to customers under investment contracts. Where u
nderlying data are not published estimates have been used. Repo includes repurchase agreements and
securities lending. The series starts in 2005. In the latest value, Nationwide’s 2015 data are used as these are the latest published full-year results. Sources: Published accounts and Bank calculations.
(w) This indicator highlights the countries where UK-owned monetary financial institutions’ (MFIs’) overall exposures are greater than 10% of UK-owned MFIs’ tangible equity on an ultimate risk basis and have grown by more than
1.5 times nominal GDP growth in that country. Foreign exposures as defined in BIS consolidated banking statistics. Uses latest data available, with the exception of tangible equity figures for 2006–07, which are estimated using
published accounts. Sources: Bank of England, ECB, IMF World Economic Outlook (WEO), Thomson Reuters Datastream, published accounts and Bank calculations.
(x) Abbreviations used are: Australia (AU), Brazil (BR), Canada (CA), Switzerland (CH), People’s Republic of China (CN), Germany (DE), Spain (ES), France (FR), Ireland (IE), Italy (IT), India (IN), Japan (JP), Republic of Korea (KR),
Cayman Islands (KY), Luxembourg (LU), Netherlands (NL), Taiwan (TW), United States (US) and South Africa (ZA).
(y) Average of major UK banks’ five-year senior CDS premia, weighted by total assets until 2014 and by half-year total assets from 2015. Series starts in 2003. In the latest value Nationwide’s senior CDS is weighted by 2016 H1
total assets as the latest published figures relate to 2016 H1. Sources: Markit Group Limited, published accounts and Bank calculations.
(z) Relates the share price with the book, or accounting, value of shareholders’ equity per share. Averages of the ratios in the peer group, weighted by end-year total assets. The sample comprises the major UK banks and National
Australia Bank between 2005 and 2015 H2, excluding Britannia, Co-operative Banking Group, and Nationwide. Northern Rock/Virgin Money is excluded from 2008. Series starts in 2000. Sources: Thomson Reuters Datastream,
published accounts and Bank calculations.
(aa) Total peer group market capitalisation divided by total peer group assets (note a discontinuity due to introduction of IFRS accounting standards in 2005, which tends to reduce leverage ratios thereafter). The sample comprises
the major UK banks, excluding Britannia, Co-operative Banking Group and Nationwide. National Australia Bank is included between 2005 and 2015 H2. Northern Rock/Virgin Money is excluded from 2008. Series starts in 2000.
Sources: Thomson Reuters Datastream, published accounts and Bank calculations.
(ab) Sample consists of Barclays Group, Co-operative Banking Group, HSBC Holdings Group, Lloyds Banking Group, Nationwide Building Society Group, RBS Group, Santander UK Group and excludes Nationwide for 2008 H2 only.
Average risk weights for residential mortgages (exposures on the Retail IRB method only) are calculated as total risk-weighted assets divided by total exposure value for all banks in the sample. Calculated on a consolidated basis,
except for Nationwide for 2014 H2/2015 H1 where only solo data were available. Series starts in 2009 and is updated half-yearly. Sources: PRA regulatory returns and Bank calculations.
(ac) Sample consists of Bank of Scotland, Barclays Bank, HSBC Bank, Lloyds Bank, National Westminster Bank, Nationwide, Santander UK, Co-operative Bank, Royal Bank of Scotland, Ulster Bank and excludes Nationwide for
2008 H2 only. Average risk weights for residential mortgages (exposures on the Retail IRB method only) are calculated as total risk-weighted assets divided by total exposure value for all banks in the sample. Calculated on an
unconsolidated basis, Royal Bank of Scotland data includes National Westminster, Ulster Bank and RBS. Historical data updated as of June 2016 to improve data series consistency. Series starts in 2009 and is updated
half-yearly. Sources: PRA regulatory returns and Bank calculations.
(ad) The disclosures the series are based on are not currently sufficient to ensure that all intra-financial activity is included in these series, nor is it possible to be certain that no real-economy activity is included. Additional data
collections would be required to improve the data in this area. The intra-financial lending and borrowing growth series are adjusted for the acquisitions of Midland by HSBC in 1992, and of ABN AMRO by RBS in 2007 to avoid
reporting large growth rates resulting from step changes in the size and interconnectedness of the major UK bank peer group. Series exclude National Australia Bank.
(ae) Lending to other banks and other financial corporations. Growth rates are year on year. Latest value shows growth rate for year to 2016 H2. Data point excludes National Australia Bank. Sources: Published accounts and
Bank calculations.
(af) Wholesale borrowing, composed of deposits from banks and non-subordinated securities in issue. Growth rates are year-on-year. Latest value shows growth rate for year to 2016 H2. Data point excludes National Australia
Bank. One weakness of the current measure is that it is not possible to distinguish between retail deposits and deposits placed by non-bank financial institutions on a consolidated basis. Sources: Published accounts and Bank
calculations.
(ag) Based on notional value of derivatives (some of which may support real economy activity). The sample includes Barclays, HSBC and RBS who account for a significant share of UK banks’ holdings of derivatives, though the sample
could be adjusted in the future should market shares change. Series starts in 2002. Growth rates are year on year. Latest value shows growth rate for year to 2016 H2. Sources: Published accounts and Bank calculations.
(ah) This indicator highlights the countries where UK-owned MFIs’ non-bank private sector exposures are greater than 10% of UK-owned MFIs’ tangible equity on an ultimate risk basis and have grown by more than 1.5 times nominal
GDP growth in that country. Foreign exposures as defined in BIS consolidated banking statistics. Overseas sectoral exposures cannot currently be broken down further at the non-bank private sector level. The intention is to
divide them into households and corporates as new data become available. Uses latest data available, with the exception of tangible equity figures for 2006–07, which are estimated using published accounts. Sources:
Bank of England, ECB, IMF World Economic Outlook (WEO), Thomson Reuters Datastream, published accounts and Bank calculations.
(ai) The twelve month growth rate of nominal credit. Defined as the four quarter cumulative net flow of credit divided by the stock of credit twelve months ago. Credit is defined as all liabilities of the household and not-for-profit
sector except for the unfunded pension liabilities and financial derivatives of the not-for-profit sector. Sources: ONS and Bank calculations.
(aj) Four-quarter growth rate of UK-resident MFIs’ loans to the real estate sector. The real estate sector is defined as: buying, selling and renting of own or leased real estate; real estate and related activities on a fee or contract
basis; and development of buildings. Non seasonally adjusted. Quarterly data. Data cover lending in both sterling and foreign currency from 1998 Q4. Prior to this period, data cover sterling only. Source: Bank of England.
(ak) Gross debt as a percentage of a four-quarter moving sum of gross disposable income of the UK household and non-profit sector. Includes all liabilities of the household sector except for the unfunded pension liabilities and
financial derivatives of the non-profit sector. Disposable income is adjusted for financial intermediation services indirectly measured (FISIM) and changes in pension entitlements. Sources: ONS and Bank calculations.
(al) Gross debt as a percentage of a four-quarter moving sum of gross operating surplus. Gross debt is measured as loans and debt securities excluding derivatives, direct investment loans and loans secured on dwellings. The
corporate gross operating surplus series is adjusted for FISIM. Sources: ONS and Bank calculations.
(am) Gross debt as a percentage of four-quarter moving sum of nominal GDP. The NBFI sector includes all financial corporations apart from monetary financial institutions (ie deposit taking institutions). This indicator additionally
excludes insurance companies and pension funds. Sources: ONS and Bank calculations.
(an) Ratio between an average of the seasonally adjusted Halifax and Nationwide house price indices and RPI housing rent. The series is rebased so that the average between 1987 and 2006 is 100. Sources: Halifax/Markit,
Nationwide, ONS and Bank calculations.
(ao) The prime (secondary) yield is the ratio between the weighted averages, across the lowest (highest) yielding quartile of commercial properties, of MSCI Inc.’s measures of rental income and capital values. Sources: MSCI Inc. and
Bank calculations.
(ap) Mean LTV (respectively LTI) ratio on new advances above the median LTV (LTI) ratio, based on loans to first-time buyers, council/registered social tenants exercising their right to buy and homemovers, and excluding lifetime
mortgages and advances with LTV above 130% (LTI above 10x). FCA Product Sales Data includes regulated mortgage contracts only. Series starts in 2005. Sources: FCA Product Sales Data and Bank calculations.
(aq) Average of the maximum offered loan to value ratios across major CRE lenders. Series starts in 2002. Sources: De Montfort University and Bank calculations.
(ar) The residential mortgage lending spread is a weighted average of quoted mortgage rates over risk-free rates, using 90% LTV two-year fixed-rate mortgages and 75% LTV tracker, two and five-year fixed-rate mortgages. Spreads
are taken relative to gilt yields of matching maturity for fixed-rate products. Spreads are taken relative to Bank Rate for the tracker product. Weights based on relative volumes of new lending. Series starts in 1997. FCA Product
Sales Data includes regulated mortgage contracts only. Sources: Bank of England, Bloomberg, Council of Mortgage Lenders, FCA Product Sales Data and Bank calculations.
(as) The CRE lending spread is the average of senior loan margins across major CRE lenders relative to Bank Rate. Series starts in 2002. Sources: Bank of England, Bloomberg, De Montfort University and Bank calculations.
48 Financial Stability Report June 2017
Table A.3 Core indicator set for LTV and DTI limits
(a)
Indicator Average, Average Minimum Maximum Previous Latest value
1987–2006
(b)
2006
(c)
since 1987
(b)
since 1987
(b)
value (oya) (as of 16 June 2017)
Lender and household balance sheet stretch
1 LTI and LTV ratios on new residential mortgages
Owner-occupier mortgage LTV ratio 90.6% 90.6% 81.6% 90.8% 86.7% 87.3% (2017 Q1)
(mean above the median)
(d)
Owner-occupier mortgage LTI ratio 3.8 3.8 3.6 4.2 4.1 4.2 (2017 Q1)
(mean above the median)
(d)
Buy-to-let mortgage LTV ratio (mean)
(e)
n.a. n.a. 63.8% 75.4% 65.9% 63.8% (2016 Q4)
2 Household credit growth
(f)
10.3% 11.2% -0.6% 19.6% 3.7% 4.5% (2016 Q4)
3 Household debt to income ratio
(g)
100.1% 141.8% 78.2% 150.5% 132.2% 135.0% (2016 Q4)
of which: mortgages
(h)
70.8% 103.8% 50.7% 113.2% 101.0% 101.0% (2016 Q4)
of which: owner-occupier mortgages
(i)
80.6% 95.0% 67.2% 100.0% 84.3% 83.6% (2016 Q4)
Conditions and terms in markets
4 Approvals of loans secured on dwellings
(j)
97,922 119,045 26,702 134,710 66,182 64,645 (Apr. 2017)
5 Housing transactions
(k)
129,508 139,039 51,660 221,978 83,050 99,910 (Apr. 2017)
Advances to homemovers
(l)
48,985 59,342 14,300 93,500 22,000 25,700 (Apr. 2017)
% interest only
(m)
53.3% 31.0% 1.8% 81.3% 1.8% 2.3% (Apr. 2017)
Advances to first-time buyers
(l)
39,179 33,567 8,500 55,800 24,800 25,400 (Apr. 2017)
% interest only
(m)
52.1% 24.0% 0.0% 87.9% 0.0% 0.0% (Apr. 2017)
Advances to buy-to-let purchasers
(l)
10,128 14,113 3,600 29,100 4,200 5,300 (Apr. 2017)
% interest only
(n)
n.a. n.a. 50.0% 74.3% 74.3% 72.4% (2017 Q1)
6 House price growth
(o)
1.8% 2.2% -5.6% 7.0% 1.4% -0.2% (May 2017)
7 House price to household disposable income ratio
(p)
3.0 4.6 2.2 4.8 4.4 4.6 (2016 Q4)
8 Rental yield
(q)
5.8% 5.1% 4.8% 7.6% 5.0% 4.8% (Apr. 2017)
9 Spreads on new residential mortgage lending
All residential mortgages
(r)
80 bps 51 bps 35 bps 379 bps 176 bps 161 bps (Apr. 2017)
Difference between the spread on high and 18 bps 25 bps 1 bps 293 bps 84 bps 101 bps (May 2017)
low LTV residential mortgage lending
(r)
Buy-to-let mortgages
(s)
n.a. n.a. 61 bps 397 bps 259 bps 253 bps (2017 Q1)
(a) A spreadsheet of the series shown in this table is available at www.bankofengland.co.uk/financialstability/Pages/fpc/coreindicators.aspx.
(b) If the series start after 1987, the average between the start date and 2006 end and the maximum/minimum since the start date are used.
(c) 2006 was the last year before the global financial crisis.
(d) Mean LTV (respectively LTI) ratio on new advances above the median LTV (LTI) ratio, based on loans to first-time buyers, council/registered social tenants exercising their right to buy and homemovers, and excluding lifetime
mortgages and advances with LTV ratio above 130% (LTI above 10x). FCA Product Sales Data includes regulated mortgage contracts only. Series starts in 2005. Sources: FCA Product Sales Data and Bank calculations.
(e) Estimated mean LTV ratio of new non-regulated lending advances, of which buy-to-let is 88% by value. The figures include further advances and remortgages. The raw data is categorical: the share of mortgages with LTV ratio
less than 75%; between 75% and 90%; between 90% and 95%; and greater than 95%. An approximate mean is calculated by giving these categories weights using the average LTV in equivalent buckets in loan level
buy-to-let data gathered by the Council of Mortgage Lenders. Series starts in 2007. Council of Mortgage Lenders data available from 2014; weights prior to this date are average LTVs across the respective buckets using all data
gathered in 2014. The share of mortgages with LTV ratio at 75% from 2014 onwards used are adjusted to estimate the LTV of each loan before any fees or charges are added. This approximates the LTV at which the loan was
originated. Source: Bank of England, Council of Mortgage Lenders and Bank calculations.
(f) The twelve month growth rate of nominal credit. Defined as the four-quarter cumulative net flow of credit divided by the stock of credit twelve months ago. Credit is defined as all liabilities of the household and not-for-profit
sector except for the unfunded pension liabilities and financial derivatives of the not-for-profit sector. Sources: ONS and Bank calculations.
(g) Gross debt as a percentage of a four-quarter moving sum of gross disposable income of the UK household and non-profit sector. Includes all liabilities of the household sector except for the unfunded pension liabilities and
financial derivatives of the non-profit sector. Disposable income is adjusted for financial intermediation services indirectly measured (FISIM) and changes in pension entitlements. Sources: ONS and Bank calculations.
(h) Total debt secured on dwellings as a percentage of a four-quarter moving sum of gross disposable income of the UK household and non-profit sector. Disposable income is adjusted for FISIM and changes in pension entitlements.
Sources: ONS and Bank calculations.
(i) Total debt associated with owner-occupier mortgages divided by the four-quarter moving sum of gross disposable income of the UK household and non-profit sector. Disposable income is adjusted for FISIM and changes in
pension entitlements. Owner-occupier mortgage debt estimated by multiplying aggregate household debt secured on dwellings by the share of mortgages on lender balances that are not buy-to-let loans. Series starts in 1999.
Sources: Council of Mortgage Lenders, ONS and Bank calculations.
(j) Data are for monthly number of house purchase approvals covering sterling lending by UK MFIs and other lenders to UK individuals. Approvals secured on dwellings are measured net of cancellations. Seasonally adjusted.
Series starts in 1993. Source: Bank of England.
(k) The number of houses sold/bought in the current month is sourced from HMRC’s Land Transaction Return. From 2008 the Return excluded properties priced at less than £40,000 (2006 and 2007 data have also been revised by
HMRC to correct for this). Data prior to 2005 comes from the Survey of Property Transactions; the UK total figure is computed by assuming that transactions in the rest of the United Kingdom grew in line with England, Wales
and Northern Ireland. Seasonally adjusted. Sources: Council of Mortgage Lenders, HMRC and Bank calculations.
(l) The number of new mortgages advanced for house purchase in the current month. Buy-to-let series starts in 2001. There are structural breaks in the series in April 2005 where the Council of Mortgage Lenders switches source.
Data prior to 2002 are at a quarterly frequency. Sources: Council of Mortgage Lenders and Bank calculations.
(m)The share of new owner-occupied mortgages advanced for house purchase that are interest only. Interest-only mortgages exclude mixed capital and interest mortgages. There are structural breaks in the series in April 2005
where the Council of Mortgage Lenders switches source. Data prior to 2002 are at a quarterly frequency. Sources: Council of Mortgage Lenders and Bank calculations.
(n) The share of non-regulated mortgages that are interest only. The data include all mortgages, not just those for house purchase. Interest-only mortgages exclude mixed capital and interest mortgages. Sources: Bank of England
and Bank calculations.
(o) House prices are calculated as the mean of the average UK house price as reported in the Halifax and Nationwide house price indices. Growth rate calculated as the percentage change three months on three months earlier.
Series starts in 1991. Seasonally adjusted. Sources: Halifax/Markit, Nationwide and Bank calculations.
(p) The ratio is calculated using a four-quarter moving sum of gross disposable income of the UK household and non-profit sector per household as the denominator. Disposable income is adjusted for FISIM and changes in pension
entitlements. Historical UK household population estimated using annual GB data assuming linear growth in the Northern Ireland household population between available data points. Series starts in 1990. Sources:
Department for Communities and Local Government, Halifax/Markit, Nationwide, ONS and Bank calculations.
(q) Using Association of Residential Letting Agents (ARLA) data up until 2014. From 2015 onwards, the series uses LSL Property Services plc data normalised to the ARLA data over 2008 to 2014, when both series are available.
Series starts in 2001. Sources: Association of Residential Letting Agents, LSL Property Services plc and Bank calculations.
(r) The overall spread on residential mortgage lending is a weighted average of quoted mortgage rates over risk-free rates, using 90% LTV two-year fixed-rate mortgages and 75% LTV tracker, two and five-year fixed-rate mortgages.
Spreads are taken relative to gilt yields of matching maturity for fixed-rate products. Spreads are taken relative to Bank Rate for the tracker product. Weights are based on relative volumes of new lending. The difference in
spread between high and low LTV lending is the rate on 90% LTV two-year fixed-rate mortgages less the 75% LTV two-year fixed-rate. Series starts in 1997. FCA Product Sales Data includes regulated mortgage contracts only.
Sources: Bank of England, Bloomberg, Council of Mortgage Lenders, FCA Product Sales Data and Bank calculations.
(s) The spread on new buy-to-let mortgages is the weighted average effective spread charged on new floating and fixed-rate non-regulated mortgages over safe rates. Spreads are taken relative to Bank Rate for the floating-rate
products. The safe rate for fixed-rate mortgages is calculated by weighting two-year, three-year and five-year gilts by the number of buy-to-let fixed-rate mortgage products offered at these maturities. Series starts in 2007.
Sources: Bank of England, Bloomberg, Moneyfacts and Bank calculations.
Index of charts and tables 49
Index of charts and tables
Charts
Executive summary i
A Major UK banks’ capital ratios ii
B Annual growth rates of consumer credit products and
household income ii
C China non-financial sector debt and growth of total
social financing iv
D International ten-year real government bond yields iv
E Commercial real estate prices in the United Kingdom
and ranges of sustainable valuations iv
Box 1
A Growth in credit to households and firms compared
with nominal GDP growth vi
Box 2
A Net inward financing flows ix
B UK CRE transactions (gross quarterly flows) ix
Part A
The FPC’s approach to addressing risks from the
UK mortgage market 1
A.1 UK house price to household income ratio 2
A.2 Completions of new dwellings in the United Kingdom 2
A.3 UK household debt to income ratio 3
A.4 Household debt to income ratio and consumption
growth over 2007–12 3
A.5 Change in consumption relative to income among
mortgagors with different LTI ratios between 2007
and 2009 3
A.6 Households in two-month arrears by mortgage DSR 4
A.7 Percentage of households with mortgage debt-servicing
ratios of 40% or greater 4
A.8 Cumulative five-year loss rates on UK mortgages in past
downturns and in stress tests 5
A.9 New mortgage lending by LTV at origination 5
A.10 UK mortgage books by indexed LTV 5
A.11 Mortgage rates on owner-occupier and buy-to-let
lending relative to risk-free rates 6
A.12 Proportion of new mortgages with no fees 6
Fig. A.1 Feedback loops between mortgage credit and house
prices can amplify a downturn 6
A.13 Composition of the outstanding mortgage stock 7
A.14 Relationship between the affordability test and the
LTI flow limit in constraining lending 8
A.15 Flow of new mortgages by LTI 9
A.16 Share of new mortgages by mortgage term 9
A.17 LTI distribution of new mortgage lending 9
A.18 Risk weights on UK ‘prime’ mortgages by LTV 11
Box 4
A Mortgages’ LTV ratio and house price to income ratio 12
UK consumer credit 14
A.19 Annual growth rates of consumer credit products
and household income 14
A.20 Spread between effective interest rates on new
personal loans and Bank Rate 15
A.21 Interest-free periods of credit card balance transfer
offers 15
A.22 Major UK banks’ average risk weights on consumer
credit exposures 16
A.23 End-2016 mortgage and personal loan portfolios of
major UK banks, by year of issuance 16
A.24 UK banks’ sterling write-offs on lending to individuals 17
A.25 Historical relationship between changes in
unemployment and write-offs on non-credit card
consumer credit exposures 17
Box 6
A Composition of the stock of consumer credit,
end-March 2017 18
B Value of annual dealership car finance for new car
purchases, and proportion of private new car
purchases funded with dealership car finance 19
Global environment 20
A.26 International annual GDP growth projections 20
A.27 China non-financial sector debt and growth of total
social financing 21
A.28 Domestic value added in exports to China 21
A.29 UK-owned banking groups’ consolidated exposures to
selected countries and regions 22
A.30 Ratio of non-performing loans and advances to total
loans and advances (2016 Q4) 22
Asset valuations 23
A.31 International ten-year real government bond yields 23
A.32 Dispersion in implied volatilities in foreign exchange,
interest rate and equity markets 24
A.33 High-yield corporate bond spreads 24
A.34 Changes in nominal ten-year interest rates since the
November Report 25
A.35 Commercial real estate prices in the
United Kingdom and ranges of sustainable valuations 25
A.36 Estimated losses in global corporate bond markets
following a 100 basis point increase in interest rates 26
A.37 UK CRE debt reported to De Montfort University
survey 26
Part B
Banking sector resilience 27
B.1 Major UK banks’ capital ratios 27
B.2 Major UK banks’ leverage ratios 28
B.3 Average interest rates on new lending and Bank Rate 30
B.4 UK banks’ share prices and FTSE All-Share index since
1 January 2016 30
B.5 UK banks’ statutory and underlying return on equity 31
B.6 UK banks’ average price to book ratio 31
B.7 Price to book ratios for major global banks compared
with expected returns on equity 31
Market-based finance 34
B.8 Net finance raised by UK private non-financial
corporations (PNFCs) 34
B.9 Dealers’ leverage ratios 35
B.10 Proportion of cash balance placed in repo of a
major European asset manager, split by counterparty 36
50 Financial Stability Report June 2017
B.11 Gilt repo rates paid by a group of pension fund
asset managers in excess of expectations of policy
interest rates 36
B.12 UK, French and German repo rates 36
B.13 Stock of property-related non-linked illiquid
assets during 2016 37
Tables
Part A
The FPC’s approach to addressing risks
from the UK mortgage market 1
A.1 The Bank has an extensive toolkit to address risks
from the UK mortgage market 7
A.2 Cuts in consumption between 2007 and 2009 among
mortgagors with different LTI ratios 10
Part B
Banking sector resilience 27
B.1 Selected measures of UK banks’ funding costs 30
Box 7
1 The FPC’s Recommendations on cyber resilience 32
Market-based finance 34
Box 8
1 Service characteristics of the renewed RTGS service 40
Annex 2: Core indicators 45
A.1 Core indicator set for the countercyclical capital buffer 45
A.2 Core indicator set for sectoral capital requirements 46
A.3 Core indicator set for LTV and DTI limits 48
Glossary and other information 51
Glossary and other information
Glossary of selected data and instruments
CDS – credit default swap.
CPI – consumer prices index.
GDP – gross domestic product.
HICP – harmonised index of consumer prices.
LCF – Living Costs and Food Survey.
RPI – retail prices index.
SVR – standard variable rate.
Abbreviations
ARF – Authorities’ Response Framework.
AT1 – additional Tier 1.
BHPS – British Household Panel Survey.
BIS – Bank for International Settlements.
CBEST – UK Government’s National Cyber Security
Programme.
CBPS – Corporate Bond Purchase Scheme.
CCyB – countercyclical capital buffer.
CCP – central counterparty.
CEIC – CEIC Data Company Ltd.
CET1 – common equity Tier 1.
CGFS – Committee on the Global Financial System.
CIS – collective insurance schemes.
CML – Council of Mortgage Lenders.
CRD IV – Capital Requirements Directive.
CRE – commercial real estate.
DSR – debt-servicing ratio.
DTI – debt to income.
ECB – European Central Bank.
EEA – European Economic Area.
EIOPA – European Insurance Occupational Pensions Authority.
EME – emerging market economy.
EU – European Union.
FCA – Financial Conduct Authority.
FDI – foreign direct investment.
FISIM – financial intermediation services indirectly measured.
FMI – financial market infrastructure.
FPC – Financial Policy Committee.
FSA – Financial Services Authority.
FSB – Financial Stability Board.
FTSE – Financial Times Stock Exchange.
G7 – Canada, France, Germany, Italy, Japan, the
United Kingdom and the United States.
G20 – The Group of Twenty Finance Ministers and Central
Bank Governors.
G-SIB – global systemically important bank.
HMRC – Her Majesty’s Revenue and Customs.
HVPS – High-Value Payment System.
ICR – interest coverage ratio.
IFRS – International Financial Reporting Standard.
IIF – Institute of International Finance.
IMF – International Monetary Fund.
IOSCO – International Organization of Securities
Commissions.
IRB – internal ratings based.
LTI – loan to income.
LTV – loan to value.
MCOB – Mortgages and Home Finance: Conduct of Business
sourcebook.
MFI – monetary financial institution.
MPC – Monetary Policy Committee.
MREL – minimum requirement for own funds and eligible
liabilities.
MSCI – Morgan Stanley Capital International Inc.
NBFI – non-bank financial institution.
NCSC – National Cyber Security Centre.
NSFR – Net Stable Funding Ratio.
OECD – Organisation for Economic Co-operation and
Development.
ONS – Office for National Statistics.
OTC – over the counter.
PCP – personal contract purchase.
PNFC – private non-financial corporation.
PPI – payment protection insurance.
PRA – Prudential Regulation Authority.
PSD – Product Sales Database.
RBS – Royal Bank of Scotland.
RFB – ring-fenced bank.
RICS – Royal Institution of Chartered Surveyors.
RoE – return on equity.
RTGS – real-time gross settlement.
SME – small and medium-sized enterprise.
SMF – Sterling Monetary Framework.
SMMT – Society of Motor Manufacturers and Traders.
SRB – systemic risk buffer.
S&P – Standard & Poor’s.
TFS – Term Funding Scheme.
WEO – IMF World Economic Outlook.