Morgan Stanley 2010 Annual Report Download - page 109

One method of evaluating the reasonableness of the Company’s VaR model as a measure of the Company’s
potential volatility of net revenues is to compare the VaR with actual trading revenues. Assuming no intra-day
trading, for a 95%/one-day VaR, the expected number of times that trading losses should exceed VaR during the
year is 13, and, in general, if trading losses were to exceed VaR more than 21 times in a year, the accuracy of the
VaR model could be questioned. Accordingly, the Company evaluates the reasonableness of its VaR model by
comparing the potential declines in portfolio values generated by the model with actual trading results. For days
where losses exceed the 95% or 99% VaR statistic, the Company examines the drivers of trading losses to
evaluate the VaR model’s accuracy relative to realized trading results.
The histogram below shows the distribution of daily net trading revenues during 2010 for the Company’s trading
businesses (these figures include revenues from the counterparty portfolio and also include net interest and
non-agency commissions but exclude certain Non-trading revenues such as primary, fee-based and prime
brokerage revenues credited to the trading businesses). During 2010, the Company experienced net trading losses
on 38 days, with zero excesses of the 95%/one-day Trading VaR.
0
10
20
30
40
50
70
60
Number of Days
Year Ended December 31, 2010
Daily Net Trading Revenues
(dollars in millions)
(Loss) Gain
<-75
-75 to -50 -50 to -25
-25 to 0
0 to 25
25 to 50
50 to 75
75 to 100
100 to 125 125 to 150
150 to 175
175 to 200 200 to 225
225 to 250
>250
Non-trading Risks.
Aggregate VaR currently incorporates certain Non-trading risks, such as the interest rate risk generated by
funding liabilities related to institutional trading positions and public company equity positions recorded as
investments by the Company. Investments made by the Company that are not publicly traded are not reflected in
the Aggregate VaR results.
VaR is one method of assessing the risk of the Company’s Non-trading portfolio; however, due to a variety of
factors (e.g., trading restrictions, illiquidity), sensitivity analysis may be a better approach to evaluating this risk.
Reflected below is a sensitivity analysis covering substantially all of the Company’s Non-trading risk.
103