Goldman Sachs 2004 Annual Report Download - page 62

management’s฀discussion฀and฀analysis
60฀฀฀G O L D M A N ฀ S A C H S ฀ 2004฀ ANNUAL฀REPO RT
60฀฀฀G O L D M A N ฀ S A C H S ฀ 2 004฀ A N N U A L ฀ R E P O RT
BES฀•฀Phone฀(201)฀635-5240฀•฀FAX฀(201)฀635-5199
BPX/S10829฀•฀Flow฀15฀•฀Proof฀9฀•฀2/4/05฀•฀0700
As part of our overall risk control process, daily trading net
revenues are compared with VaR calculated as of the end of the
prior business day. Trading losses incurred on a single day
exceeded our 95% one-day VaR on one occasion during 2004.
Distressed฀Asset฀Portfolios
The market risk associated with distressed asset portfolios in
FICC that cannot be properly measured in VaR (primarily due
to inadequate historical data on the underlying assets in the
aggregate) is measured based on a potential 10% decline in the
asset value of such portfolios. The market values of the underly-
ing distressed asset positions are sensitive to changes in a
number of factors, including discount rates and the projected
timing and amount of future cash flows. As of November 2004,
the potential impact of a 10% decline in the asset value of these
portfolios was $416 million compared with $228 million as of
November 2003.
Nontrading฀Risk
smfg฀– ฀The market risk of our investment in the convertible
preferred stock of SMFG is measured using a sensitivity analysis
that estimates the potential reduction in our net revenues associ-
ated with a 10% decline in the SMFG common stock price. As
of November 2004, the sensitivity of our investment to a 10%
decline in the SMFG common stock price was $236 million
compared with $75 million as of November 2003. The change
is primarily due to an increase in the SMFG common stock
price and the passage of time in respect of the transfer restric-
tions on the underlying common stock. This sensitivity should
not be extrapolated to other movements in the SMFG common
stock price, as the relationship between the fair value of our
investment and the SMFG common stock price is nonlinear.
othe r ฀ p rinc i pa l฀ inv e stmen t s฀– ฀The market risk for
financial instruments in our nontrading portfolio, including our
merchant banking investments but excluding our investment in
the convertible preferred stock of SMFG, is measured using a
sensitivity analysis that estimates the potential reduction in our
net revenues associated with a 10% decline in equity markets.
This sensitivity analysis is based on certain assumptions regard-
ing the relationship between changes in stock price indices and
changes in the fair value of the individual financial instruments
in our nontrading portfolio. Different assumptions could produce
materially different risk estimates. As of November 2004,
the sensitivity of our nontrading portfolio (excluding our
investment in the convertible preferred stock of SMFG) to a
10% equity market decline was $118 million compared with
$104 million as of November 2003, primarily reflecting an
increase in the carrying value of our public principal investments.
CREDIT฀RISK
Credit risk represents the loss that we would incur if a counter-
party or an issuer of securities or other instruments we hold fails
to perform under its contractual obligations to us. To reduce
our credit exposures, we seek to enter into netting agreements
with counterparties that permit us to offset receivables and
payables with such counterparties. In addition, we attempt to
further reduce credit risk with certain counterparties by entering
into agreements that enable us to obtain collateral from a coun-
terparty or to terminate or reset the terms of transactions after
specified time periods or upon the occurrence of credit-related
events, by seeking third-party guarantees of the counterparty’s
obligations, and through the use of credit derivatives and other
structures and techniques.
For most businesses, counterparty credit limits are established
by the Credit Department, which is independent of the revenue-
producing departments, based on guidelines set by the Firmwide
Risk Committee and the Credit Policy Committee. For most
products, we measure and limit credit exposures by reference to
both current and potential exposure. We typically measure
potential exposure based on projected worst-case market move-
ments over the life of a transaction within a 95% confidence
interval. For collateralized transactions, we also evaluate poten-
tial exposure over a shorter collection period, and give effect to
the value of collateral received. We further seek to measure
credit exposure through the use of scenario analyses, stress tests
and other quantitative tools. Our global credit management
systems monitor current and potential credit exposure to indi-
vidual counterparties and on an aggregate basis to counterpar-
ties and their affiliates. The systems also provide management,
including the Firmwide Risk and Credit Policy Committees,
with information regarding overall credit risk by product,
industry sector, country and region.
As of both November 2004 and November 2003, we held U.S.
government and federal agency obligations that represented 5%
and 6% of our total assets, respectively. In addition, most of our
securities purchased under agreements to resell are collateral-
ized by U.S. government, federal agency and other sovereign
obligations. As of November 2004 and November 2003, we did
not have credit exposure to any other counterparty that
exceeded 5% of our total assets. However, over the past several
years, the amount and duration of our credit exposures have
been increasing, due to, among other factors, the growth of our
lending and OTC derivatives activities. A further discussion of
our derivative activities follows below.