Basel 2: The end of the beginning.

1
Basel 2 The end of the beginning.

• Warren Mockett

2
Basel 2 in the EU and USA

• EU Capital Requirements Directive (CRD aka CAD
  3/RBCD) The EU implementation of Basel 2
• Extends Basel requirements to credit institutions
  investment firms, not directive as to approach
  - up to each bank
• Live now - for SA and FIRB, AIRB next year
• Includes a great deal of prescription on controls

• USA Congressional oversight over federal agency
  rule making powers. Political issues - hence
  still at the drafting stage
• Proposing mandatory Basel 2 AIRB only option for
  core/opt-in (250bn total assets or 10bn foreign
  exposure) and Basel 1A for all other banks
• BUT.still seeking views on smaller banks vs.
  bigger banks - Basel 1, Basel 1A, Basel 2 AIRB,
  Basel 2 Standardised
• Time now very tight for January 2009 start

3
Basel 2 Timeline - where we are now
CRD Agreed
Implementation through EU legislatures
Target end of decision making on UK applications
Jan Start of Parallel Run
EU start date AIRB
EU start date SA , FIRB, retail AIRB
implementation
EU
2005
2006
2007
2008
2009
Q3 Draft B2 NPR
US start date for implementation
Start of parallel run for core/opt-in banks
USA
Q3 Draft B1A NPR
March 26 Comments on NPRs close
AT RISK
4
Illustrative IRB Impact on Capital - PD LGD
driven

• B1/B2 benchmark c1 PD, LGD of c.45, maturity
  of 2.5 years
• NB NOT including capital floors, EL charge, op
  risk, or pillar 2

Price increase?
Capital change within 10
Price reduction pressure?
5
But pricing impacts are uncertain

• Pragmatic implementation approach by supervisors
  means uncertainty in the early years. E.g. UK FSA
  may
• Require FSA consent before later roll out to
  certain portfolios
• Require delayed roll out of portfolios with
  shortcomings
• Apply more conservative parameters in Pillar 1
• Apply a bespoke capital buffer or multiplier to
  models in Pillar 1
• Apply a capital floor to a particular model.
• What to do with portfolio level charges for
  credit expected loss, op risk, and pillar 2?
• Banks manage to a different capital requirements
  than the regulatory minimum anyway

6
LGD - future opportunities

• LGD - observations still very low without data
  pooling (with little to no downturn measurement)
• Loss data sharing will become more important -
  e.g. Pan European Credit Data Consortium
  (www.pecdc.org)
• Lending policies and internal policies on default
  management will become much more important - they
  will be the basis of adjustment/interpretation of
  pooled data - likely to be increasingly
  controlled
• Credit Risk Mitigation recognition requirements -
  AIRB Banks must be generally consistent with
  these, FIRB banks and SA must be completely
  compliant. Control infrastructure requires
  upgrade in order to support CRM recognition in
  LGD

7
EAD - future opportunities

• EAD Current balance () (Remaining Drawing
  Capacity () x Conversion Factor ())
• Could be the parameter that makes the most
  difference between banks
• Not as well understood across the industry
• Little consensus on how it should be done, or
  what drivers should referenced
• It is THE fundamental building block of the
  calculation - affects the calculation of LGD AND
  is a multiplier in the calculation BUT there is
  virtually nothing written about it
• For example, it determines when there is an
  exposure to measure.

8
What does Basel 1 say about when exposure starts?

• An exposure exists 30 days after offering a
  facility to a customer

9
What does Basel 2 (and the CRD) say about
exposure?

• This page intentionally blank...

10
EAD - a principles based approach

• Principle 1 A firm should treat a facility as
  an exposure from the earliest date at which a
  customer is able to make drawings under it.
• Principle 2 Where a customer can draw without
  reference a conversion factor is required
• Principle 3 Where a bank can control drawing a
  conversion factor is not required
• Zero can be the right answer - e.g. creditor
  accounts, unadvised/marketing lines, potential
  excesses
• Conditional drawings where conditions are not met
  - CF based on potential drawing over next 12
  months or zero if the bank has option to decline
• Contingent obligations - based on probability of
  line capacity being drawn (through issuance) x
  probability of claim (against issued)

11
EAD Impact

• Banks with strong systems and controls can make
  cases for lower CCFs
• Unadvised lines attractive from capital
  perspective - but not from risk perspective
• Conditional drawings more attractive - not just
  to manage risk
• Only banks which can control conditionality can
  benefit
• Controls will have to recognise draw
  conditionality, fulfilment and timing
• Greater use of conditions in order to create
  regulatory firebreaks
• Contingent obligations - likely to require
  further investment in systems and data capture to
  better understand P(Use) vs. P(Claim)

12
EAD conclusion

• Not just about modelling, the benefits rely on
  the control framework
• Stronger transparent well documented control
  frameworks will allow more flexibility
• Have to accept more prescription, perhaps less
  flexibility
• May bring pillar 2 benefits as well
• Probably not an issue in the short term - but
  next 3 years these issues are going to become
  ever more important
• Transforms far more than the capital calculation
  rules - and this is of course intended

Far from the only issue 1st January 2007 not the
end of Basel 2 ..rather the end of the beginning