One should make sure that businesses that take on operational risk incur a transparent
capital charge. The methodology for translating operational risk into capital is
typically developed by the Raroc group in partnership with the operational risk
management group.
Operational risks can be divided into those losses that are expected and those that
are unexpected. Management, in the ordinary course of business, knows that certain
operational activities will fail. There will be a ‘normal’ amount of operational loss that
the business is willing to absorb as a cost of doing business (such as error correction,
fraud, etc.). These failures are explicitly or implicitly budgeted for in the annual
business plan and are covered by the pricing of the product or service.
The focus of this chapter, as illustrated in Figure 12.13, has been on unexpected
failures, and the associated amount of economic capital that should be attributed to
business units to absorb the losses related to the unexpected operational failures.
However, as the figure suggests, unexpected failures can themselves be further
subdivided:
Ω Severe but not catastrophic losses. Unexpected severe operational failures, as
illustrated in Table 12.7, should be covered by an appropriate allocation of
operational risk capital These kinds of losses will tend to be covered by the
measurement processes described in the sections above.
Ω Catastrophic losses. These are the most extreme but also the rarest forms of operational
risk events – the kind that might destroy the bank entirely. Value-at-Risk
(VaR) and Raroc models are not meant to capture catastrophic risk, since potential
losses are calculated up to a certain confidence level and catastrophic risks are by
their very nature extremely rare. Banks will attempt to find insurance coverage to
hedge catastrophic risk since capital will not protect a bank from these risks.
Although VaR/Raroc models may not capture catastrophic loss, banks can use
these approaches to assist their thought process about insurance. For example, it
might be argued that one should retain the risk if the cost of capital to support the
asset is less than the cost of insuring it. This sort of risk/reward approach can bring
discipline to an insurance program that has evolved over time into a rather ad hoc
set of policies – often where one type of risk is insured while another is not, with very
little underlying rationale.
Banks have now begun to develop databases of historical operational risk events
in an effort to quantify unexpected risks of various types. They are hoping to use the
databases to develop statistically defined ‘worst case’ estimates that may be applicable
to a select subset of a bank’s businesses – in the same way that many banks
already use historical loss data to drive credit risk measurement.
A bank’s internal loss database will most likely be extremely small relative to the
major losses in certain other banks. Hence, the database should also reflect the
experience of others. Blending internal and external data requires a heavy dose of
management judgement. This is a new and evolving area of risk measurement.
Some banks are moving to an integrated or concentric approach to the ‘financing’
of operational risks. This financing can be achieved via a combination of external
insurance programs (e.g. with floors and caps), capital market tools and self-insurance.
If the risk is self-insured, then the risk should be allocated economic capital.
How will the increasing emphasis on operational risk and changes in the financial
sector affect the overall capital attributions in banking institutions? In the very
broadest terms, we would guess that the typical capital attributions in banks now
stand at around 20% for operational risk, 10% for market risk, and 70% for credit
risk (Figure 12.14). We would expect that both operational risk and market risk
might evolve in the future to around 30% each – although, of course, much depends
on the nature of the institution. The likely growth in the weighting of operational risk
can be attributed to the growing risks associated with people, process, technology and
external dependencies. For example, it seems inevitable that financial institutions will
experience higher worker mobility, growing product sophistication, increases in
business volume, rapid introduction of new technology and increased merger/
acquisitions activity – all of which generate operational risk. 363
Hiç yorum yok:
Yorum Gönder