Revised Capital Rules from the Perspective of Emerging Markets

1
Revised Capital Rules from the Perspective of
Emerging Markets

• David Carse
• Deputy Chief Executive
• Hong Kong Monetary Authority
• 24 November 1999

2
Basel Committees relationship with the emerging
(non-G10) markets

• The Basel Committee (BC) is increasingly, and
  more explicitly, becoming the standard setter for
  banking supervisors around the world
• The Core Principles have given added impetus to
  the BCs liaison with non-G10 supervisors

3
The Core Principles Liaison Group

• Comprises both G10 and non-G10 supervisors (plus
  IMF, World Bank, ECB)
• Original purpose was to assist in the
  drafting/consultation process on the Core
  Principles
• Has since developed into a broader consultative
  role, including on the new Accord
• CPLG Working Group on Capital set up for this
  purpose

4
Other channels for liaison

• Ad hoc participation by non-G10 experts in BCs
  working group meetings
• More workshops for G10 and non-G10 supervisors on
  a variety of subjects
• Regular meetings with chairmen of the regional
  supervisory groups such as the EMEAP Working
  Group on Banking Supervision which Hong Kong
  chairs
• The non-G10 views expressed in this
  presentation are based on those of CPLG and EMEAP
  Working Group members

5
Reaction of the non-G10 supervisors

• The non-G10 supervisors generally welcome the
  opportunity to engage in the activities of the BC
• There are however some skeptics who see it only
  as a means of the BC to claim legitimacy for
  decisions by G10 countries
• It is important therefore for the BC to take
  non-G10 views fully into account
• It may be necessary in due course to widen the
  membership of the BC

6
Hong Kongs position under the new Accord

• HK is non-OECD but rated A by SP
• We aim to comply fully with the Core Principles
  and we subscribe to the SDDS
• We already apply the Pillar 2 and 3 approach
• The capital ratio for the system as a whole is
  currently around 19
• The impact of the new Accord is estimated to
  reduce the ratio by less than 1

7
Overview of non-G10 reactions to the new Accord

• Non-G10 supervisors generally support the demise
  of the OECD Club Approach
• They support the idea of more closely aligning
  capital with underlying risk
• They acknowledge the importance of supervisory
  review and market discipline
• BUT, there are important doubts/reservations
  about various aspects of the new Accord
• some of these concerns are shared by G10 countries

8
Areas of concern

• The impact on cost of funds and flow of funds to
  non-G10 countries
• Coverage of the Accord
• Use of external credit rating agencies
• Option 1 versus Option 2
• The linkage with international standards
• The applicability of the internal ratings
  approach
• The charge for interest rate and other risks

9
The impact on non-G10 countries

• There is a view among some non-G10 countries that
  the new Accord has been developed primarily for
  the benefit of G10 sovereigns, banks and
  corporates
• Many non-G10 countries are still struggling to
  meet the old Accord and improve quality of
  supervision and transparency
• For such countries, the concern is that their
  access to international markets and cost of funds
  will be adversely affected

10
The coverage of the Accord

• What is the target audience of the new Accord?
• The BCs continued focus on internationally
  active banks seems unrealistic
• the new Accord will become a de facto world
  standard and the interests of the non-G10
  countries should therefore be taken into account
• The non-G10 majority preference is for a unified
  Accord to which they can subscribe
• a separate version for non-G10 countries would be
  seen as second-class
• but there should be options within the Accord
  that would be suitable for non-G10 countries

11
Use of external credit rating agencies

• The use of external CRAs is the most
  controversial aspect of the new Accord
• Some of the opposition is simply a natural
  dislike of criticism from outsiders (such as
  CRAs) that is perceived to be unfair
• But the role of the CRAs during the Asian crisis
  has raised legitimate questions about giving them
  a greater role in the regulatory process
• Even the BC itself seems lukewarm about the use
  of CRAs

12
Issues arising from use of CRAs

• The credibility and track record of the CRAs,
  particularly as regards sovereign risk
• The greater procyclical bias that would be
  introduced into the capital charge
• Competitive inequality caused by limited
  penetration of external ratings in the non-G10
• The accountability of the CRAs (and the possible
  need for regulation)
• Surrender of responsibility by regulators to CRAs

13
Procyclicality

• The use of credit ratings means that capital
  requirements would increase as borrowers were
  downgraded
• Arguable that banks should set aside more capital
  as risk increases
• But the concern (shared by the IMF) is that every
  bank would do this at exactly the same time
• increase the tendency for banks to rush for the
  exit
• would have made it more difficult to maintain the
  standstill for Korean banks at end-1997

14
Anticyclicality

• There is general agreement that supervisors
  should try to encourage banks to build up a
  capital cushion in the good times to allow for a
  decline in capital ratios in the bad times
• Dynamic provisioning can also help to insure
  banks against future downturns
• But this does not remove the threat that a
  ratings downgrade would cause a sharp portfolio
  reallocation in times of stress

15
The case of Korea

• Date SP rating Risk weight
• 21/6/97 AA- 0
• 24/10/97 A (-1) 20
• 25/11/97 A- (-3) 20
• 11/12/97 BBB- (-6) 50
• 22/12/97 B (-10) 100
• cumulative change in rating notches if the
  new Accord had been in effect

16
The case against the CRAs performance in the
Asian crisis

• The allegation is that the CRAs were
  over-optimistic in their ratings prior to the
  crisis and then too aggressive in downgrading
• This is said to have contributed to the abrupt
  reversal of capital flows to the Asian countries
• Rather than being an important independent
  stabilizing force, the major credit rating
  agencies did not behave very differently from the
  vast majority of market participants. (IMF
  International Capital Markets Report 1999)
• Use of the CRAs ratings for capital purposes
  would not have led to capital being built up in
  advance of the crisis against countries like Korea

17
Have the CRAs learned the right lessons?

• In the light of this, should the BC make use of
  the CRAs for sovereign and bank risk?
• The CRAs have of course defended their role in
  the crisis, but they have also acknowledged that
  there are lessons to be learned, eg
• need for greater focus on banking sector weakness
• greater emphasis on the risks of leverage and
  particularly of short-term debt
• increased sensitivity to contagion risk
• increased focus on transparency
• need to devote more resources to sovereign
  ratings

18
Alternatives to the CRAs

• Notwithstanding these changes, the non-G10
  countries remain opposed to the use of CRAs
• Nor are they keen on the use of export credit
  rating agencies
• Some have suggested that the IMF should take on
  the role of issuing sovereign ratings
• but the IMF does not have the resources to do
  this and it would create conflicts of interest
• One option would be to retain the present
  standardized approach despite its flaws - but the
  BC seems determined to change it
• Hence, there seems no alternative to CRAs

19
Option 1 versus Option 2

• If the CRAs are to be used, a number of detailed
  issues arise on the structure of the risk weights
• In particular, there is the question of which
  option should be used for interbank exposures
• In general, non-G10 countries favour the use of
  Option 2, ie ratings directly assigned to banks
• no necessary correlation between the risk of a
  bank and that of its sovereign (to the extent
  that there is a link, it is already built into
  the banks rating)
• Option 1 too directly implies Government support
  for banks

20
Higher risk categories

• Some non-G10 countries are worried that the 150
  weight will require higher capital on top of
  higher provisions as creditworthiness
  deteriorates
• Others (including Hong Kong) see it as a useful
  means of differentiating credit risk
• There is the issue of which exposures should fall
  within this category
• one suggestion is that it should include exposure
  to HLIs which fail to provide sufficient
  information to lenders or which are unrated

21
Other CRA issues

• The criteria for recognition of CRAs and whether
  this should be left to national discretion
• The incentives in the proposed structure for some
  borrowers not to be rated
• The types of credit ratings that will be used -
  in particular, will unsolicited ratings be
  eligible?
• The sharp discontinuity in some of the risk
  weights (eg between AA- and A rated corporates)

22
The internal ratings approach

• It is still unclear how this will work and to
  whom it will be applicable
• The BC originally seemed to envisage that it will
  apply only to some sophisticated banks
• But a number of sophisticated banks in non-G10
  countries may also wish to use this option
• need for consistent international standards
• non-G10 supervisors will need to have adequate
  resources to validate banks rating systems and
  processes

23
The linkage of the capital charge with
international standards

• This proposed linkage with the Core Principles
  and the IMFs SDDS is controversial
• There is greater acceptance of use of the Core
  Principles as a precondition for lower capital
  requirements for claims on banks
• but the precondition is currently somewhat vague
  since it allows for countries simply to be in the
  process of implementing the Core Principles
• some supervisors stress that there is a need for
  clear independent assessment of implementation
• others argue that compliance with the Principles
  is not supposed to be a simple matter of pass/fail

24
The linkage with SDDS

• There is much more opposition to any linkage with
  SDDS for sovereign risk
• only 47 countries have subscribed to SDDS
• difficult to assess compliance with SDDS
• some countries believe that it is fundamentally
  incorrect to link capital standards to SDDS - no
  necessary connection between transparency and
  relative financial strength
• The Asian crisis however shows that lack of
  transparency can destabilize both a country and
  its banks, and increase the riskiness of claims
  on these

25
Interest rate risk and other risks

• There are doubts about whether risk-sensitive
  methodologies for measuring the capital charge
  can be developed, particularly for operational
  risk
• If there is a charge for interest rate risk it
  should be applied not simply to outliers
• However, there is a view that the main
  supervisory concern with such risks should be
  with banks systems of control rather than with
  attempts to quantify a capital charge
• The capital aspects should be dealt with by
  requiring a bigger cushion of capital under
  Pillar 2

26
The other two pillars

• These are not particularly controversial among
  the non-G10 countries
• A number of supervisors (like Hong Kong) already
  mandate minimum capital ratios above the 8
  minimum and set different ratios for different
  banks
• However, there is concern that some supervisors
  may not yet possess the skills and resources to
  implement the recommendations fully

27
Conclusions

• The non-G10 countries generally support the
  direction in which the BC is moving
• The problem is that they disagree with a number
  of the important details released so far
• This is going to pose a problem for the BC - the
  more widely it consults, the more divergent views
  it is going to receive
• But to raise supervisory standards worldwide, it
  is important that the BC tries to take the
  broader supervisory community with it - as it did
  with the Core Principles