Chapter 23: Operational Risk
23.12.
Suppose that there is a 1% probability that operational risk losses of a certain type exceed $10
million. Use the power law to estimate the 99.97% worst-case operational risk loss when the _
parameter equals (a) 0.25, (b) 0.5, (c) 0.9, and (d) 1.0.
(a) In this case K × 10e−0.25 = 0.01 so that K = 0.01778. The 99.97% worst case loss
is (in millions of dollars) x where 0.0889x−0.25 = 0.0003. In this case x = 12,345,679.
23.13.
Consider the following two events: (a) a bank loses $1 billion from an unexpected lawsuit
relating to its transactions with a counterparty and (b) an insurance company loses $1 billion
because of an unexpected hurricane in Texas. Suppose that you have the same investment in
shares issued by both the bank and the insurance company. Which loss are you more concerned
about? Why?
You should be more concerned about the bank loss. Most other insurance companies are likely to
have suffered from the same type of loss as the insurance company in question and insurance
23.14. (Spreadsheet Provided)
The worksheet used to produce Figure 23.2 is on the author’s web site. How does the loss
distribution change when the loss severity has a beta distribution with upper bound of 5, lower
bound of zero, and the other parameters both 1?
The spreadsheet used to calculate Figure 23.2 is in the software section on my web site. In the
to