Bank of Montreal 2013 Annual Report Download - page 78

At a minimum, the following are considered when determining
appropriate valuation adjustments: credit valuation adjustments,
closeout costs, uncertainty, administrative costs, and liquidity and model
risk. Also, a fair value hierarchy is used to categorize the inputs used in
the valuation of securities, liabilities, derivative assets and derivative
liabilities. Level 1 inputs consist of quoted market prices, Level 2 inputs
consist of models that use observable market information and Level 3
inputs consist of models without observable market information. Details
of Level 1, Level 2 and Level 3 fair value measurements can be found in
Note 29 on page 178 of the financial statements.
Our models are used to determine market risk Economic Capital for
each of our lines of business and to determine regulatory capital. For
capital calculation purposes, longer holding periods and/or higher con-
fidence levels are used than are employed in day-to-day risk manage-
ment. Prior to use, models are subject to review under the Model Risk
Corporate Standard by our Model Risk and Vetting group. The Model Risk
Corporate Standard outlines minimum requirements for the identi-
fication, assessment, monitoring and management of models and model
risk throughout the enterprise and is described on page 96.
We measure the market risk for trading and underwriting portfolios
that meet regulatory criteria for trading book capital treatment using the
Internal Models Approach. We also apply this approach in measuring the
market risk for money market portfolios that are subject to AFS
accounting rules under IFRS and are accorded banking book regulatory
capital treatment. For trading and underwriting portfolios covered by the
Internal Models Approach, VaR is computed using BMO’s Trading Book
VaR model. This is a Monte Carlo scenario simulation model, and its
results are used for market risk management and reporting of
exposures. The model computes one-day VaR results using a 99% con-
fidence level and reflects the correlations between the different classes
of market risk factors.
We use a variety of methods to verify the integrity of our risk
models, including the application of back-testing against hypothetical
losses. This process assumes there are no changes in the previous day’s
closing positions and then isolates the effects of each day’s price
movements against those closing positions. Models are validated by
assessing how often the calculated hypothetical losses exceed the VaR
measure over a defined period. This testing result is in line with
regulatory-defined expectations and confirms the reliability of our
models. The correlations and volatility data that underpin our models
are updated monthly, so that VaR measures reflect current levels
of volatility.
The total trading VaR decreased over the year due to reduced
exposure in equity and credit risk factors, coupled with increased
diversification. The total AFS VaR increase was the result of enhanced
risk capture at the beginning of the year, additional assets and the
impact of higher interest rates. Total trading SVaR increased modestly
despite the benefit of increased diversification between risk factors.
Model inputs for SVaR are calibrated to historical data from a period of
significant financial stress, whereas model inputs for VaR are calibrated
to data from a trailing one-year period.
Total Trading Value at Risk (VaR) Summary ($ millions)*
As at
For the year ended October 31, 2013 Oct. 31,
(pre-tax Canadian equivalent) Year-end Average High Low 2012
Commodity VaR (0.4) (0.6) (1.0) (0.2) (0.6)
Equity VaR (6.1) (6.3) (8.5) (4.4) (6.6)
Foreign exchange VaR (0.5) (1.6) (4.4) (0.1) (0.2)
Interest rate VaR (4.6) (4.9) (10.6) (2.3) (4.5)
Credit VaR (5.0) (5.4) (9.4) (4.1) (5.5)
Diversification 7.5 9.3 nm nm 6.7
Total trading VaR (9.1) (9.5) (15.8) (6.7) (10.7)
Total AFS VaR (10.1) (11.0) (14.5) (7.2) (8.9)
Total Trading Stressed Value at Risk (SVaR) Summary
($ millions)* **
As at
For the year ended October 31, 2013 Oct. 31,
(pre-tax Canadian equivalent) Year-end Average High Low 2012
Commodity SVaR (4.7) (3.0) (6.3) (1.1) (2.1)
Equity SVaR (9.8) (9.4) (16.0) (6.2) (10.5)
Foreign exchange SVaR (0.8) (3.2) (7.0) (0.4) (0.3)
Interest rate SVaR (9.5) (10.0) (15.3) (4.9) (11.4)
Credit SVaR (11.0) (10.8) (14.2) (7.7) (9.3)
Diversification 19.9 21.3 nm nm 18.9
Total trading SVaR (15.9) (15.1) (24.1) (10.6) (14.7)
* The tables reflect updated first quarter 2013 metrics.
** Stressed VaR is produced weekly.
nm – not meaningful
MD&A
Material in blue-tinted font above is an integral part of the 2013 annual consolidated financial statements (see page 77).
BMO Financial Group 196th Annual Report 2013 89