Wells Fargo 2013 Annual Report Download - page 92

Risk Management – Asset/Liability Management (continued)
monitored at both the business unit level and at aggregated
levels on a daily basis. Our corporate market risk management
function aggregates all Company exposures to monitor whether
risk measures are within our established risk appetite. Changes
to the Company’s market risk profile are analyzed and reported
on a daily basis. The Company monitors various market risk
exposure measures from a variety of perspectives, which include
line of business, product, risk type and legal entity.
Value-at-Risk Overview VaR is a statistical risk measure used to
estimate the potential loss from adverse moves in the financial
markets. We utilize VaR models to measure market risk on an
aggregate basis as well as on a disaggregated basis for each
individual line of business. The VaR measures assume that
historical changes in market values (historical simulation
analysis) are representative of the potential future outcomes and
measure the expected loss over a given time interval (for
example, 1 day or 10 days) within a given confidence level. The
historical simulation analysis approach uses historical changes
of the risk factors from each trading day in the previous
12 months. The risk drivers of each trading position with respect
to interest rates, credit spreads, foreign exchange rates, and
equity and commodity prices are updated on a daily basis. We
measure and report VaR for a 1-day holding period and a 10-day
holding period at a 99% confidence level. This means that we
would expect to incur single day losses greater than predicted by
VaR estimates for the measured positions one time in every
100 trading days. We treat data from all historical periods as
equally relevant and consider utilizing data for the previous
12 months as appropriate for determining VaR. We believe using
a 12 month look back period helps ensure the Company’s VaR is
responsive to current market conditions.
VaR measurement between different financial institutions is
not readily comparable due to modeling and assumption
differences from company to company. VaR measures are more
useful when interpreted as an indication of trends rather than an
absolute measure to be compared across institutions.
The VaR model is subject to limitations which are well
established in the industry. Some of the primary limitations
include availability of historical data and determining the
appropriate mathematical model assumptions. These limitations
are monitored by a management committee of the Market Risk
Committee and Corporate Model Risk Committee (CMoR). The
CMoR consists of senior executive management and reports on
material model risk issues to the Risk Committee of the Board.
Sensitivity Analysis Overview Sensitivity analysis is the measure
of exposure to a single risk factor, such as a one basis point
increase in rates or a 1% increase in equity prices. We conduct
and monitor sensitivity on interest rates, credit spreads,
volatility, equity, commodity, and foreign exchange exposure.
Since VaR is based upon previous moves in market risk factors
over recent historical periods, it may not provide accurate
predictions of future market moves. Sensitivity analysis
complements VaR as it provides an indication of risk relative to
each factor irrespective of historical market moves.
Stress Testing Overview While VaR captures the risk of loss due
to adverse changes in markets using recent historical market
data, stress testing captures the Company’s exposure to extreme,
but low probability market movements. Stress scenarios
estimate the risk of losses based on management’s assumptions
of abnormal but severe market movements such as severe credit
spread widening or a large decline in equity prices. These
scenarios also assume that the market moves happen
instantaneously and no repositioning or hedging activity takes
place to mitigate losses as events unfold (although experience
demonstrates otherwise).
An inventory of scenarios is maintained representing both
historical and hypothetical stress events that affect a broad range
of market risk factors with varying degrees of correlation and
differing time horizons. Historical scenarios utilize an event-
driven approach: the stress scenarios are based on plausible but
rare events, and the analysis addresses how these events might
affect the risk factors relevant to a portfolio. Hypothetical
scenarios assess the impact of large movements in financial
variables on portfolio values. Typical examples include a
100 basis point increase across the yield curve or a 10% decline
in stock market indexes. However, this analysis lacks historical
and economic content, which can limit its usefulness.
The Company’s stress testing framework is also used in
calculating results in support of the Federal Reserve Board’s
Comprehensive Capital Analysis & Review (CCAR) and internal
risk measures. Stress scenarios are regularly reviewed and
updated to address potential market events or concerns. For
more detail on the CCAR process, see the “Capital Management”
section in this Report.
Market Risk Monitoring Trading VaR is the VaR measure used
to provide insight into the market risk exhibited by the
Company’s trading positions. The Company calculates Trading
VaR for risk management purposes to establish line of business
risk limits. Trading VaR is calculated based on all trading
positions classified as trading assets or trading liabilities on our
balance sheet. In addition, the Company monitors and manages
a variety of sensitivity exposures and stress testing estimates.
Table 46 shows the results of the Company’s Trading VaR by
risk category. As presented in the table, average Trading VaR
was $21 million for the quarter ended December 31, 2013,
compared with $18 million for the quarter ended
September 30, 2013. The increase was primarily driven by
changes in portfolio composition.
90