LESSONS FROM THE NORTHERN ROCK EPISODE

1
LESSONS FROM THE NORTHERN ROCK EPISODE

• David G Mayes and Geoffrey Wood
• Auckland University Cass Business
  School
• University of Buckingham

2
MOTIVATION

• First bank run in UK since 1866 Overend Gurney
• How could it happen?
• Strategy and risk well-known
• Northern Rock thought solvent
• UK had Lender of Last Resort/Emergency Liquidity
  Arrangements (home of Bagehot)
• UK had extensive deposit insurance
• Experienced central bank, unified supervisor,
  tripartite arrangements
• Was theory or practice at fault?
• Crises usually caused by something new
• In this case drying up of wholesale markets with
  US sub-prime problems

3
ISSUES

• Problems with LOLR/ELA
• Why did deposit insurance not stop a run
• Diamond-Dybvig
• Why was there not more action earlier?
• Keeping a failed institution open
• Temporary nationalisation a solution not a
  problem
• Problems of co-ordination
• Certainty

4
CERTAINTY

• People will rush for safety if they do not know
  what may happen.
• Future is uncertain cannot prescribe specific
  solutions
• 4 general sources of certainty
• Unlimited ELA against acceptable collateral
• No taxpayer bailouts closure or takeover by the
  authorities to keep critical functions
  operating/minimise the loss
• Insured depositors have no material break in
  access to funds
• Regulatory authority forced to act early
  accelerating scale of severity

5
CERTAINTY

• 4 general sources of certainty
• Unlimited ELA against acceptable collateral
• No taxpayer bailouts closure or takeover by the
  authorities to keep critical functions
  operating/minimise the loss
• Insured depositors have no material break in
  access to funds
• Regulatory authority forced to act early
  accelerating scale of severity
• None of these applied ELA late cautious, viewed
  as penalty government stepped in major break in
  access to deposits and share of loss, lack of
  action (no PCA)

6
EMERGENCY LIQUIDITY ASSISTANCE

• ELA has worked in the sense that there has been
  adequate collateral for the Bank of England to
  provide all the necessary liquidity to meet the
  withdrawal of retail deposits
• The government should not lose money from
  nationalisation unless there is a problem with
  compensating shareholders not a problem with
  ELA
• However it did not prevent the run
• Two main drawbacks highlighted
• Transparency could not put problem right
  quietly
• Stigma use of ELA interpreted as disaster not
  salvation

7
TRANSPARENCY

• Ideally problem and solution are revealed
  simultaneously.
• Is it possible to hold the discussions in secret?
• Is it permissible? Duty to inform shareholders?
• Are other switches in funding disclosed?
• Is effect on share price different from
  takeovers?
• Price of NR fell by 50 between Feb07 and Sept
  fell by further third after news broke

8
(No Transcript)
9
TRANSPARENCY

• Ideally problem and solution are revealed
  simultaneously.
• Is it possible to hold the discussions in secret?
• Is it permissible? Duty to inform shareholders?
• Are other switches in funding disclosed?
• Is effect on share price different from
  takeovers?
• Price of NR fell by 50 between Feb07 and Sept
  fell by further third after news broke
• Creditors normally protect themselves if a
  company looks in trouble banks different
  because funds are an essential input cannot
  reschedule safety net especially deposit
  insurance should reassure
• Opacity only justifiable if it is in interests of
  creditors and depositors and all know there will
  be disclosure afterwards
• Key is ability to close bank early before there
  are losses to creditors if recapitalisation fails

10
STIGMA

• A dilemma
• Should access to ELA be sufficiently frequent
  that it is not regarded as a signal of disaster?
• If ELA is common then it may increase risk taking
  moral hazard
• Needs to be at an above market rate to enable the
  market to work first should this be described
  as a penalty rate risk premia are common
• No progressive structure either using ELA or
  not
• Difficult to match by general lending to the
  market
• Is the problem unusual due to banks offloading
  the risk from normal borrower-lender
  relationship? Safety net not so well designed for
  market risk as idiosyncratic risk?

11
DEPOSIT INSURANCE

• Issues
• Does coinsurance help improve incentives?
• Must all deposits be insured?
• How long can access to funds be interrupted?
• Does prefunding matter?

12
DEPOSIT INSURANCE

• Issues
• Does coinsurance help improve incentives?
• UK unusual in having coinsurance after first
  2000 10 of next 33000 others zero or 100 to
  limit
• NZ experience suggests people do not check
• NR shows exposure leads to a run
• Theory OK but does not work if there is threat of
  loss
• NR has killed coinsurance
• Must all deposits be insured?
• How long can access to funds be interrupted?
• Does prefunding matter?

13
DEPOSIT INSURANCE

• Issues
• Does coinsurance help improve incentives?
• Must all deposits be insured?
• Current UK limit gives 100 coverage to over 90
  of depositors. Are remainder well informed? Do
  they exit early before run? US experience
  suggests no. Can they afford the loss (not widows
  and orphans)
• Raise to US levels? Already high by European
  standards regime competition within single
  market?
• NR shows that having no insurance is likely to
  result in a blanket guarantee also Nordic
  crises
• How long can access to funds be interrupted?
• Does prefunding matter?

14
DEPOSIT INSURANCE

• Issues
• Does coinsurance help improve incentives?
• Must all deposits be insured?
• How long can access to funds be interrupted?
• If there is any material break people will have
  problems but Irish Bank strike example
• Keep core functions operating NZ approach
• US tries to get full access within a week EU 3
  months permitted renewable twice
• In NR people shifted banks rather than into cash
  access is crucial
• Implementation difficult need considerable
  prior knowledge of banks computer systems. Bank
  systems must identify insured deposits on an
  ongoing basis (problem of aggregation) Must avoid
  netting.
• Does prefunding matter?

15
DEPOSIT INSURANCE

• Issues
• Does coinsurance help improve incentives?
• Must all deposits be insured?
• How long can access to funds be interrupted?
• Implementation difficult need considerable
  prior knowledge of banks computer systems. Bank
  systems must identify insured deposits on an
  ongoing basis (problem of aggregation) Must avoid
  netting.
• Can transfer all deposits to another bank, can
  take over bank and keep deposits in good bank,
  bridge bank in US
• Does prefunding matter?

16
DEPOSIT INSURANCE

• Issues
• Does coinsurance help improve incentives?
• Must all deposits be insured?
• How long can access to funds be interrupted?
• Does prefunding matter?
• Without prefunding cannot payout in a hurry or
  offer other quick solutions
• Funding increases credibility of the fund
• If problems occur in a downturn rest of banking
  system not in good shape to provide funds in a
  hurry.
• Cost of prefunding small particularly if pay
  interest after cost recovery

17
GETTING EARLIER ACTION

• Prompt Corrective Action
• In US a required series of increasingly harsh
  interventions within a timetable as capital falls
  ending in closure before capital is exhausted.
• However NR shows that intervention must occur
  before capital falls risk taking and liquidity
  problems. Need Pillar 2 and Pillar 3 to require
  action
• NR share price falling FSA clear that reliance
  on wholesale market a risk.
• How is this turned into clear rules? PCA limits
  discretion
• NR shows that authorities hold off in the hope of
  achieving a less difficult solution

18
GETTING EARLIER ACTION

• Prompt Corrective Action
• Prompt action requires intervention by the
  private sector well before regulatory limits are
  breached
• Midland and National Westminster are UK examples
• Can only be encouraged not compelled main
  encouragement is that failure to act will be
  unattractive
• Action on risk needs to be based on quantitative
  assessment NR shows standard approaches can be
  underestimates

19
GETTING EARLIER ACTION

• Prompt Corrective Action
• The authorities must have the power to step into
  a troubled institution should it get too close to
  failure
• They have to be capable of forming a rapid
  judgement about the extent of the losses and the
  sensible action
• They have to be capable of acting fast enough to
  be able to assign the losses and keep the bank
  operating without a material break
• Furthermore if the bank does close the
  authorities have to be able to act fast enough to
  ensure that insured depositors have access to
  their funds without any significant break.
• All these must be believed to apply in advance

20
KEEPING CORE FUNCTIONS OPERATING

• In general banks that cannot perform to the rules
  should be closed and closed early before there
  are losses
• However, some banks are too important to be
  allowed to close because of the disruption it
  will cause to financial markets and the wider
  economy
• Not simply too big to fail but too important in
  certain functions market maker for example
• NR shows this applies to smaller institutions
  than previously thought
• US have good way of achieving this through ending
  legal personality and reopening in whole or
  essential parts as a bridge bank under temporary
  government control without a break time limit
  laid down
• UK does not have legal power but recommended by
  BoE/FSA/HMT paper avoids costly litigation and
  taxpayer being exposed to full extent of losses

21
COORDINATION FAILURE

• Government by committee
• Cross-border arrangements

22
COORDINATION FAILURE

• Government by committee
• 3 players in the UK BoE, FSA, HMT
• In US FDIC has clear lead role UK deposit
  insurer FSCS has none others only involved if
  Systemic Risk Exemption invoked never used
• NR showed some problems with each being reluctant
  to act
• Treasury Committee recommends Office of Deputy
  Governor in BoE to lead
• Cross-border arrangements
• Much worse when more countries involved
• Who should lead? ECB? Has possibility under the
  Treaty. European Deposit Insurance Corporation
  only for banks with systemic implications - 30-50
• How shall different interests be balanced?
• Whose jurisdiction applies

23
COORDINATION FAILURE

• Government by committee
• Cross-border arrangements
• Much worse when more countries involved
• Who should lead?
• How shall different interests be balanced?
• Whose jurisdiction applies?
• Need agreement about how joint arrangements shall
  run in normal times to avoid recrimination if
  things go wrong
• Lead of consolidating supervisor of college of
  supervisors, with single database, able to
  exercise similar powers (single rule book)
• Agreed PCA respecting all countries financial
  stability
• Closure at positive capital before any losses to
  argue over
• Currently do not have power to take account of
  other countries interests

24
LESSONS

• deposit insurance needs to be designed so that
• the large majority of all individuals balances
  are fully covered
• depositors can all have access to their deposits
  without a material break
• the activation of emergency liquidity assistance
  arrangements needs to give confidence that those
  being assisted will survive, and should be seen
  as the system working as it should, rather than
  signalling some breakdown
• there needs to be a regime of prompt corrective
  action for supervisors whereby prescribed actions
  of increasing severity are required within short
  time periods according to a set of triggers based
  on capital adequacy and risks of failure

25
LESSONS (2)

• there needs to be a legal framework such that the
  functions of systemic importance in banks that
  fail can be kept operating without a material
  break
• such failure should occur before the bank
  becomes insolvent so that there is little chance
  of losses to the taxpayer
• this will normally involve a special insolvency
  regimes for banks
• some designated institution needs to be in charge
  of intervention in failing banks to ensure rapid
  and concerted action
• At a European level far greater coherence among
  the legislation and authorities of member states
  is required if these provisions for the handling
  of problems in domestic banks are to be equally
  successfully handled in the case of large
  cross-border banks

26
REMARKS

• If 1 to 5 had been in place there would have been
  no run on Northern Rock
• But if NR had been a more complex cross-border
  bank the problems could have been much worse
• This unfortunate and unnecessary incident will
  help authorities not just in the UK but round the
  world avoid much greater threats to financial
  stability
• If distinguished and well thought out authorities
  have a problem so will others
• No safety net can be completely successful