HSBC 2009 Annual Report Download - page 253

251
risk management policies and measurement
techniques. Each major operating entity has an
independent market risk management and control
function which is responsible for measuring market
risk exposures in accordance with the policies
defined by Group Risk, and monitoring and
reporting these exposures against the prescribed
limits on a daily basis.
Each operating entity is required to assess the
market risks which arise on each product in its
business and to transfer these risks to either its local
Global Markets unit for management, or to separate
books managed under the supervision of the local
Asset and Liability Management Committee
(‘ALCO’). The aim is to ensure that all market risks
are consolidated within operations which have the
necessary skills, tools, management and governance
to manage such risks professionally. In certain cases
where the market risks cannot be adequately
captured by the transfer process, simulation
modelling is used to identify the impact of varying
scenarios on valuations and net interest income.
HSBC uses a range of tools to monitor and limit
market risk exposures. These include sensitivity
analysis, value at risk (‘VAR’) and stress testing.
Sensitivity analysis
(Unaudited)
Sensitivity measures are used to monitor the market
risk positions within each risk type, for example,
present value of a basis point movement in interest
rates, for interest rate risk. Sensitivity limits are set
for portfolios, products and risk types, with the depth
of the market being one of the principal factors in
determining the level of limits set.
Value at risk
(Audited)
VAR is a technique that estimates the potential losses
that could occur on risk positions as a result of
movements in market rates and prices over a
specified time horizon and to a given level of
confidence.
The VAR models used by HSBC are based
predominantly on historical simulation. These
models derive plausible future scenarios from past
series of recorded market rates and prices, taking
account of inter-relationships between different
markets and rates such as interest rates and foreign
exchange rates. The models also incorporate the
effect of option features on the underlying
exposures.
The historical simulation models used by HSBC
incorporate the following features:
• potential market movements are calculated with
reference to data from the past two years;
• historical market rates and prices are calculated
with reference to foreign exchange rates and
commodity prices, interest rates, equity prices
and the associated volatilities; and
• VAR is calculated to a 99 per cent confidence
level and for a one-day holding period.
HSBC routinely validates the accuracy of its
VAR models by back-testing the actual daily profit
and loss results, adjusted to remove non-modelled
items such as fees and commissions, against the
corresponding VAR numbers. Statistically, HSBC
would expect to see losses in excess of VAR only
1 per cent of the time over a one-year period. The
actual number of excesses over this period can
therefore be used to gauge how well the models are
performing.
Although a valuable guide to risk, VAR should
always be viewed in the context of its limitations.
For example:
• the use of historical data as a proxy for
estimating future events may not encompass all
potential events, particularly those which are
extreme in nature;
• the use of a one-day holding period assumes that
all positions can be liquidated or the risk offset
in one day. This may not fully reflect the market
risk arising at times of severe illiquidity, when a
one-day holding period may be insufficient to
liquidate or hedge all positions fully;
• the use of a 99 per cent confidence level, by
definition, does not take into account losses that
might occur beyond this level of confidence;
• VAR is calculated on the basis of exposures
outstanding at the close of business and
therefore does not necessarily reflect intra-day
exposures; and
• VAR is unlikely to reflect loss potential on
exposures that only arise under significant
market moves.