Definition
of capital:
The
Committee adopted a definition of capital encompassing both expected (EL) and
unexpected loss (UL). This contrasts with banks’ practice, where economic
capital is held against UL only, and with ISDA’s earlier recommendations. The
Association recognizes however that adopting a definition covering both EL and
UL is defensible, considering the discrepancy in the definition of expected
loss for risk management and provisioning purposes, and the lack of a generally
accepted accounting and tax treatment for general provisions. It would however
be advisable to re-focus the capital requirements on UL as soon as a harmonized
definition of EL has been adopted. Furthermore, for assets/activities where
losses occur routinely (and hence are factored into the margin applied), ISDA
suggests that the Basel Committee excludes EL from the scope of Pillar 1 and
reviews its adequacy as part of Pillar 2.
Calibration:
The
Committee is proposing to benchmark the new capital charge against the global
amount of regulatory capital currently held in the banking system. A more valid
objective should be to ensure that the charges implied for unexpected loss in
the new framework are not disproportionately high compared to the banks’
internal economic capital calculations.
Having
compared the Committee’s IRB charge with the banks’ own estimates, ISDA found a
2:1 ratio between the two. Some of the discrepancy is explained by the
inclusion of an arbitrary 1.5 multiplier in the IRB function. ISDA hopes that
the new impact study launched by the Committee will offer proof of the need to
remove this multiplier. Absent this re-adjustment: (i) the new IRB capital
charge will probably exceed the current credit risk charge for many banks; (ii)
there is little incentive for banks to move from the standardized approach to
IRB treatment.
Credit
risk:
ISDA
particularly appreciates the recognition by the Committee of the fact that
maturity is an important parameter in the setting of banks’ credit risk capital
requirements. ISDA hopes that the ability to use “maturity” as a risk factor is
not confined to the advanced internal rating based (IRB) approach, but is also
available as an option under the foundation IRB approach. ISDA further
recommends, for those banks not using maturity as a risk driver, replacing the
current three-year average tenor by the true “average economic maturity” in the
banks’ books.
ISDA
welcomes the possibility offered to banks of moving towards an advanced
approach under which values can be assigned to the variables in the regulatory
formula by the firms themselves. However, ISDA cautions against the risk of
setting the hurdle too high, by imposing the need for a seven-year history of
loss given default (LGD) and exposure at default (EAD) data to qualify for
treatment under this approach. A five-year history (as for default
probabilities) would be more appropriate.
ISDA also questions the rationale for imposing a 90% floor on the
advanced IRB charge. Not only will the floor calculation impose material costs
on banks, but the risk of seeing a substantial decrease in regulatory capital
stem from a move to the advanced approach seems tenuous, as in practice,
average internal LGD measures revolve around 40%.
Counterparty
risk in the trading book:
It
is unclear in the consultative paper whether the Committee intends to review
the counterparty risk treatment of trading book instruments, in light of
changes brought to the calculation of the credit risk charge in the banking
book. ISDA would appreciate more
clarity on this issue, and has sought, jointly with the BBA and LIBA, to
outline ways in which the treatment of OTC derivatives and repos could be
improved.
One of the recurrent themes has been the need for better recognition of netting effects in the credit equivalent exposure (CEE) calculation. On a counterparty by counterparty basis, the net portfolio effect of repos, as well as OTC derivatives, should be recognized in order to define the size of the exposure. ISDA suggests that greater reliance be placed on the models used by firms to derive CEEs. The technology used is in essence close, or can be related to VaR, which regulators already recognize, subject to validation requirements, in setting banks’ market risk capital charges.
Credit
risk mitigation:
ISDA
supports the Committee’s objective of promoting the use of credit risk
mitigation instruments, but questions whether this objective is compatible with
some of the measures envisaged in the consultative paper, particularly the
introduction of a charge for residual (principally legal) risk. ISDA contends
that legal risk falls within the definition of operational risk and hence is already
specifically charged for under the proposed framework. ISDA does not, based on
existing case law and reported loss rates, see any grounds for suggesting that
legal risk in relation to credit risk mitigation instruments is greater than in
relation to other forms of contracts which banks enter into.
Furthermore,
on a more detailed note, ISDA is extremely concerned by the disparity in
treatment between bank guarantees (0 “W” charge) and credit derivatives (15%
“W” charge). The Committee seems to be lending a premium to the less
standardized, less transparent side of the unfunded credit risk protection
market, for reasons that have not been satisfactorily spelt out. A consequence
may be that protection buyers seek to re-structure transactions as guarantees,
rather than credit default swaps, leading to market fragmentation and a
substantial rise in the cost of credit protection.
Operational
risk:
ISDA
welcomes the progress made towards a clearer definition of operational risk and
its treatment. The Association believes, however, that further dialogue –
including a formal, scheduled impact review – will be necessary in order to
ensure that the primary and shared objective of genuine risk-sensitivity is
attained. The industry takes the management of operational risk very seriously
and remains concerned about ensuring that any charge is proportionate and
effective.
Two
major outstanding items are: 1) the overall calibration is currently set
considerably high and 2) basing capital requirements on the size of a firm will
mean larger firms are overcharged relative to their levels of risk. This issue
is discussed in a joint ISDA-BBA-LIBA paper.
ISDA
is also concerned about double counting (via the ‘W’ factor, as discussed
above) and failure to recognize the beneficial effects of i) earnings, ii)
diversification and iii) risk transfer, each of which have a major impact on
the level of risk actually incurred by firms.
The Association supports a quantitative approach that incorporates
qualitative factors, developed according to the principles laid out in ISDA’s
‘Operational Risk Regulatory Approach Discussion Paper’ published in September
2000.