Professional Documents
Culture Documents
CHAPTER 1
RISK & ITS MANAGEMENT
LESSON 1:
INTRODUCTION TO RISK
Expected loss
Expected loss
Risk
Do
cu
11D.571.3
1
Risk Is Costly
Regardless of the specific meaning of risk being used, greater
risk usually implies greater cost. To illustrate the cost of risk we
use a simple example: Suppose that two identical homes are in
different but equally attractive locations. The structures have the
same value, say $100,000, and initially there is no risk of damage
to either house. Then scientists announce that a meteor might
hit the earth in the coming week and that one house is in the
potential impact area. We would naturally say that one house
now has greater risk than the other.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
business.
al
management methods.
11.5811
Chapter Objectives
Price Risk
Credit Risk
Pure Risk
Damage to Assets
Legal Liability
Commodity Price Risk
Worker Injury
al
Employee Benefit
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Extra operating
expenses
Bankruptcy costs
(Legal fees)
Price Risk
Price risk refers to uncertainty over the magnitude of cash flows
due to possible changes in output and input prices. Output
price risk refers to the risk of changes in the prices that a firm
can demand for its goods and services. Input price risk refers to
the risk of changes in the prices that a firm must pay for labor,
materials, and other inputs to its production process. Analysis
of price risk associated with the sale and production of existing
and fu-ture products and services plays a central role in strategic
management.
Three specific types of price risk are commodity price risk, exchange
rate risk, and in-terest rate risk. Commodity price risk arises from
fluctuations in the prices of commodities, such as coal, copper,
oil, gas, and electricity, which are inputs for some firms and outputs
for others. Given the globalization of economic activity, output
and input prices for many firms also are affected by fluctuations in
foreign exchange rates. Output and input prices also can fluctuate
due to changes in interest rates. For example, increases in interest
rates may alter a firms revenues by affecting both the terms of
credit allowed and the speed with which customers pay for products
purchased on credit. Changes in interest rates also affect the firms
cost of borrowing funds to finance its operations.
Credit Risk
The risk that a firms customers and the parties to which it has lent
money will delay or fail to make promised payments is known as
credit risk. Most firms face some credit risk for account receivables.
The exposure to credit risk is particularly large for financial institutions, such as commercial banks, that routinely make loans that
are subject to risk of default by the borrower. When firms borrow
money, they in turn expose lenders to credit risk (i.e., the risk that
the firm will default on its promised payments). As a consequence,
borrowing exposes the firms owners to the risk that the firm will
be unable to pay its debts and thus be forced into bankruptcy, and
the firm generally will have to pay more to borrow money as credit
risk increases.
Pure Risk
The risk management function in medium-to-large corporations
(and the term risk man-agement) has traditionally focused on the
management of what is known as pure risk. As summarized in
11D.571.32
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Personal Risk
The risks faced by individuals and families can be classified in a
variety of ways. We classify personal risk into six categories:
earnings risk, medical expense risk, liability risk, physical asset
risk, financial asset risk, and longevity risk. Earnings risk refers
to the potential fluctuation in a familys earnings, which can
occur as a result of a decline in the value of an in-come earners
productivity due to death, disability, aging, or a change in
technology. A familys expenses also are uncertain. Health care
costs and liability suits, in particular, can cause large unexpected
expenses. A family also faces the risk of a loss in the value of
the phys-ical assets that it owns. Automobiles, homes, boats,
and computers can be lost, stolen, or dam-aged. Financial
assets values also are subject to fluctuation due to changes ill
inflation and changes in the real values of stocks and bonds.
Finally, longevity risk refers to the possibility that retired people
will outlive their financial resources. Often individuals obtain
advice about personal risk management from professionals,
such as insurance agents, accountants, lawyers, and financial
planners.
Do
cu
11D.571.3
4. Losses from pure risk usually are not associated with offsetting
gains for other par-ties. In contrast, losses to businesses that
arise from other types of risk often are associated with gains to
other parties. For example, an increase in input prices harms
the purchaser of the inputs but benefits the seller. Likewise, a
decline in the dollars value against foreign currencies can harm
domestic importers but benefit domestic exporters and foreign
im-porters of U S. goods. One implication of this difference
between pure risk and price risk is that losses from pure risk
reduce the total wealth in society, whereas fluctuations in output and input prices need not reduce total wealth. In addition,
and as we hinted above, the fact that price changes often produce
losses for some firms and gains for others in many cases allows
these firms to reduce risk by taking opposite positions in
derivative contracts.
While many of the details concerning pure risk and its management
differ from other types of risk, it is nonetheless important for you
to understand that pure risk and its man-agement are conceptually
similar, if not identical, to other types of risk and their management. To make this concrete, consider the case of a manufacturer
that uses oil in the production of consumer products. Such a firm
faces the risk of large losses from product liability lawsuits if its
products harm consumers, but it also faces the risk of potentially
large losses from oil price increases. The business can manage the
expected cost of product lia-bility settlements or judgments by
making the products design safer or by providing safety
instructions and warnings. While the business might not be able
to do anything to reduce the likelihood or size of increases in oil
prices, it might be able to reduce its exposure to losses from oil
price increases by adopting a flexible technology that allows low
cost con-version to other sources of energy. The business might
purchase product liability insurance to reduce its liability risk; it
Figure 1.3, the major types of pure risk that affect businesses
include:
Do
cu
Insurance
Increased precautions
Other contractual Risk transfers
Hedging
Investments in Information
Loss Control
al
While the concepts and broad risk management strategies are the
same for pure risk and other types of business risk, the specific
characteristics of pure risk and the significant re-liance on insurance
contracts as a method of managing these risks generally lead to
their management by personnel with specialized expertise. Major
areas of expertise needed for pure risk management include risk
analysis, safety management, insurance contracts, and other
methods of reducing pure risk, as well as broad financial and
managerial skills. The insurance business, with its principal
function of reducing pure risk for businesses and in-dividuals,
employs millions of people and is one of the largest industries
in the United States (and other developed countries). In addition,
pure risk management and insurance have a major effect on many
other sectors of the economy, such as the legal sector, medical
care, real estate lending, and consumer credit.
Loss Financing
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
might hedge its risk of loss from oil price increases using oil
futures contracts.
Many types of loss control influence both the frequency and severity
of losses and can-not be readily classified as either loss prevention
or loss reduction. For example, thorough safety testing of
consumer products wi1l likely reduces the number of injuries,
but it also could affect the severity of injuries. Similarly, equipping
automobiles with airbags in most cases should reduce the severity
of injuries, but airbags also might influence the frequency of
injuries. Whether injuries increase or decrease depends on whether
the number of injuries that are completely prevented for accidents
that occur exceeds the number of injuries that might be caused by
airbags inflating at the wrong time or too forcefully, as well as any
in-crease in accidents and injuries that could occur if protection by
airbags causes some driv-ers to drive less safely.
Viewed from another perspective, there are two general approaches
to loss control: (1) reducing the level of risky activity, and (2)
increasing precautions against loss for ac-tivities that are undertaken.
First, exposure to loss can be reduced by reducing the level of risky
activities, for example, by cutting back production of risky products
or shifting atten-tion to less risky product lines. Limiting the level
of risky activity primarily affects the fre-quency of losses. The
main cost of this strategy is that it forgoes any benefits of the
risky activity that would have been achieved apart from the risk
involved. In the limit, exposure to losses can be completely
eliminated by reducing the level of activity to zero; that is, by not
engaging in the activity at all. This strategy is called risk avoidance.
As a specific example of limiting the level of risky activity, consider
a trucking firm that hauls toxic chemicals that might harm people
or the environment in the case of an accident and thereby produce
claims for damages. This firm could reduce the frequency of liability
11D.571.3
Concept Checks
Do
cu
11D.571.3
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
of scale in arranging loss financing. Moreover, many risk management decisions are strategic in nature, and centralization
facilitates effective interaction between the risk manager and senior
management.
Summary
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Firms also vary in the extent to which the risk management function
is centralized, as opposed to having responsibility spread among
the operating units. Centralization may achieve possible economies
6
11D.571.3
UNIT I
CHAPTER 2
OBJECTIVE OF RISK MANAGEMENT
LESSON 2:
RISK MANAGEMENT OBJECTIVE &
COST OF RISK
value.
Discuss possible conflicts between business and societal
objectives.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
Most and perhaps all of these factors can increase the price that the
firm will need to charge for the drug, thus reducing demand. For
a given price, the risk of injury also might discourage some doctors
from prescribing the drug. The risk of injury also might cause the
firm and the medical profession to distribute the drug only to the
most severe cases of the disease, or the firm might even decide
not to introduce the drug. As a result, from the com-panys
perspective, the risk of consumer injury could have a significant
effect on the value of introducing the drug.
Now consider the risk that you will be involved in an auto accident,
which could cause physical harm to you and your vehicle, as well as
exposing you to the risk of a lawsuit for harming someone else.
The possibility of being involved in an accident reduces the value
of driving. Other things being equal, people obviously would
prefer to have a lower likeli-hood of accident. But other things are
11D.571.3
Chapter Objective
Direct Losses
Increased
Precautions
Indirect
Losses
Reduces
Activity
Cost of Residual
Uncertainty
Div ersification
Effects on
Shareholders
Investments in
information
Effects on other
Stakeholders
Insurance
Hedging
Do
cu
Note that when losses are insured, the cost of loss financing
through insurance only re-flects the loading in the policys premium
for the insurers administrative expenses and re-quired expected
profit. The amount of premium required for the expected value
of insured losses is included in the firms expected cost of losses.
Cost of Internal Risk Reduction Methods
Insurance, hedging, other contractual risk transfers, and certain
types of loss control can re-duce the uncertainty associated with
losses; that is, these risk management methods can make the
cost of losses more predictable. Uncertainty also can be reduced
through diversification and investing in information to obtain
better forecasts of losses. The cost of internal risk reduction
includes transaction costs associated with achieving
diversification and the cost associated with managing a
diversified set of activi-ties. It also includes the cost of
obtaining and analyzing data and other types of information to
obtain more accurate cost forecasts. In some cases this may
involve paying another firm for this information; for example,
the pharmaceutical company may pay a risk management
consultant to estimate the firms expected liability costs.
al
Cost of Loss
Control
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Expected
Loss
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
The third major tradeoff is that which often occurs between the
costs of loss financ-ing/internal risk reduction and the cost of
residual uncertainty. For example, if the firm in-curs higher loss
financing costs by purchasing insurance, residual uncertainty
declines. Greater and more costly internal risk reduction also reduces
residual uncertainty.
Concept Checks
3. Assume that you want to fly across the country and that for a
price of $400 the proba-bility of a fatal crash is one in a million
trips. To reduce this probability to one in 1.5 mil-lion trips, the
price of a ticket would increase to $800. Would you be willing
to pay the extra $400?
Cost of Other Types of Risk
We illustrated the cost of risk concept using a business
perspective and analyzing pure risk. However, the cost of risk is
a general concept. With some modification, our discussion of
11D.571.3
al
Large increases in the price of oil could cause indirect costs if, for
example, production is reduced, alternative sources of energy need
to be arranged, or profitable investment is curtailed. The possibility
of indirect costs increases the expected cost of using oil in the production process. Expenditures on loss control, such as redesigning
the production process to allow for the substitution of other
sources of energy, would decrease the expected cost of oil use and
indirect losses.
With regard to loss financing, the manufacturer might choose to
reduce its exposure to the risk of oil price changes with futures
contracts. The ap-propriate use of futures will produce a profit if
oil prices increase, thus offsetting all or part of the loss to the
firm. (If oil prices drop, all or part of the gain that the firm
otherwise would experience will be offset by a loss on its futures
contracts.) However, the use of futures con-tracts involves
transaction costs that are analogous to the loading in insurance
premiums. The firm also might engage in internal risk reduction
by diversifying its activities to reduce the sensitivity of its profits
to oil price changes or by investing in information to obtain better forecasts of oil prices.
You can see from this simple example that the cost of risk concept
illustrated in Figure 2.1 is quite general. This concept provides a
useful way of thinking about and evaluating all types of risk
management decisions.
Firm Value Maximization and the Cost of Risk
Determinants of Value
A businesss value to shareholders depends fundamentally on
the expected magnitude, timing, and risk (variability) associated
with future net cash flows (cash inflows minus cash outflows)
that will be available to provide shareholder-s with a re-turn on
their investment.
Business value and the effects of risk on value reflect an ex ante
perspective: Value de-pends on expected future net cash flows and
risk associated with these cash flows. Cash in-flows primarily result
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Because most investors are risk averse, the risk of cash flows
reduces the price that they are willing to pay for the firms stock
and thus its value (provided that this risk cannot be eliminated
by investors holding a diversified portfolio of investments). For
a given level of expected net cash flows, this reduction in the
firms stock price due to risk increases the expected return from
buying the stock. In other words, the variation in net cash flows
causes investors to pay less for the rights to future cash flows,
which increases the expected return on the amount that they
invest. Thus, a fun-damental principle of business valuation is
that risk reduces value and increases the ex-pected return required
by investors. The actual return to investors in any given period
will depend on realizations of net cash flows during the period
and new information about the expected future net cash flows
and risk.
Maximizing Value by Minimizing the Cost of Risk
Unexpected increases in losses that are not offset by cash
inflows from insurance con-tracts, hedging arrangements, or
other contractual risk transfers increase cash outflows and often
reduce cash inflows, thus reducing the value of a firms stock.
The effects of risk and risk management on firm value before
losses are known reflect their in-fluence on (1) the expected
value of net cash flows and (2) the compensation required by
shareholders to bear risk. Much of basic financial theory deals
with the kind of risk for which investors demand
compensation and the amount of compensation required. We
will have more to say about how risk affects expected cash
flows, risk, and required compen-sation in later chapters. For
now, it is sufficient for you to understand that making risk
man-agement decisions to maximize business value requires an
understanding of how risk and risk management methods
affect (1) expected net cash flows and (2) the compensation for
risk that is required by shareholders.
Do
cu
(2.1)
(2.2)
10
11D.571.3
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
is needed on insurance, loss control, or other methods of reducing the likelihood of financial distress.
Do
cu
11D.571.3
11
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
While the efficiency concept is abstract and the benefits and costs
of risk management are often difficult to measure, the efficiency
goal is nonetheless viewed as appropriate by many people
(especially economists). The main reason for this is that maximizing
the value of resources by minimizing the cost of risk makes the
total size of the economic pie as large as possible. Other things
being equal, this permits the greatest number of economic needs
to be met.
12
11D.571.3
and workers can accurately assess the risk of injury, they can influence
the business to consider the risk of harm by reducing the price
they are willing to pay for pr9ducts and increasing the wages
demanded in view of the risk of injury.
Summary
The overall objective of risk management is to minimize the
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
cost of risk.
Components of the cost of risk include: (1) the expected cost
of losses, (2) the cost of loss con-trol, (3) the cost of loss
financing, (4) the cost of internal risk reduction, and (5) the
cost of any residual uncertainty that remains after loss con-trol,
loss financing, and internal risk reduction methods have been
implemented.
al
Do
cu
(that is, if the private cost of risk is less than the social cost),
the total cost of risk in society will not be minimized when
businesses maximize value. A major function of business liability and workplace injury law is to align private costs with
social costs so that business value max-imization will minimize
the social cost of risk.
Notes:
11D.571.3
13
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Chapter Objective:
UNIT I
CHAPTER 3
INTEGRATED RISK MANAGEMENT &
VALUE CREATION
al
LESSON 3 :
INTRODUCTION TO INTEGRATED RISK
MANAGEMENT
clients.
They engage in financial intermediation and the management
of risk.
Therefore, a banks economic performance, and hence value,
depends on the quality of the provided services and the efficiency
14
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
To fill this gap, some of the leading banks have developed a set
of practical heuristics called Risk-Adjusted Performance Measures
(RAPM) or also better known, named after their most famous
representative, as RAROC (risk-adjusted return on capital). These
measures can be viewed as modified return on equity ratios and
take a purely economic perspective. Since banks are concerned about
unexpected losses and how they will affect their own credit rating,
they estimate the required amount of (economic or) risk capital
that they optimally need to hold and that is commensurate with
the (overall) riskiness of their (risk) positions. To do that, banks
employ a risk measure called value at risk (VaR), which has evolved
as the industrys standard measure for lower tail outcomes (by
choice or by regulation). VaR measures the (unexpected) risk
contribution of a transaction to the total risk of a banks existing
portfolio. The numerator of this modified ROE ratio is also
based on economic rather than accounting numbers and is,
therefore, adjusted, for example, for provisions made for credit
losses (so-called expected losses). Consequently, normal credit
losses do not affect a banks performance, whereas unexpected
credit losses do.
al
Do
cu
11D.571.3
15
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Risk Management
Whatever name they put on itbusiness ... holistic ... strategic ...
enterpriseleading organizations around the world are breaking out of
the silo mentality and taking a comprehensive approach to dealing
with all the risks they face.
-Towers Perrin
For the purposes of the Integrated Risk Management Framework:
Integrated risk management is a continuous, proactive and
systematic process to understand, manage and communicate
risk from an organization-wide perspective. It is about
11D.571.3
scanning;
assessed; and
al
government.
The four elements of the Integrated Risk Management
Framework are presented as they might be applied: looking
outward and across the organization as well as at individual
activities. This comprehensive approach to managing risk is
intended to establish the relationship between the organization
and its operating environment, revealing the interdependencies
of individual activities and the horizontal linkages.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
tools.
all levels;
Results of risk management practices at all levels are integrated
ongoing.
11D.571.3
17
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
18
11D.571.3
management;
Having senior managers champion risk management;
Encouraging innovation, while providing guidance and
risk management;
appraisals;
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
organization;
Introducing risk management components into existing
and
Improving control and accountability systems and processes
11D.571.3
Reporting on Performance
The development of evaluation and reporting mechanisms for
risk management activities provides feedback to management
and other interested parties in the organization and
government-wide. The results of these activities ensure that
integrated risk management is effective in the long term. Some
of these activities could fall to functional groups in the
organization responsible for review and audit. Responsibility
may also be assigned to operational managers and employees
to ensure that information affecting risk that is collected as part
of local reporting or practices is incorporated into the
environmental scanning process. Reporting could take place
through normal management channels (performance reporting,
ongoing monitoring, appraisal) as part of the advisory and
challenge functions associated with risk management.
Reporting facilitates learning and improved decision-making by
assessing both successes and failures, monitoring the use of
resources, and disseminating information on best practices and
lessons learned. Organizations should evaluate the effectiveness
of their integrated risk management processes on a periodic basis.
In collaboration with departments, the Treasury Board of Canada
Secretariat will review the effectiveness of the Integrated Risk
Management Framework and make the necessary adjustments to
ensure sustained progress in building a risk-smart workforce and
environment.
Building Organizational Capacity
19
Human Resources
Building awareness of risk management initiatives and culture;
Broadening skills base through formal training including
experiences; and
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
and external.
A Common Process
A common, continuous risk management process assists an
organization in understanding, managing and communicating
risk. Continuous risk management has several steps. Emphasis
on various points in the process may vary, as may the type, rigor
or extent of actions considered, but the basic steps are similar.
In the exhibits that follow, Exhibit 1 illustrates an example of a
continuous risk management process that focuses on an
integrated approach to risk management, while Exhibit 2
presents a risk management decision-making process in the
context of public policy.
20
11D.571.3
7. Selecting a Strategy
Choosing a strategy, applying decision criteriaresults-oriented,
problem/opportunity driven.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Organizations may vary the basic steps and supporting tasks most
suited to achieving common understanding and implementing
consistent, efficient and effective risk management. A focused,
systematic and integrated approach recognizes that all decisions
involve management of risk, whether in routine operations or
for major initiatives involving significant resources. It is important
that the risk management process be applied at all levels, from the
corporate level to programs and major projects to local systems
and operations. While the process allows tailoring for different
uses, having a consistent approach within an organization assists
in aggregating information to deal with risk issues at the corporate
level.
Exhibit 2: Risk Management in Public Policy: A
Decision-Making Process
Do
cu
11D.571.3
21
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
work environment.
Learning plans are built into organizations risk management practices.
Results of risk management are evaluated to support innovation, capacity
Do
cu
government.
Promote learning
22
exercises.
Demonstrate management leadership
By selecting leaders who are coaches, teachers and good stewards;
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
decision-making;
further action.
Notes:
11D.571.3
23
al
Frauds
Some examples of frauds are:
Custodial
Custodial risk is the loss of prime documents due to theft, fire,
water, termites etc. This risk is enhanced when the documents
are in transit.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
LESSON 4:
DEFINING OPTIMAL RISK LEVEL
Settlement Risk
Settlement risk is the risk of Counter party defaulting on the
day of the settlement. The risk in this case would be 100% of
the exposure if the corporate gives value before receiving value
from the Counter party. In addition the transaction would have
to be redone at the current market rates.
Do
cu
Operating Risk
Operational risk is the risk that the organization may be
exposed to financial loss either through human error,
misjudgment, negligence and malfeasance, or through
uncertainty, misunderstanding and confusion as to
responsibility and authority. Following are the different kinds
of operating risks:
Legal
Regulatory
Legal
Legal risk is the risk that the organisation will suffer financial
loss either because contracts or individual provisions thereof are
unenforceable or inadequately documented, or because the
precise relationship with the counter party is unclear.
Regulatory
24
Front running
Circular trading
Systems
Systems risk is due to significant deficiencies in the design or
operation of supporting systems; or inability of systems to
develop quickly enough to meet rapidly evolving user
requirements; or establishment of a great many diverse,
incompatible system configurations, which cannot be effectively
linked by the automated transmission of data and which
require considerable manual intervention.
Determining Optimal Risk
Seasoned traders know the importance of risk management. If
you risk little, you win little. If you risk too much, you
eventually run to ruin. The optimum, of course, is somewhere
in the middle. Here, Ed Seykota of Galt Capital and Dave Druz
of Tactical Investment Management, present a method to
measure risk and return.
Placing a trade with a predetermined stop-loss point can be
compared to placing a bet: The more money risked, the larger the
bet. Conservative betting produces conservative performance, while
bold betting leads to spectacular ruin. A bold trader placing large
bets feels pressure or heat from the volatility of the portfolio.
A hot portfolio keeps more at risk than does a cold one. Portfolio
heat seems to be associated with personality preference; bold traders
prefer and are able to take more heat, while more conservative
traders generally avoid the circumstances that give rise to heat.
In portfolio management, we call the distributed bet size the heat
of the portfolio. A diversified portfolio risking 2% on each of five
instrument & has a total heat of 10%, as does a portfolio risking
5% on each of two instruments.
Our studies of heat show several factors, which are:
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
The result after n sets of head/ tail cycles is just the result of
al
Coin Flipping
One way to understand portfolio heat is to imagine a series of
coin flips. Heads, you win two; tails, you lose one is a fair
model of good trading. The heat question is: What fixed
fraction of your running total stake should you bet on a series
of flips?
Do
cu
11D.571.3
FIGURE 2: Plotting the return versus the heat illustrates that the
optimal amount bet is 25% of the stake. The curve has a peak
(point of zeroslope) at 25%
25
2. Setting the heat level is far and away more important than
fiddling with trade timing parameters.
al
= 1 + Bet -2Bet 2
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Five instruments: Soybean oil, live cattle, sugar, gold and Swiss
francs
Time span: December 19, 1979, through January 28, 1992
Do
cu
26
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Introduction
The Case
Mark Edwards has recently joined STC Trucking Corporation
(SIC) as Assistant Treasurer. In addition to other traditional
financial duties, Edwards new responsibilities involve the risk
management function, including purchase of insurance for the
corporation. This activity was formerly handled by Al Avery, the
present chief financial officer, a principal and cofounder of STC.
Mr. Avery has charged Edwards to ... get a handle on the
rising costs of property and liability insurance. Edwards
enthusiastically accepted the challenge; however, he knew this
would involve more than insurance alone.
Do
cu
Background
STC has been in operation for four years. It is a closely held
corporation, owned by its President, John Schmidt (55%), Vice
President and Treasurer, Al Avery (25%), and two other
investors (10% each), who do not participate actively in the
management of STC.
STC has been very successful in its various operations during its
brief time in business. The principal business functions of SIC
are the operation; leasing and management of long- haul units
used in transportation of non-hazardous cargo, as a common
carrier. The companys operations are divided between its ownership
and operation of various tractor and trailer units, and the
management of other investor-owned trucking units. The trucking
units are either dispatched from one of the companys own two
terminals, or from one of the terminals of a large trucking/
transportation company which contracts with STC and other
similar trucking companies. STC presently owns some 100-truck
units. All remaining units are owned by other independent investor
syndicates, but are managed by STC under contract. Contract drivers
who are paid by the mile for operating the trucks drive all trucks.
Although these drivers are not considered to be employees of
STC, because of contract requirements with several of the shippers,
workers compensation coverage is carried voluntarily for these
drivers by STC. No other employee benefits are provided for the
contract drivers by STC. Recent court decisions in the state in
11D.571.3
27
LESSON 5:
CASE STUDY OF STANDARD TRUCKING CORPORATION
ASSETS
LIABILITIES
Current Assets
Fixed Assets
$299
7,900
8199
SHAREHOLDERS EQUITY
Stock
Capital
1,841
1,000
2,841
lst Year
2nd Year
3rd Year
Amt.
Amt.
Amt.
Present Year
(Est.)
#
Amt.
WORKERS COMPENSATION
$ 100 - 500
(4) 700
(10) 2,500
(15) 3,100
(22)
3,600
501 - 2,000
( 6) 6,200
2,001 - 10,000
( 8)
8,100
( 5) 16,300 ( 7)
24,000
( 1) 12,500 ( 1) 28,000 ( 2)
38,100
73,800
$ 100 - 500
(1) 420
( 4) 1,610
( 6) 1,800
( 8)
2,800
501 - 2,000
(10) 7,800
(0) 0
FLEET LIABILITY
2,001 - 10,000
10,000 & Above
TOTALS
(13)
14,300
32,340
(0) 0
167,5OO
( 2) 36,500 ( 3) 96,000 ( 2)
865,000
Gross Income
8,975,000
$ 100- 500
(2) 750
Cost of Operations
6,462,720
501-2,000
( 5) 3,110
(10)
7,090
Gross Profit
2,512,280
2,001- 5,000
( 6) 24,800 (11)
34,700
Other Expenses
1,307,425
5,001- 10,000
114,800
10,001- 20,000
77,200
1,204,855
(0) 0
79,400
(12)
( 1) 60,200 ( 2) 101,700 ( 2)
GENERAL LIABILITY
750
3rd
2,000
3,970
Premium $ 12,570
29,700
57,800
98,060
GENERAL LIABILITY
22,000
13,750
16,500
29,000
22,000
22,000
22,000
50
82,500
180,000
360,000
80
120
160
410,000
618,000
Deductible $ 1,000
UMBRELLA
1,000
2,500
2,500
36,000
75,000
16,000
PROPERTY
32,200
185,190
233,400
212,200
(0)
( 1) 714
(0) 0
( 1) 1,600
(0) 0
(0) 0
( 1) 850
(0)
WORKERS COMPENSATION
Payroll
316,870
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
NET of Deductible s
PROPERTY
3,680
al
( 5) 1,510
TOTAL (incldg deductibles) (10) 39,760 (28) 208,500 (39) 307,580 (60)
Value
216,940
Other Income
Do
cu
28
11D.571.3
Do
cu
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
LESSON 6:
INTERACTIVE SESSION
11D.571.3
29
LESSON 7:
BASIC CONCEPT OF RISK
MEASUREMENT
UNIT I
CHAPTER 4
RISK IDENTIFICATION & MEASUREMENT
risk exposures.
Review concepts from probability and statistics.
Apply mathematical concepts to understand the frequency and
al
severity of losses.
Explain the concepts of maximum probable loss and value at
risk.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Risk Identification
The five major steps in the risk management decision-making
process are: (1) identify all significant risks that can cause loss;
(2) evaluate the potential frequency and severity of losses; (3)
develop and select methods for managing risk; (4) implement
the risk management methods chosen; and (5) monitor the
suitability and performance of the chosen risk management
methods and strategies on an ongoing basis. This chapter
focuses on the first two steps of this process.
Do
cu
Chapter Objective
30
Indirect losses also can arise from damage to property that will
be repaired or replaced. For example, if a fire shuts down a
plant for four months, the firm not only incurs the cost of
replacing the damaged property, it also loses the profits from
not being able to produce. In addition, some operating
expenses might continue despite the shutdown (e.g., salaries for
certain managers and employees and advertising expenses).
These exposures are known as business income exposures
(or, sometimes, business interruption exposures), and they
frequently are insured with business interruption insurance. Note
that business interruption losses also might result from
property losses to a firms major customers or suppliers that
prevent them from transacting with the firm. This exposure can
be insured with contingent business interruption insurance.
Firms also may suffer losses after they resume operations if
previous customers that have switched to other sources of supply
do not return. In the event that a long-term loss of customers
would occur and/or a shutdown temporarily would impose large
11D.571.3
Liability Losses
As we analyze in detail in later chapters, firms face potential legal
liability losses as a result of relationships with many parties,
including suppliers, customers, employees, shareholders, and
members of the public. The settlements, judgments, and legal
costs associated with liability suits can impose substantial losses
on firms. Lawsuits also may harm firms by damaging their
reputation, and they may require expenditures to minimize the
costs of this damage. For example, in the case of liability to
customers for injuries arising out of the firms products, the
firm might incur product recall expenses and higher marketing
costs to rehabilitate a product.
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
31
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
for the random variable and the probability of the outcomes. For
the coin flipping
example, Table 3.2 gives the probability distribution for X.
Do
cu
32
11D.571.3
al
Expected Value
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Since the area under the curve in Figure 3.3 equals 1, we can
graphically identify the probability that profits are within a certain
interval. For example, the probability that profits are greater than
$40 million is the area under the curve to the right of $40 million.
The probability that profits are less than $0 is the area under the
curve to the left of $0. The probability that profits are between
$10 and $30 million is the area under the curve between $10 and
$30 million. Thus, the bell-shaped curve in Figure 3.3 tells us that
for the automaker, there is a relatively high probability that profits
will be between $10 and $30 million. In contrast, while very low
profits and very high profits are possible, they do not have a high
probability of happening.
Concept Checks
1. What information is given by a probability distribution? What
are the two ways of describing a probability distribution?
2. Earthquakes are rare, but the property damage can be very large
when they occur. Illustrate these features by drawing a
11D.571.3
33
al
For each of the loss distributions in Table 3.4, the expected value
is $500 (you should verify this for yourself), but the variances of
the three distributions differ. Loss distribution 2 has a larger
variance than distribution 1, because the extreme outcomes for
distribution 2 are farther from the expected value than they are for
distribution 1. Distribution 3 has a larger variance than distribution
2, because even though the outcomes are the same for distributions
2 and 3, the extreme outcomes are more likely with distribution 3
than with distribution 2. That is, the probability of having a loss
far from the expected value ($500) is greater with distribution 3
than with distribution 2. The comparison of distributions 2 and
3 illustrates that the variance depends not only on the dispersion
of the possible outcomes but also on the probability of the
possible outcomes.
The mathematical definitions of variance and standard
deviation show precisely how the probabilities of the different
outcomes and the deviation of each outcome from the expected
value affect these measures of risk. The definitions are:
Concept Check
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
34
11D.571.3
Concept Checks
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
Notes:
11D.571.3
35
al
Skewness
Another statistical concept that is important in the practice of
risk management is the skewness of a probability distribution.
Skewness measures the symmetry of the distribution. If the
distribution is symmetric, it has no skewness. For example,
consider the two distributions for accident losses illustrated in
Figure 3.7. The distribution at the top of Figure 3.7 is
symmetric; it has zero skewness. However, the distribution at
the bottom is not symmetric; it has positive skewness. Many
of the loss distributions that are relevant to risk management
are skewed.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
LESSON 8:
BASIC CONCEPTS OF STATISTICS & PROBABILITY
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
11D.571.3
Correlation
37
Concept Check
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
38
Frequency
The frequency of loss measures the number of losses in a
given period of time. If historical data exist on a large number
of exposures, then the probability of a loss per exposure (or
the expected frequency per exposure) can be estimated by the
number of losses divided by the number of exposures. For
example, if Sharon Steel Corp. had 10,000 employees in each of
the past five years and over the five-year period there were 1,500
workers injured, and then an estimate of the probability of a
particular worker becoming injured would be 0.03 per year
(1,500 injuries/50,000 employee-years). When historical data do
not exist for a firm, frequency of losses can be difficult to
quantify. In this case, industry data might be used, or an
informed judgment would need to be made about the
frequency of losses.
Severity
The severity of loss measures the magnitude of loss per
occurrence. One way to estimate expected severity is to use the
average severity of loss per occurrence during a historical period.
If the 1,500 worker injuries for Sharon Steel cost $3 million in
total (adjusted for inflation), then the expected severity of
worker injuries would be estimated at $2,000 ($3,000,000/
1,500). That is, on average, each worker injury imposed a $2,000
loss on the firm. Again due to the lack of historical data and
the infrequency of losses, adequate data may not be available to
estimate precisely the expected severity per occurrence. With a
little effort, however, risk managers can estimate the range of
possible loss severity (minimum and maximum loss) for a
given exposure.
Expected Loss and Standard Deviation
When the frequency of losses is uncorrelated with the severity
of losses, the expected loss is simply the product of frequency
and severity. Thus, the expected loss per exposure in our
example can be estimated by taking expected loss severity per
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
Tables 3.6 and 3.7 both show that the standard deviation of losses
for high frequency, low severity losses is low, while the standard
deviation is high for low frequency losses with high potential
severity. This relationship is fairly general: Infrequent but potentially
large losses are less predictable and pose greater risk than more
frequent, smaller losses. Using the type of information illustrated
in these tables, firms pay particular attention to exposures that can
produce potentially large, disruptive losses, either from a single
event or from the accumulation of a number of smaller but still
significant losses during a given period.
11D.571.3
39
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
40
11D.571.3
A3.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Level III: The third level aggregates risk across different business
lines, such as banking and insurance subsidiaries. This leads to the
composite picture or cumulative loss distribution at the top
(holding company) level.
Do
cu
A2.
The concentration of those risk positions, or their relative
weights in a portfolio
11D.571.3
41
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Notes:
al
At Level III, there are only a few risk factors; some are likely to
Do
cu
42
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
We have someone take off his shoes and stand with his back
against a wall that is marked with a scale of inches. We note the
number of inches correspond to the top of the persons head.
This process is a measure. The number obtained is a measurement.
We interpret the number as the persons height. The
interpretation is a metric.
al
Do
cu
Value at Risk
Value-at-risk (VaR) is a category of risk measures that describe
probabilistically the market risk of a trading portfolio. VaR is
widely used by banks, securities firms, commodity and energy
merchants, and other trading organizations. Such firms could
track their portfolios market risk by using historical volatility as
a risk metric. They might do so by calculating the historical
volatility of their portfolios market value over a rolling 100
trading days. The problem with doing this is that it would
provide a retrospective indication of risk. The historical volatility
would illustrate how risky the portfolio had been over the
previous 100 days. It would say nothing about how much
market risk the portfolio was taking today.
For institutions to manage risk, they must know about risks
while they are being taken. If a trader mis-hedges a portfolio, his
employer needs to find out before a loss is incurred. VaR gives
institutions the ability to do so. Unlike retrospective risk metrics,
such as historical volatility, VaR is prospective. It quantifies market
risk while it is being taken.
Measure time in trading days. Let 0 be the current time. We know
a portfolios current market value . Its market value
in one
trading day is unknown. It is a random variable. We may ascribe it
11D.571.3
43
LESSON 9:
MEASURING RISK AND VAR
As with its power, the challenge of VaR also stems from its
generality. In order to measure market risk in a portfolio using
VaR, some means must be found for determining the probability
distribution of that portfolios market value. Obviously, the more
complex a portfolio isthe more asset categories and sources of
market risk it is exposed tothe more challenging that task
becomes.
It is worth distinguishing between three concepts:
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
44
11D.571.3
[1]
[4]
al
1.645 + ( )
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
11D.571.3
(7)
[3]
Copy Right: Rai University
45
al
[8]
[9]
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
[11]
Do
cu
46
11D.571.3
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
The first piece of the puzzle is the key factors . Because they are
observable financial variables, historical data should be available
for them. Based on this, we can characterize the joint distribution
of
. We may do so with standard deviations
and
correlations
for the , or we may do so in some other manner.
Our problem, then, is to convert that characterization of the
distribution of into a characterization of the distribution of . On
its own, our characterization of the distribution of is not enough
to do this. Because it is independent of the portfolios
composition, it cannot, on its own, tell us how risky the portfolio
is.
Do
cu
The second piece of the puzzle is the portfolio mapping [8] that
relates to . That formula will change over time, evolving to reflect
the portfolios changing composition. Formula [8] contributes to
our analysis what the characterization of the distribution of does
not. It reflects the portfolios composition. On its own, however,
it cannot tell us how risky the portfolio is, for it contains no
information relating to market volatility.
11D.571.3
Transformation procedure.
47
VaR Metric
It assumes familiarity with concepts described in the articles
value-at-risk and measuring VaR. It is worth distinguishing
between three concepts:
A VaR measure is an algorithm with which we calculate a portfolios
VaR.
A VaR model is the financial theory, mathematics, and logic that
motivate a VaR measure. It is the intellectual justification for the
computations that are the VaR measure.
A VaR metric is our interpretation for the output of the VaR
measure.
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Linear transformations,
Quadratic transformations,
Do
cu
time 0; and
[1]
measures)
[2]
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
base currency.
percentage.
Do
cu
base currency.
percentage.
For example, we may speak of a portfolios
1-day standard deviation of simple return USD VaR,
2-week 95% JPY VaR, or
1-week 90% ETL GBP VaR, etc.
Notes:
11D.571.3
49
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
LESSON 10:
BEYOND VAR & CASH AT RISK
Do
cu
50
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Incremental Decomposition
Incremental decomposition is similar to regression analysis.
Express the return r on a portfolio in terms of the changes in
(or returns to) K factors, i.e.
al
Do
cu
Marginal Decomposition
11D.571.3
51
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Evidence suggests that the GPD provides a good fit to the tails
of the distributions of changes in individual market rates and
prices (see, Neftci (2000)); and EVT appears to be useful in
measuring credit risk when there is a single important factor (Parisi
(2000)). However, the available empirical evidence does not bear
directly on the question of whether EVT is useful for measuring
the VaR of portfolios that depend (perhaps nonlinearly) on
multiple sources of risk. Classical EVT is univariate, i.e. it does
not characterize the joint distribution of multiple risk factors. To
apply EVT to portfolios that depend on multiple sources of risk
one must estimate the distribution of P/L by historical simulation,
and then fit the GPD to the tail of the distribution of P/L. We are
not aware of any results on the performance of this approach.
Coherent Measures of Risk
VaR was criticized from the outset because it says nothing about
the magnitude of losses greater than the VaR. A more subtle
criticism of VaR is that it does not correctly capture the effect of
diversification (even though capturing the benefits of
diversification is one of the commonly cited advantages of VaR).
To see this, suppose the portfolio contains short digital puts and
calls on the same underlying. Each option has a notional amount
of $10 million, a 4 percent probability of being exercised, time to
52
11D.571.3
But even if one can use standard VaR-like tools to model the
quantitative impact of Dells FX exposure, this risk is likely to be
second-order relative to, say, the risk that Dell does a poor job of
marketing and customer support and loses significant market
share to Gateway and Compaq. The bottom line is that while one
canand some consultants doattempt to implement a bottomup VaR analogue to companies like Dell, there is a danger that
such an approach will simply leave out some important sources
of risk, badly mis-measure others, and thus lead to a highly
inaccurate estimate of overall C-FaR.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
11D.571.3
53
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
We begin by assembling quarterly income statement and balancesheet data from Compustat. In our baseline analysis, we pool
together data from firms in all non-financial industries. However,
our model can also be applied to individual well-defined industries
in which there are enough firms. Electricity companies represent
one such industry; we will review this industry-specific application
later.
55
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
But to define a shock, one needs a benchmark for what cash flow
is expected to be in the absence of a shock. In other words, our
shocks correspond to forecast errorsto deviations of cash flow
from their expected values. Thus the cash flow forecasts we
construct are not an end in and of themselves, but rather a necessary
ingredient to construct these forecast errors.
Do
cu
56
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
11D.571.3
57
one finds the year-ahead number for Dell. This number is 28.50,
which should be read as saying that, in a five-percent worst-case
year, Dells EBITDA would fall short of expectations by $28.50
for every $100 of book assets that it has. For example, applying
the model at the start of Dells 1999 fiscal year, when its book
assets (net of cash and securities) stood at $3,696 million,11 the
conclusion is that a five-percent worst-case scenario for 1999 would
involve an EBITDA shortfall of $1,053 million (1,053 = 3696
.285) To get a sense of proportion, this figure can be compared to
Dells actual realized 1999 EBITDA of $2,419 million.
Another question that one might ask is: how do the C-FaR
distributions vary for electricity companies with different
characteristics? Naturally, we no longer have enough data to chop
this much-smaller pool of forecast errors into 81 separate bins.
But there is no longer any need to. Firstly, the observations are all
from firms in the same industry, so there is no need to do an
industry cut. Moreover, most are relatively large (as compared to
the overall Compustat sample), so there is less need to sort on
market cap as well. Instead, we streamline our sorting procedure
so that we just do a pair of two-way sortsone on profitability
and one on stock-price volatilitythereby dividing the pool of
forecast errors into four sub samples.
al
Notes:
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
that shocks to cash flow are normally distributed. Finally, once the
model is built, it can be easily and at relatively low cost applied to
any number of non-financial companies.
Of course, none of this is to claim that our approach dominates
the alternative of building a company-specific C-FaR model from
the bottom up. Rather, the two approaches can be thought of as
complementary. For example, our model can be used to provide a
reality check on the results produced by an in-depth bottom-up
analysis. Again, the analogy to valuation practices is informative.
Comparables methods are widely (though not exclusively) used
by practitioners to value companies. And in spite of their inability
to factor in certain types of company-specific information, it would
be hard to argue that they do not represent an important part of
the pragmatic persons valuation toolkit. We hope that our
approach to estimating C-FaR will prove to be similarly useful.
58
11D.571.3
LESSON 11:
RISK REDUCTION THROUGH POOLING
LOSSES
risk.
Show how correlation in losses affects the amount of risk that
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
The first column of Table 4.2 lists the possible outcomes for
Emily and Samantha with pooling. If neither woman has an
accident, total accident costs are zero and each woman pays zero. If
either of the women has an accident, total accident costs are $2,500
and each woman pays $1,250. If both women have an accident,
total accident costs equal $5,000 and each pays $2,500.
al
Do
cu
11D.571.3
Probability
0.80
0.20
Total Cost
Average Loss
Probability
(0.8)(0.8) = 0.64
2500
1250
(0.2)(0.8) = 0.16
2500
1250
(0.2)(0.8) = 0.16
5000
2500
(0.2)(0.2) = 0.04
59
Chapter Objective:
UNIT I
CHAPTER 5
DIVERSIFICATION OF RISK
Standard deviation =
{0.64 X ($0 - $500)2 + 0.32 X ($1,250 - $500)2 + 0.04 X ($2,500 - $500)2}1/2 = $707
Do
cu
60
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
11D.571.3
a.
b.
c.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
losses, and one persons unexpectedly high losses are less likely to
be offset by another persons unexpectedly low losses.
Do
cu
11D.571.3
61
al
Concept Check
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Figure 4.4 also illustrates the intermediate case, where losses are
characterized by less than perfect positive correlation; As can be
seen, the standard deviation of average losses decreases as the
number of participants increases, but the standard deviation does
not ap-proach zero. The amount of risk (standard deviation)
cannot be reduced as much by adding participants when losses are
positively correlated; the greater the degree of correlation, the less
is the reduction in risk.
b.
The expected loss for each participant is $500, and losses
for the 100 participants are positively correlated.
c.
The expected loss for each participant is $1,500, and losses
for the 100 participants are independent.
d.
The expected loss for each participant is $1,500, and losses
for the 100 participants are positively correlated.
Insurers as Managers of Risk Pooling Arrangements
As we just learned, individuals or businesses can reduce their
risk by forming a pooling arrangement. As a result, risk-averse
individuals 1!Pd businesses that value lower risk would have
strong incentives to participate in pooling arrangements if they
could be organized at zero cost. However, risk-pooling
arrangements obviously are not costless to operate. Indeed, the
cost of organizing and operating pooling arrangements is the
main reason why insurance companies exist and why most
pooling arrangements take place indirectly through insurance
contracts. In essence, insurance contracts are a way of lowering
the costs of operating pool-ing arrangements.
Types of Contracting Costs
Do
cu
62
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
There are many ways that people and businesses diversify risk in
addition to pooling arrangements through insurance contracts.
Stock markets, for example, provide a mecha-nism for
entrepreneurs to share risk associated with new business ventures
with other peo-ple. A share of stock entitles the owner of the
share to a portion of a companys dividends. If the company
does well, then dividends and/or stock prices will increase, and
the owner will gain accordingly. If the company does poorly then
the dividends and/or stock prices will decline, and the owner of
the share will lose accordingly. Thus the owner of the share of
stock shares in risk of the venture.
al
Do
cu
11D.571.3
63
UNIT I
CHAPTER 6
RISK DECISIONS
LESSON 12:
RISK REDUCTION DECISIONS
risk retention/reduction.
Summarize evidence indicating which types of firms are more
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
reduction activities.
Explain the advantages and disadvantages of following a
disaggregated approach to risk reduction.
al
Do
cu
Chapter Objectives
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Firms also can provide insurers with any available evidence that
their expected claim costs might be lower than predicted by the
insurer.
al
Do
cu
11D.571.3
65
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Product Characteristics
When consumers expect future services from the provider of
products and services, then the demand for those products and
services will depend on con-sumers perceptions about the
likelihood that the provider will be able to provide the future
services. Of course, the likelihood that a firm will be able to
provide futures services is in-versely related to the likelihood of
bankruptcy. Consumer durables, such as electronic equipment
and cars, and financial services, such as insurance, are examples
of products and services for which consumer demand is likely
to be especially vulnerable to consumers per-ceptions about the
providers probability of bankruptcy. Thus, firms in industries
such as these tend to benefit more from risk reduction than
firms in industries that produce prod-ucts for which future
services are not expected.
Do
cu
Many of the arguments for why firms should reduce risk suggest
a more aggregate fo-cus. For example, the progressive tax rate
argument implies that firms should focus on tax-able income,
which depends on many sources of risk, including property losses,
exchange rates, and so on. If a more aggregate approach is adopted,
then firms need to consider in-teractions between the various
sources of risk. This section has two objectives: First, we highlight
the level at which risk reduction would take place under each of
the arguments we outlined in the previous chapters for why firms
should reduce risk. Second, we discuss the advantages and
disadvantages of the disaggregated versus aggregated approach
to risk reduction.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Some studies have found that firms with relatively greater research
and development expenses are more likely to use derivatives. This
finding is consistent with one of the reasons for reducing risk
(hedging) discussed earlier. Firms that make large investments in
research and development need funds on a consistent basis. If
internal funds are not available, then these firms will have to either
raise costly external capital or forgo some research and development expenditures. To ensure that internal funds are
available, firms with greater re-search and development are more
likely to hedge.
Do
cu
There is also evidence that firms are more likely to hedge as their
financial leverage increases. One study examined the hedging
practices of oil and gas producers. The output of these firms is
subject to oil and gas price risk. If the price of gas decreases, then
all else equal, revenues decrease. Fortunately, this risk can be hedged
with derivative contracts. Among oil and gas producers that use
derivatives, the extent of hedging (the proportion of expected
output) increases as the firms financial leverage ratio increases.
Aggregated or Disaggregated Risk Management?
Assuming risk reduction is appropriate, firms must decide
where to focus their risk reduc-tion activities. Should firms take
a disaggregated or micro approach and hedge (insure) each
individual risk exposure separately? Or, should firms hedge
(insure) some aggregate or macro measure of performance,
$ 50 million
$ 25 mi llion
0 million
$ 50 million
$ 25 million
0 million
To capture the idea that firms often want to avoid large losses,
assume that the managers do not want total retained losses to
exceed some critical value, say $40 -million (perhaps be-cause the
firm would then be forced to raise costly external capital or violate
a debt covenant). The firm can insure each loss exposure to achieve
its objective, but assume that contracts are priced so that the firm
must pay 120 percent of the contracts expected pay-out, implying
a 20 percent loading or transaction cost. As you will see below, this
propor-tional transaction cost can make the cost of managing
each exposure separately greater than the cost of managing the
bundled exposure.
Recall that the firm was willing to retain losses up to $40 million.
The important point to notice is that in some cases the coverage
provided by the separate contracts results in a pay-out from the
insurer even though retained losses are less than $40 million. In
these cases, the firm has purchased coverage that, ex post, it did
not really need. We refer to this extra cover-age as unnecessary
coverage, and report it in the final column. The problem with
purchasing unnecessary coverage under these assumptions is that
there is a positive loading associated with purchasing coverage.
Thus, the unnecessary coverage is costly for the firms owners.
Now suppose that the firm was able to purchase an insurance
policy that would indem-nify the firm based on total losses. To
achieve its objective of not having retained losses ex-ceed $40
million, the firm could use one policy under which the insurer
would pay aggregate (sum of property and liability) losses in
excess Of $40 million. We refer to a policy like this that bundles
multiple exposures as a bundled policy. The outcomes with the
bundled policy are summarized in Panel B of above table. The
important point is that with the bundled policy, there is no
unnecessary coverage.
al
Liability Loss =
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
68
11D.571.3
Several of the reasons for reducing risk imply that a firm should
al
Notes:
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Moral Hazard
A completely bundled policy would only have an aggregate
retention level and an aggregate limit; consequently, the source
of a loss would not matter for the contracts payoff. The
problem with such a policy is that once a firms aggregate
retention level was reached, an additional loss (up to the
aggregate limit) would be covered. Such a policy therefore
would greatly reduce the insureds incentive to reduce additional
losses once the retention level was reached. To mitigate this
moral hazard problem, per occurrence deductibles for each type
of loss exposure would likely be included in any bundled policy.
large losses that could cause financial distress or cause the firm
to raise costly external capital.
Do
cu
11D.571.3
69
The Portfolio
In fact, Bob Citron was implementing a big bet that interest rates
would fall or stay low. The $7.5 billion of investor equity was
leveraged into a $20.5 billion portfolio. Through reverse repurchase
agreements, Citron pledged his securities as collateral and reinvested
the cash in new securities, mostly 5-year notes issued by
government-sponsored agencies. One such agency is the Federal
National Mortgage Association, affectionately known as Fannie
Mae. The portfolio leverage magnified the effect of movements
in interest rates. This interest rate sensitivity is also known as
duration.
Introduction
In December 1994, Orange County stunned the markets by
announcing that its investment pool had suffered a loss of $1.6
billion. This was the largest loss ever recorded by a local
government investment pool, and led to the bankruptcy of the
county shortly thereafter.
al
Summary
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
LESSON 13:
CASE STUDYORANGE COUNTY CASE:
USING VALUE AT RISK TO CONTROL FINANCIAL RISK
Do
cu
70
Value at Risk
What is VAR? VAR is a method of assessing risk that uses
standard statistical techniques routinely used in other technical
fields. Formally,
VAR is the maximum loss over a target horizon such that there
is a low, prespecified probability that the actual loss will be
larger. Based on firm scientific foundations, VAR provides users
with a summary measure of market risk. For instance, a bank
might say that the daily VAR of its trading portfolio is $35 the
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
1. Duration approximation.
The state auditor reported the effective duration of the pool as
7.4 years in December 1994. This high duration is the result of
two factors: the average duration of individual securities of 2.74
years (most of the securities had a maturity below 5 years), and
the leverage of the portfolio, which was 2.7 at the time. In 1994,
interest rates went up by about 3%. Compute the loss predicted
by the duration approximation and compare your result with
the actual loss of $1.64 billion.
Advanced (2)
Next, we check whether the assumption of a conditional
normal distribution seems adequate for changes in yields.
Compute the number of exceptions at the 1-tailed 95% level,
using the monthly volatility forecast just computed and the
actual increase in yield. Test whether the number of exceptions
is in line with what was expected.
(For the exception test, you can use the normal approximation
to the binomial distribution. Also, be careful to match the
volatility forecast with the subsequent change in yield.)
al
Questions
Let us place ourselves in the position of the county
Supervisors, who had to decide in December of 1994 whether
to liquidate the portfolio or maintain the strategy (obviously,
based on past information only). At that time, interest rates
were still on an upward path. A Federal Open Market
Committee meeting was looming on December 20, and it was
feared that the Fed would raise rates further. To assess the
possibility of future gains and losses, VAR provides a simple
measure of risk in terms that anybody can understanddollars.
using the actual distribution for yield changes (HistoricalSimulation method). Compare the VAR obtained using the
two methods.
1. Hedging.
On that day, the March T-bond futures contract closed at 9905. The contract has notional amount of $100,000. Its duration
duration can be measured by that of the Cheapest-To-Deliver
(CTD) bond, which is assumed to be 9.2 years. Compute the
number of contracts to buy or sell to hedge the Orange County
portfolio.
Do
cu
It seems that both in 1994 and 1995, interest rate swings were
Notes:
Advanced (1)
Compute a time-varying volatility of changes in yields using the
Risk Metrics approach to see if the recent volatility is abnormally
high. The exponential model (as used in Risk metrics) is:
Var[dy(t)] = Var[dy(t-1)] * k + [dy(t-1)*dy(t-1)]*(1-k)
where Var[dy(t)] is the conditional, predicted variance for time t
and k is the decay factor, usually selected as 0.97 for monthly
data. The model states that the variance forecast is a combination
of the previous month forecast and of the latest squared
innovation. For the starting value of the variance (at time t=0),
use the average variance over the whole period.
11D.571.3
71
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Notes:
Do
cu
LESSON 14:
INTERACTIVE SESSION
72
11D.571.3
UNIT I
CHAPTER 7
HEDGING WITH DERIVATIVES
LESSON 15:
EXPOSURE MANAGEMENT OVERVIEW
al
Introduction
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
73
Chapter Objectives
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
Financial Risks
A companys activities face different kinds of risks. In order to
be able to introduce financial risks, a general definition of risk
conception is needed. Risk, according to Oxelheim and
Wihlborg (1997, p.18), is a measure of unanticipated changes.
In our paper, we brake down every type of risk that a company
might face into two groups: financial and non-financial. We will
leave non-financial risks, as we are not concern with them, and
concentrate on financial ones. Financial risk is the likelihood and
magnitude of unanticipated changes in interest, exchange and
inflation rate risks. As one might expect, financial risk might be
broken down into the interest rate, exchange rate and inflations
rate risks. According to Oxelheim and Wihlborg (1997, p.27- 28)
the above-mentioned risks are defined in the following way:
Interest rate risk refers to the magnitude and likelihood of
risk.
Exchange, interest and inflation changes in the market are very
interrelated and usually have a high degree of correlation. The
main reason why these three factors recently became of major
concern is the effect they were having on the firms value. The
above mentioned factors are the main causes of the companys
financial risk exposure and value volatility. In other words, they
might influence the companys value in a positive way, when the
company is worth more than expected (upside risk), or in the
74
11D.571.3
balance sheet,
Transaction risk comes from future sales and purchases certain
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
11D.571.3
together with price changes can alter the amounts and riskiness
of a companys future revenue and cost streams, i.e. operating
cash flows.
75
sale,
Balance sheet risk= mismatch between assets and liabilities in a
given currency.
Kenyon (1990) was citing his previous book as follows: Kenyon
further suggested that any of these financial risks could be viewed
and managed either in accounting terms, i.e. as accounting or
translation risk, or in cash terms, i.e. as transaction risks, but that
these two concepts referred to different ways in which management
looks at the same risks, rather than two different risks.
One contrast between Shapiro and Kenyons (1981) classifications
stands out: Shapiro regards the main division as being between
the accounting model and the economic or cash flow model,
whereas Kenyon (1981) gives primacy to the contrast between
risks from the real and nominal exchange rates, a contrast also
stressed by Shapiro.
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Problem Discussion
Among the above-mentioned financial exposures, it is especially
the exchange rate risk exposure that becomes more and more
important in light of world markets globalisation and
internationalization. Foreign exchange exposure (FEE) comes
from the international trade and financial activities, such as
foreign loans, guarantees etc. As an example, one big
multinational company buys its raw material in the domestic
market and sells its final product in both domestic and foreign
markets. Assume that the situation in the markets changes, and
as a consequence the foreign currency becomes cheaper in
relation to the domestic one. What will happen to such a
company? If the company cant increase the price, its products
to be sold in the foreign market, will generate less income than
earlier, because the domestic currency as well as the final product,
will become more expensive in comparison with the foreign
currency and prices level. Following the same logic, it is not
difficult to realize, that the foreign competitors of our company
will get the competitive advantage, being able to offer the lower
price for the same product in our domestic market. Therefore,
the company will incur double losses: it will lose part of the
domestic market and part of the foreign market.
Not only big multinational companies, but also small firms having
only domestic trade operations, become increasingly dependant
on the world market main currencies fluctuations. With common
76
Lets start with a simple example, which hopefully will make the
introduction of some main concepts clearer. One Swedish
company that buys raw material in Germany pay in DEM, and has
90 days deferred payment. The companys main activity is in
Sweden, and the biggest part of its cash inflows is in SEK. It is
not difficult to realize that if the DEM suddenly and unexpectedly
increases in price just before the maturity of the payment to the
German supplier, the company incurs losses, as it is forced to pay
more SEK than was expected for the same amount of DEM. In
other words, the company is exposed to DEM price changes. We
arrive at the main definition in this chapter, i.e. exchange rate exposure,
which according to A.C.Shapiro (1996, p. 277) is the degree to
which a company is affected by exchange rate change.
Following the Shapiro way of exchange rate exposure classification
in the coming chapters, we will present it, describing accounting
versus economic exposure and then breaking down exchange rate
exposure into translation, transaction and operating exposures
providing the description of every single one of them.
Accounting practice and economic reality
Accounting exposure arises from the need, for purpose of
reporting and consolidation, to convert the financial statements
of foreign operations from local currencies (LC) involved to the
home currency (HC) (Shapiro, 1996, p. 237). Big multinational
companies usually have foreign subsidiaries and a lot of foreign
operations. As a consequence, foreign currency denominated
assets and liabilities as well as revenues and expenses take place
in their values. However, the investors and the other interested
part of society need values expressed in one currency in order to
get a clear understanding about the companys overall financial
results. Therefore, in accordance with accounting standards at
the end of accounting period (quarter, year) all foreign
subsidiaries values are translated in to HC. Assets and liabilities
might be translated in current (post change) exchange rate and
are considered to be exposed, or at historical (pre-exchange) rate,
11D.571.3
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Translation Exposure
The best definition of Translation exposure we found in
Eitemans book (1997, p. 187). The later follows Shapiros point
of view and states that translation exposure also called accounting
exposure, is the potential for accounting derived changes in
owners equity to occur because of the need to translate
foreign currency financial statements of foreign affiliated into a
single reporting currency to prepare worldwide consolidated
financial statements.
11D.571.3
77
Do
cu
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Transaction exposure
contract, and
foreign currencies.
Now we will go back to the example that we present in the
beginning of this chapter to explain how those exposures fit into
the firms transaction life span. Below (figure No. 2), we present
a basic framework of one business transaction, which starting at
Time 1 ends at Time 4, repeating the cycle again.
78
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
From the above analysis, we can see that both the transaction
exposure and economic exposure focus on the aggregate effect of
both the direct effect and indirect effect. Direct exposure, captured
by transaction and translation exposure, potentially exists whenever
a firm sells or buys its productions or sources in a foreign currency.
Indirect exposure exists when a firm has a supplier, customer or
competitor that is exposed. (Pringle J.1995). The following table
(table 1) shows the effect of the home country currency appreciation
or depreciation. From the table we can see the exchange rate change
impact on the company.
Before the buyer signs the contract with the seller (Time 1), the
seller is already exposed to the risk, even though, in this period,
the exposure is not reflected in the accounting numbers and, at
this moment, the exposure will only be an estimation. Neither
size nor time of the exposure may be known at this time. From
this point of view, some companies will identify the estimated
sales volumes as a transaction exposure, others may treat it as
economic exposure. Usually, firms can estimate their sales volume
on the base of the historical performance. Basic on the transactions
from long-term contracts with permanent customers, companies
forecast the future sales volumes, keeping in mind possible
deviations. The transaction will start in Time 1 and will not end
until the transaction cycle is finished (Time 4). The economic
exposure for one transaction is equal to the transaction exposure,
though from the whole companys value point of view economic
exposure can be broken down into transaction and translation
exposures.
Do
cu
The following figure (figure no. 3) will give us a clear picture of the
relationship between transaction, translation and operating
exposures.
Transaction exposure
Impact of having outstanding obligations, that were set before
change in exchange rates, but to be settled after change in
exchange rate change.
Another important concept of FEE is the time horizon. FEE
might be broken down into short and long-term exposures. Shortterm exposure is related to the cash flow management, while long
79
(price elasticity = 1), we get the estimate that the cash flow will fall
by another SEK 2,037.5. In this case, since we can only estimate
future sales volume and sales amount, it appears that the firm are
faced with the economic exposure and transaction exposure.
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Purpose
Based on the above problem analysis, we found out that both
transaction and operating exposures measure the exchange rate
change effect on the firms cash flows. The main differences
between operating and transaction exposures are the following:
Do
cu
Notes:
80
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
al
81
11D.571.3
Why hedge?
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Today, more and more firms try to manage the exposure through
hedging. Hedging is the taking of a position either acquiring a
cash flow or an asset or a contract (including a forward contract)
that will rise (fall) in value and offset a drop (rise) in value of an
existing position. Therefore, the main purpose of a hedge is to
reduce the volatility of existing position risks caused by the exchange
rate movement (smoother effect). Figure No.4 shows how the
firms expected value E [V] in the home currency looks before and
after hedging. Hedging narrows the distribution of the firms
value about the mean of the distribution. From the figure, we can
see that unless the hedging shifts the mean of distribution to the
right it cant increase the firms value, what means that the hedging
not only protects the purchaser against loss, but also eliminates
any gain that might result from changes in exchange rates. At the
same time hedging is not free; the firm must use their resources to
undertake hedging activity. In order to add value through hedging,
the result must not only shift the mean to the right, but also
needs a net right hand shift given
costs are 17,5% (Altman, 1984) of the companys value oneyear prior the bankruptcy.
al
Do
cu
LESSON 16:
EXPOSURE MANAGEMENT STRATEGIES
82
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
11D.571.3
Non-linear
Convertibles
One can add some other instruments to both of the two classes.
For example, bonds can be viewed as non-linear derivatives with
the interest rate being a non-tradable underlying. During the course
we will talk about each of the listed above types of derivative
products. Although the main goal of my course is not to teach
how to use various derivatives, but rather how to price them, I
will try to explain the most common applications of some of the
derivatives.
Forwards and Futures
Forward is a contract between two parties agreeing that at
certain time in the future one party will deliver a pre-agreed
quantity of some underlying asset (or its cash equivalent in the
case of non-tradable underlyings) and the other party will pay a
pre-agreed amount of money for it. This amount of money is
called the forward price. Once the contract is signed, the two
parties are legally bound by its conditions: the time of delivery,
the quantity of the underlying and the forward price.
While the delivery time and the delivery quantity of the underlying
asset can be fixed without any problem, the question is how the
parties can agree on the future price of the underlying when the
83
At the end of the year, we will deliver the asset to the buyer of the
forward contract who will pay us the forward price n X F. From
this amount we have to repay the bank our loan which obviously
grew to (1 + r) X S, where r is the one-year interest rate quoted by
our bank. Thus, at the end of the year our cash flow is]
F (1+r) X S
Since we started with no money, we have to end with no money.
Otherwise, by selling or buying forward contracts we would be
able to make unlimited profit without taking any risk. This is not
possible in practice. Therefore, we have to impose the following
constrain
F (1+r) X S = 0
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
At the end of the third day, the seller of the futures contract is
better off by 2.
Forwards and futures are designed to reduce risks related to the
uncertainty of future market prices for both sellers and buyers of
underlying assets. By entering this type of contracts, both sides
achieve complete certainty about their future positions, which may
help them to have a better control over their financial resources.
However, many traders take futures positions for purely speculative
reasons. For instance, if we sell an uncovered futures contract (i.e.,
when we do not have a long position in the underlying asset),
then when the asset price goes down, our futures position will
gain a profit and vice versa.
In what follows we will be using the following definitions. A
long position in an asset is a position that benefits from price
increases in that security (an investor who buys a share has a long
position, but an equivalent long position can also be established
with derivatives). A short position benefits from price decreases
in the security. A short position is often established through a
short sale. To sell a security short, one borrows the security and
sells it. When one unwinds the short sale, one has to buy the
security back in the market to return it to the lender. One then
benefits from the short sale if the assets price is lower when one
buys it back than it was when one sold it.
Swaps
Swaps, as the name suggests, are instruments, which allow a
swap holder to receive a floating interest rate from and pay a
fixed interest rate to a swap seller for a certain period of time.
The interest rates are paid on the same fixed notional principal.
84
11D.571.3
Put options give the right to sell the underlying asset. Calls give
the right to buy the underlying assets. There exist also chooser
options, when the option holder has the right to chose between
call and put payoffs.
There are hundreds of different types of options, which differ in
their payoff structures, path-dependence, and payoff trigger and
termination conditions.
al
Bonds
Bonds are securities, which pay a certain fixed amount on a
certain fixed date in the future. Since we know how much we
will get on some future date, we can find the present value of
the notional by discounting this amount to the present time
with respect to a certain interest rate. If the rate was known in
advance, the price of the bond would be very easy to calculate.
For example, if the rate is fixed and equal to 5% per annum,
then the one-year bond with the notional value 1000 should
now cost
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Where
is the interest earned from the time T i 1 to T i
discounted to the present time.
Taking into account that B (t, T 0) = 1, we find All in all, the value
of the swap is given by We can now see that a swap from the
floating-rate receiver side can be presented as a combination of
short positions in bonds with different maturities. Normally, the
fixed rate is chosen so that the swap present value is equal to zero.
Swaps are extremely liquid instruments and, therefore, their market
values can be used to price bonds.
Options
Options are the most flexible of all derivatives because they give
an option holder a multiple choice at various moments during
the lifetime of the option contract. However, an option seller
does not have such flexibility and always has to fulfill the
option holders requests. For this reason, the option buyer has
to pay a premium to the option seller.
11D.571.3
85
Practical strategies
Pricing strategy
The case that we presented in chapter one explained how the
exchange rate change affect the firms cash flow. Pricing strategy
and demand sensitivity to competitors price are two important
factors, which affect the firms exchange exposure. Therefore, it
would be logical to presume that if we set a flexible pricing
strategy, then the firm can handle the exchange rate exposure
easily. However thats not always the case. As a matter of fact,
some industries such as chemical, petroleum and mining
businesses have few pricing decisions to make relative to the
currency risk, since those industries are very large depend on
economies of scale which means they are pricing taker instead
of pricing setter. For example, in the SKF case, the company
whose activity we are going to analyze later, the buyer not the
seller dictates the price. Additionally, there still exists some costs
associated with pricing changing policy; such as: long term
customer relationship, the customers loyalty to the firm, and so
on.
al
Netting
An alternative method to the previous one is to use a netting
system. This system is often based on a re-invoice center
establishment, where each separate subsidiary deals only with its
own currency, leaving all the transaction exposure to re-invoicing
center. There are some advantages of re-invoice center:
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
It is easy to control the overall firms activity when all the currency
Do
cu
Diversification
From above mentioned Pringles analysis, we may get an
impression that the firms can manage the currency exposure
through diversification of both operating and financial policies.
From the first sight, we may say that diversification of both
strategies gives a lot of choices. The firm can diversify its
operations through, such branches of its activity as, sales,
location of production facilities, raw material sources, while
financial policy diversification can be done using funds in more
than one capital market and in more than one currency.
However, its not always an easy way. Some industry may require
large economies of scale that it are not feasible to diversify its
production location, maybe some firm are too small to be
known by the international investors or lenders. Thereby,
especially operating strategys diversification can be used mostly
depend on the firms characteristic. In the later case study, we
will look if this kind of strategy is feasible.
Notes:
11D.571.3
LESSON 17:
CASE STUDY IN SKF
Overview of SKF
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
87
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
88
11D.571.3
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
(October+November+December)=(13000+12000+20000)*4=170
000
The performance for last six months:(July,August,,December)
=(9000+17000+18000+13000+12000+20000)*2=178 000
Do
cu
11D.571.3
89
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Expenses are made and the risk of incurring losses appears, creating
the credit risk, in the case the buyer wont pay the bill. To our
knowledge, during Time 1-Time 2, SKF is making only some
kind of interest and credit exposures management, but since
interest rate and credit risk exposure are out of our concern we are
not going to go deeper into that topic. At Time 3 SKF ships the
goods and bills the buyer. Starting from that moment the amount
of cash flow receivables appears in the SKF accounting books. All
above described exposures, which appeared at Time 1 and Time 2
increase by shipment and any Other cost related with the
transporting (in the case the SKF is paying).
Do
cu
90
11D.571.3
Risk sharing
Do
cu
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
91
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Notes:
Do
cu
92
11D.571.3
LESSON 18:
CASE STUDY IN ELOF HANSSO
Overview of Elof Hansson
Elof Hansson is a trading company with an even longer history
than the SKFs one. The company was founded by the
merchant Elof Hansson in 1897. The Board of Directors and
Managing Director gives the following presentation of the
company (in annual report of the fiscal year 1999):
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
11D.571.3
All above given and the other numbers presented in this chapter
93
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
94
11D.571.3
al
Do
cu
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
95
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
96
In SKF the biggest part of the bearings are produced from the
About 50% of SKFs contracts are long term with long standing
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
The hedging period, especially for SKF, has one more aspect. The
company is hedging every three-months. So, at the end on the
hedged period they are faced with the big risk of being exposed to
exchange rate changes, if any. Therefore, the conclusion and
suggestion at the same time is that, in the case of SKF, they
would have less risk if they were to hedge every month.
Do
cu
At this stage, the first important step for the firm is to confirm
the starting moment of the hedging strategy, based on the
firms operating characteristics. Since SKF has a big percentage
of long-term contracted relationship, it becomes very important
to set a certain hedging percentage in the quotation period and
use 100 percent hedging after the contract signing. Otherwise,
they are exposed to exchange rate risk during these periods.
Since EH has small percentage of fixed long term contracted
relationships, the hedging after the contract (order) has been
confirmed is the most important one. The possibility to
forecast future cash flows before that moment is very small.
Thats why the hedging before the contract is placed might
turned out to be more costly then useful, although in theory
the company is incurring risk from Time 1 to Time 2. So in this
sense, different firms might have different most fitting starting
moment of hedging depending on this net of specific features.
11D.571.3
97
starts to hedge when the seller ships products and bills the buyer
(Time 3) and EHs starting point is when the buyer places a contract
(order) with the seller (Time 2).
Thus far, we have only been able to place a lower and upper
bound around the value of an option prior to its expiration. To
produce an exact formula, we will need to make specific
assumptions about the way the underlying asset price and
riskless return evolve over time.
Do
cu
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
LESSON 19:
OPTION PRICING MODELS
For European calls and puts, while the lower bounds must be
loosened, the upper bounds can be tightened:
Sd -t C max [0, Sd -t - Kr-t]
Kr-t P max [0, Kr-t - Sd -t]
11D.571.3
Do
cu
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
99
[(log u)2 + (log d)2 + (log u)2 + (log u)2 + (log d)2]/5 = (log u)2
The sample variance of the second path is:
al
[(log d)2 + (log d)2 + (log d)2 + (log u)2 + (log u)2]/5 = (log u)2
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
rule this out, in effect, we are assuming that volatilities (1) and (2)
are the same.
Moreover, investors all think the next up and down moves at
every node in the tree will be the same everywhere in the tree,
and that u = 1/d. Thus log u = - log d, so that (log u)2 = (log
d)2. This means that along any path in the tree the sampled
(logarithmic) volatility around a zero mean will be the same. For
example, consider two paths in a five move tree: u, d, d, u, d and
d, d, u, u, u. The sample variance of the first path is:
100
11D.571.3
Gamma
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Delta
Graph #1
Do
cu
Vega
Graph #2
11D.571.3
101
rate. In reality there is no such thing as the risk-free rate, but the
discount rate on U.S. Government Treasury Bills with 30 days left
until maturity is usually used to represent it. During periods of
rapidly changing interest rates, these 30-day rates are often subject
to change, thereby violating one of the assumptions of the model.
Graph #6
al
Graph #3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Graph #4
Graph #7
Graph #5
Do
cu
Graph #8
102
11D.571.3
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Volatility
The volatility figure input into an option-pricing model reflects
the assumptions of the person using the pricing model.
Volatility is defined technically in various ways, depending on
assumptions made about the underlying assets price
distribution. For the regular option trader it is sufficient to
know that the volatility a trader assigns to a stock reflects
expectations of how the stock price will fluctuate over a given
period of time.
Hedging
Delta measures the sensitivity of an option value, ceteris
paribus, to a small change in its underlying asset price. So it
makes sense to calculate the delta by taking the first partial
derivative of the option value, as expressed by the Black-Scholes
formula, with respect to the underlying asset price. Other
hedging parameters including gamma and vega, can also be
derived from the Black-Scholes formula by taking the
appropriate partial derivatives.
Do
cu
11D.571.3
103
Graph #9
Forward Contracts
al
Introduction
Despite the recent adverse press they have received, derivative
securities provide a number of useful functions in the areas of
risk management and investments. In fact, derivatives were
originally designed to enable market participants to eliminate
risk. A wheat farmer, for example, can fix a price for his crop
even before it is planted, eliminating price risk. An exporter can
fix a foreign exchange rate even before beginning to
manufacture the product, eliminating foreign exchange risk. If
misused, however, derivative securities are also capable of
dramatically increasing risk.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
LESSON 20:
HEDGING WITH FORWARD / FUTURE CONTRACTS
Do
cu
F = S0 e(q+r)T
104
Position
-S 0
ST
-S0 (eqT - 1 )
Borrow
S 0 eqT
-S0 e(q+r)T
F - ST
F - S 0 e (q+r)T
11D.571.3
(e0.06x0.5
e(0.06+0.05)T
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Position
Do
cu
105
Margin
Although futures contracts require no initial investment, futures
exchanges require both the buyer and seller to post a security
deposit known as margin. Margin is typically set at an amount
that is larger than usual one-day moves in the futures price. This
is done to ensure that both parties will have sufficient funds
available to mark to market. Residual credit risk exists only to
the extent that (1) futures prices move so dramatically that the
amount required to mark to market is larger than the balance of
an individuals margin account, and (2) the individual defaults
on payment of the balance. In this case, the exchange bears the
loss so that participants in futures markets bear essentially zero
credit risk. Margin rules are stated in terms of initial margin
(which must be posted when entering the contract) and
maintenance margin (which is the minimum acceptable
balance in the margin account). If the balance of the account
falls below the maintenance level, the exchange makes a margin
call upon the individual, who must then restore the account to
the level of initial margin before the start of trading the
following day. Below Example illustrates the margining
procedure.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
106
Example: Margin
Suppose a contract requires initial margin of $7,000 and maintenance margin of $5,000. The following table
illustrates the margining procedure and the cash flows required for the buyer of a futures contract.
Value
Time Futures
Contract
of Margin
Margin
Balance before
Call
Calls
Margin
Balance after
Calls
25,000
7,000
7,000
24,000
6,000
6,000
22,000
4,000
3,000
7,000
24,500
7,000
7,000
Note that when the margin balance falls below the maintenance margin, it must be restored to the initial level.
Note also that when the future moves favorably (as at time 3) the marking to market cash inflow can be
immediately withdrawn - it need not remain in the margin account.
0 1
... T
0 0
... S T-FO 0
0 FU1 -FU 0
FU 2-FU1
... F UT - FUT- 1
11D.571.3
where we note that FUT = S T. The payoff on the bond is FU0 e rT.
Therefore, the overall initial investment required for this strategy
is $FU0 and the overall payoff at time T is S 0erT.
Now consider the strategy of buying erT forward contracts on day
o and investing $FO 0 in a riskless bond (where FO represents the
price of a forward contract). The overall initial investment required
for this strategy is $FO 0 and the overall payoff at time T is:
erT(ST - FO0) + FO0 erT = S0 erT
Since both of these strategies have the same payoff, they must
cost the same. That is FO0 = FU0. The following table illustrates
the cash flows associated with the two strategies.
Forward-Futures Equivalent
Time
Net Cash Flow
r
The interest rate (annual continuously compounded TBill rate)
d
The dividend yield on the index (continuously
compounded annual rate)
T
This is the same as equation (1) except that +q has been replaced
by -d as the cost of carry (storing wheat) has been replaced by a
benefit (dividends). To see why this relationship must hold,
consider the strategy of (1) borrowing e-dTS0 through time T, (2)
using this to purchase e-dT units of the index and reinvesting all
dividends back into the index, and (3) selling a futures contract
that matures at time T. If interest rates are constant, the futures
contract is equivalent to a forward contract, which simplifies the
analysis. In particular, the (equivalent) cash flows associated with
this strategy are tabulated in the following table. Note that
reinvestment of the dividends has resulted in the initial investment
of e-dT units of the index growing at a rate of d to amount to one
unit by maturity.
Two Examples given below illustrate how to execute a riskless
arbitrage if this equality does not hold.
Arbitrage Relationship Between Spot and Futures Contract
Position
Time 0
Time T
Borrow
e-dTS0
-e-dTS0
ST
F - ST
Net Position
F - S0e( r- dT)
e2r
Do
cu
STer T
er
T
Contract erT (ST
FO 0erT
FU 1-FU0
-(FU1-FU0 )
er(t- 1)
FO0) erT(S T
FU 0)
erT[FU0 + (FU 1 - F U0) + ... + (FUT-1 - F UT- 2)]
S TerT
Arbitrage Relationships
For the remainder of this module, we assume that interest rates
are indeed constant over the period of the contract and hence
the futures price equals the forward price. That is, we can
consider the price and payoffs of a futures contract to be
identical to those of a forward contract. This simplifies things
because a forward contract has only a single payoff at maturity.
Consider, for example, the valuation of a futures contract on
the S&P 500 stock index. This contract, which trades on the
Chicago Mercantile Exchange (CME) entitles the buyer to receive
the cash value of the S&P 500 stock index at the end of the
contract period. There are always four contracts in effect at any
one time expiring in March, June, September, and December. In
11D.571.3
-FU 0
er t
FU t-FU t-1
-(FUt-FU t- 1)
S0
-FO 0
Time:
t
Number of Contracts Purchased 0
Cash Flow from Contract 0
Investment in Bonds
0
Time:
Cash Flow from
Payoff from Bonds
Time:
1
Number of Contracts Purchased 0
Cash Flow from Contract 0
Investment in Bonds
0
Net Cash Flow
where
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
2r
F = S 0 e (r-d)T
al
Once again, since this strategy requires no initial cash outlay, the
cash flow at maturity must also be zero or an arbitrage opportunity
exists. In particular, if F > S 0 e(r-dT) the strategy of buying the index
and selling the futures generates an arbitrage profit. Conversely, if
F < S 0 e(r-dT) the strategy of selling the index and buying the futures
generates an arbitrage profit.
Two Examples given below illustrate how to execute a riskless
arbitrage if this equality does not hold.
Example: Futures arbitrage: Buy index - Sell futures
Suppose the S&P 500 stock index is at $295 and the six-month
futures contract on that index is at $300. If the prevailing T-Bill
rate is 7% and the dividend rate is 5%, an arbitrage opportunity
exists because F=300 > S e(r-d)T = 297.96. The arbitrage can be
executed by buying low and selling high. In this case, the futures
contract is relatively overvalued, so we sell the futures and buy the
index.
In particular, the strategy is to
107
Position
Time 0
Time T
Borrow
287.72
-297.96
-287.72
ST
300 - S T
Net Position
2.04
Short sell e-dT units of the S&P index generating Se-dT = 292.59.
Time 0
Time T
292.59
-S T
-292.59
303.02
ST -300
3.02
Do
cu
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Use this $287.72 to buy e-dT = 0.975 units of the S&P index, and
reinvest all dividends in the index. Sell a futures contract for delivery
of the index in six months.
First note that no money changes hands at the time you buy the
contract. This is the nature of all futures contracts. The November
15 price of 94.86 implies that the LIBOR rate of interest was 10094.86=5.14% at that time. If LIBOR falls 100 basis points by the
time the December contract expires, LIBOR will then be 4.14%.
Therefore, the expiration futures price will be 100-4.14=95.86.
The total gain is therefore:
0.25 (1,000,000)(Ft-F0) = 0.25 (1,000,000)(0.9586-0.9486) = $2,500
That is, to settle the contract, your counter party will give you
$2,500.
Example given below contains a detailed illustration of how the
Eurodollar futures contract can be used to hedge interest rate risk.
Example: Hedging with the Eurodollar Futures Contract.
It is currently November 15 and your company is aware that it
needs to borrow $1 million on December 16 to pay a liability,
which falls due on that day. The loan can be repaid on March 16
when an account receivable will be collected. The current LIBOR
rate is 5.14%. Your company is concerned that interest rates will
rise between now and December 16, in which case you will pay a
higher rate of interest on your loan. How can your company lock
in the current rate of 5.14%?
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
If we could lock in the rate of 5.14%, the total interest on the loan
would be 0.0514($1 million)/4 = $12,850.
First, suppose that on December, 16 LIBOR is 6.14%. Interest
on the loan will be
0.0614($1 million)/4 = $15,350, and the gain on the futures
position will be -10000(93.86-94.86)/4 = $2,500. This yields a
net cash outflow of -$15,350+$2,500 = -$12,850, which is the
same as 3-months interest on $1 million at 5.14%.
Now suppose that on December, 16 LIBOR is 4.14%. Interest
on the loan will be
0.0414($1 million)/4 = $10,350, and the gain on the futures
position will be -10000(95.86-94.86)/4 = -$2,500. This yields a
net cash outflow of -$10,350-$2,500 = -$12,850, which is the
same as 3-months interest on $1 million at 5.14%.
First note that at the December futures price of 383.50, the return
on the index, since the beginning of the year, is 383.5/306.80-1 =
25%. If the manager is able to lock in this return on his fund, the
value of the fund will be 1.25($76.7 million) = $95.875 million.
Since the notional amount underlying an S&P 500 futures contract
is 500(383.50) = $191,750, the manager can lock in the 25% return
by selling 95,875,000/191,750=500 contracts. To illustrate that
this position does indeed form a perfect hedge, we examine the
net value of the hedged position under two scenarios.
al
Do
cu
First, suppose the value of the S&P 500 index is 303.50 at the end
of December. In this case, the value of the fund will be (303.50/
383.50)95.875 million = 75.875 million. The gain on the futures
position will be -500(500)(303.50-383.50) = 20 million. Hence the
total value of the hedged position is 75.875+ 20 = 95.875 million,
locking in a 25% return for the year.
Now suppose that the value of the S&P 500 index is 403.50 at the
end of December. In this case, the value of the fund will be
403.50/383.50(95.875 million) = 100.875 million. The gain on the
futures position will be -$500(500)(403.50-383.50) = -5 million.
Hence the total value of the hedged position is 100.875-5=95.875
million, again locking in a 25% return for the year.
Hedging Foreign Exchange Risk
Another source of risk that an individual or organization may
wish to hedge is foreign exchange risk. For example, a person
who will be traveling overseas in the coming months may wish
to hedge the value of the amount of money he intends to
spend abroad against a devaluation of his domestic currency
relative to the foreign currency. An exporter who sells goods
overseas on credit may wish to hedge against a devaluation of
the foreign currency in which payment occurs.
A number of foreign currency futures contracts trade on the
International Monetary Market division of the Chicago Mercantile
Exchange. The currencies on which contracts are based, and the
underlying notional amount are listed in the following Table.
Delivery months for all contracts are March, June, September, and
December. Prices are quoted as US dollars per unit of foreign
currency. For example, if one Swiss franc buys 69.15 US cents, the
price will be quoted as 0.6915.
Denomination of Foreign Currency Futures Contracts
The gain on your futures position is $500(Ft-F0) = $500(393.50383.50)=$5,000. That is, to settle the contract, your counter party
will give you $5,000.
Example given below contains a detailed illustration of how the
S&P 500 futures contract can be used to hedge stock market risk.
Example: Hedging with the S&P 500 Futures Contract.
Currency
Underlying Amounts
British Pound
62,500 L
Canadian Dollar
100,000 C$
German Mark
125,000 DM
Japanese Yen
12,500,000 Y
Swiss Franc
125,000 SF
French Franc
250,000 FF
Australian Dollar
125,000 A$
109
with Treasury bond futures; basis risk arises due to the uncertainty
of the yield differential at the time the hedge is lifted.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Note that since (1) the total exposure is one million Swiss Francs
and (2) each futures contract is for 125, 000 Francs, eight contracts
are required to hedge the exposure. Further, since (1) the company
stands to lose if the Swiss Franc depreciates (each Swiss Franc can
be converted back into a smaller number of Dollars) and (2) the
futures contracts decrease in value if the Swiss Franc depreciates
(since the basis of the contract is Swiss Francs per Dollar), the
contracts should be sold.
al
Do
cu
From the above equation, the net revenue from the hedge position
is composed of (1) a certain component that depends upon the
futures price at the time of the hedge (F0) and (2) an uncertain
component that depends upon the difference between the price
received for bonds in the spot market and the futures price at
closeout (BT-FT). The difference between the spot and the futures
price is called the basis. Thus, uncertainty about the net-hedged
revenue arises if there is uncertainty about the basis. To quote
Holbrook Working, hedging is speculation in the basis.
There are many reasons for the basis to be
uncertain.
First, the good or instrument being hedged may be different
from the good or instrument for which there is a futures contract.
This would be the case if a corporate bond offering is hedged
110
Fourthly, with most futures contracts, the seller has the choice of
the date of delivery within the delivery month. This choice is an
uncertain value and thus contributes to basis risk.
Finally, the mark to market aspect of futures results in hedging
risk. The uncertainty is about the amount of interest earned or
forfeited due to the daily transfers of profits and losses. In fact,
the equations for net revenue are not exactly right due to the
omission of interest earned (lost) on futures profits (losses).
The Volatility of Futures
A common mistake made is to assume that futures are much
more volatile than stocks. Percentage changes of futures prices
are generally less volatile than the percentage changes of a typical
stock. Annualized standard deviations for most futures
contracts are in the 15-20% range whereas a typical stocks is
about 30%.
There is no reason that the futures should be played in a high-risk
manner by a large investor. Of course, if the futures investor does
not have enough capital (5-8 times margin), then he is required to
play with considerable leverage or not at all. Before taking great
leverage, the small investor should consider looking at a smaller
contract (grain on CBT is 5,000 bushels whereas Mid-America
contract is 1,000 bushels).
The effect of leverage is to increase volatility. Borrowing to meet
the margin requirements will increase gains but also increase losses.
Setting aside larger amounts of capital, which are invested, in a
safe asset will decrease the volatility.
Risk in the Futures Markets
As we have already seen, one the most important applications
of the futures is for hedging. Futures contracts were initially
introduced to help farmers that did not want to bear the risk of
price fluctuations. The farmer could short hedge in March (agree
to sell his crop) for a September delivery. This effectively locks in
the price that the farmer receives. On the other side, a cereal
company may want to guarantee in March the price that it will
pay for grain in September. The cereal company will enter into a
long hedge.
11D.571.3
F = S0 e(r-d)T
The price of a forward contract when there is a dividend benefit d.
When interest rates are constant, the same relationship holds for
a futures contract.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Notes:
The most obvious case is when you are trying to hedge a cash
position with futures positions in different instruments. This is
the case that we introduced in one of the first lectures when we
hold the Ginnie Mae security and want to hedge this security with
a combination of T-Bonds and Euros. It is unlikely, however,
that at the expiration of the futures contract, the cash price of the
T-Bond and Euros will equal the Ginnie Mae. This is the basis
risk.
Do
cu
11D.571.3
111
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
LESSON 21:
OTHER DERIVATIVES CONTRACTS
Do
cu
NeedOil can obtain its desired payoff by buying a call option with
an exercise price of $15 and selling a call option with an exercise
price of$18. To see this, you simply need to graph the payoff on
each option separately and then vertically add the payoffs. Figure
given below illustrates the payoffs from the two options with
dashed lines.
112
11D.571.3
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Equity Swaps
An equity swap is a contract providing for the delivery of cash
flows associated with shares of equity (or an equity index) in
exchange for the cash flows associated with another asset (such
as a debt or index instrument). For example, an investor
wishing to relieve himself of risk associated with shares he is
currently holding, without selling and exposing himself to
capital gains tax liability, may agree to deliver to another investor
the cash flows (dividends and capital gains for a specified
period) associated with his shares. The second investor, in turn,
agrees to deliver cash flows associated with a treasury bond to
the stock investor. Equity swaps are used to exploit apparent
mis-pricing in equity markets, to manage risks associated with
domestic or foreign equity investment, to circumvent dividend
withholding tax requirements in foreign countries and to
speculate in foreign equity markets when direct ownership is not
permitted.
Do
cu
11D.571.3
113
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
received by the exporter will vary. The Asian option enables the
exporter to stabilize its cash flows without entering the derivatives
market on a daily basis. The cash flow structures of these options
vary from contract to contract. For example, some contracts call for
the payoff to be related to the difference between the time T spot
rate and the average exchange rate realized during the life of the
option.
An Interest Rate Cap pays its owner a value based on the difference
between the market rate and the cap strike rate if the market rate
rises above the strike rate. A Swaption gives its owner the right
(but not the obligation) to enter into a swap arrangement at a later
date.
Markets for Derivatives
Do
cu
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
Commodity Derivatives
Derivative contracts on agricultural commodities have existed
for a long time. For example, the Chicago Board of Trade has
traded futures contracts since 1865, and forwards and op-tions
on agricultural products date back several centuries. Users and
producers of com-modities such as metals and oil also
frequently trade both OTC and exchange-traded derivatives. The
use of electricity derivatives also has grown significantly in recent
years due in part to deregulation of the industry.
Equity Derivatives
Equity derivatives are contracts derived from stock market
indexes like the Standard & Poors 500. Futures contracts exist
that are based on US stock market indexes and on for-eign stock
market indexes, such as the Nikkei index for the Japanese stock
market. In ad-dition, options have traded on individual stocks
for some time. The notional principal on futures and options
in 2002 equaled about $2.2 trillion.
Notes:
11D.571.3
115
LESSON 22:
MEASURING & HEDGING INTEREST
RATE RISK IN BANKS
Introduction
Interest rate risk is the risk to earnings or capital arising from
movement of interest rates. It arises from differences between
the timing of rate changes and the timing of cash flows (re
pricing risk); from changing rate relationships among yield
curves that affect bank activities (basis risk); from changing rate
relationships across the spectrum of maturities (yield curve
risk); and from interest-rate-related options embedded in bank
products (option risk). The evaluation of interest rate risk must
consider the impact of complex, illiquid hedging strategies or
products, and also the potential impact on fee income that is
sensitive to changes in interest rates.
servicing fees.
Do
cu
116
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Chapter Objective
UNIT I
CHAPTER 8
CREDIT & INTEREST RATE RISKS
Banks whose re-pricing asset maturities are longer than their repricing liability maturities are said to be liability sensitive, because
their liabilities will re-price more quickly. The earnings of a liabilitysensitive bank generally increase when interest rates fall and decrease
when they rise. Conversely, an asset sensitive bank (asset re-pricings
shorter than liability re-pricings) will generally benefit from a rise
in rates and be hurt by a fall in rates.
Re-pricing risk is often, but not always, reflected in a banks current
earnings performance. A bank may be creating re-pricing imbalances
that will not be manifested in earnings until sometime into the
future. A bank that focuses only on short-term re-pricing
imbalances may be induced to take on increased interest rate risk
by extending maturities to improve yield. When evaluating repricing risk, therefore, it is essential that the bank consider not
only near term imbalances but also long-term ones. Failure to
measure and manage material long-term re-pricing imbalances
can leave a banks future earnings significantly exposed to interest
rate movements.
Basis Risk
Basis risk arises from a shift in the relationship of the rates in
different financial markets or on different financial instruments.
Basis risk occurs when market rates for different financial
instruments, or the indices used to price assets and liabilities,
change at different times or by different amounts. For example,
11D.571.3
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
balance-sheet positions needs to consider how the off-balancesheet contracts cash flows may change with changes in interest
rates and in relation to the positions being hedged or altered.
Derivative strategies designed to hedge or offset the risk in a
balance sheet position will typically use derivative contracts
whose cash flow characteristics have a strong correlation with the
instrument or position being hedged. The bank will also need
to consider the relative liquidity and cost of various contracts,
selecting the product that offers the best mix of correlation,
liquidity, and relative cost. Even if there is a high degree of
correlation between the derivative contract and the position
being hedged, the bank may be left with residual basis risk
because cash and derivative prices do not always move in
tandem. Banks holding large derivative portfolios or actively
trading derivative contracts should determine whether the
potential exposure presents material risk to the banks earnings
or capital.
Certain pricing indices have a built-in lag feature such that the
index will respond more slowly to changes in market interest
rates. Such lags may either accentuate or moderate the banks shortterm interest rate exposure. One common index with this feature
is the 11th District Federal Home Loan Bank Cost of Funds
Index (COFI) used in certain adjustable rate residential mortgage
products (ARMs). The COFI index, which is based upon the
monthly average interest costs of liabilities for thrifts in the 11th
District (California, Arizona, and Nevada), is a composite index
containing both short and long-term liabilities. Because current
market interest rates will not be reflected in the index until the
long-term liabilities have been re-priced, the index generally will
lag market interest rate movements.
Do
cu
117
Risk Measurement
Accurate and timely measurement of interest rate risk is
necessary for proper risk management and control. A banks risk
measurement system should be able to identify and quantify
the major sources of the banks interest rate risk exposure. The
system also should enable management to identify risks arising
from the banks customary activities and new businesses. The
nature and mix of a banks business lines and the interest rate
risk characteristics of its activities will dictate the type of
measurement system required. Such systems will vary from
bank to bank.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
11D.571.3
al
Aggregation
The amount of data aggregated from transaction systems for
the interest rate risk model will vary from bank to bank and
from portfolio to portfolio within a bank. Some banks may
input each specific instrument for certain portfolios. For
example, the cash flow characteristics of certain complex CMO
or structured notes may be so transaction-specific that a bank
elects to model or input each transaction separately. More
typically, the bank will perform some preliminary data
aggregation before putting the data into its interest rate risk
model. This ensures ease of use and computing efficiency.
Although most bank models can handle hundreds of
accounts or transactions, every model has its limit. Because
some portfolios contain numerous variables that can affect their
interest rate risk, additional categories of information or less
aggregated information may be required. For example, banks
with significant holdings of adjustable rate mortgages will need
to differentiate balances by periodic and lifetime caps, the reset
frequency of mortgages, and the market index used for rate
resets. Banks with significant holdings of fixed rate mortgages
will need to stratify balances by coupon levels to reflect
differences in prepayment behaviors.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Gathering Data
The first step in a banks risk measurement process is to gather
data to describe the banks current financial position. Every
measurement system, whether it is a gap report or a complex
economic value simulation model, requires information on the
composition of the banks current balance sheet. In modeling
terms, gathering financial data is sometimes called providing
the current position inputs. This data must be reliable for the
risk measurement system to be useful. The bank should have
sufficient management information systems (MIS) to allow it
to retrieve appropriate and accurate information in a timely
manner. The MIS systems should capture interest rate risk data
on all of the banks material positions, and there should be
sufficient documentation of the major data sources used in the
banks risk measurement process. Bank management should be
alert to the following common data problems of interest rate
risk measurement systems:
branches.
Lack of information on off-balance-sheet positions and on
caps and floors incorporated into bank loan and deposit
products.
Inappropriate levels of data aggregation.
Information to Be Collected
To describe the interest rate risk inherent in the banks current
position, the bank should have, for every material type of
financial instrument or portfolio, information on:
Do
cu
Sources of Information
11D.571.3
119
of assumptions.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
often move more slowly than market rates, including rates such as
the banks prime rate, and rates it pays on consumer deposits.
al
Banks should use interest rate scenarios with at least a 200-basispoint change-taking place in one year. Since 1984, rates have twice
changed that much or more in that period of time. The OCC
encourages banks to assess the impact of both immediate and
gradual changes in market rates as well as changes in the shape of
the yield curve when evaluating their risk exposure. The OCC also
encourages banks to employ stress tests that consider changes
of 400 basis points or more over a one-year horizon. Although
such a shock is at the upper end of post-1984 experience, it was
typical between 1979 and 1984. Banks with significant option risk
should include scenarios that capture the exercise of such options.
For example, banks that have products with caps or floors should
include scenarios that assess how the banks risk profile would
change should those caps or floors become binding. Some banks
write large, explicitly priced interest rate options. Since the market
value of options fluctuates with changes in the volatility of rates
as well as with changes in the level of rates, such banks should
also develop interest rate risk assumptions to measure their
exposure to changes in volatility.
Do
cu
rates.
The bank also must estimate how rates that are administered or
managed by bank management (as opposed to those that are
purely market driven) might change. Administered rates, which
120
11D.571.3
behavior.
strategies.
Do
cu
11D.571.3
al
interest rate risk exposure. Banks that have not updated risk
measurement techniques for changes in business strategies and
products or acquisition and merger activities can experience this
problem.
Bank management does not understand the models methods
and assumptions. Banks that purchase a vendor model and
fail to obtain current user gudes and source documents that
describe the models implied assumptions and calculation
methods may misinterpret model results or have difficulties
with the measurement system.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Only one person in the bank is able to run and maintain the
121
Risk Monitoring
Interest rate risk management is a dynamic process. Measuring
the interest rate exposure of current business is not enough; a
bank should also estimate the effect of new business on its
exposure. Periodically, institutions should reevaluate whether
current strategies are appropriate for the banks desired risk
profile. Senior management and the board should have
reporting systems that enable them to monitor the banks
current and potential risk exposure and to ensure that those
levels are consistent with their stated objectives.
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
11D.571.3
exposure.
Do
cu
Risk Control
A banks internal control structure ensures the safe and sound
functioning of the organization generally and of its interest rate
risk management process in particular. Establishing and
maintaining an effective system of controls, including the
enforcement of official lines of authority and appropriate
separation of duties, is one of managements more important
responsibilities. Persons responsible for evaluating risk
monitoring and control procedures should be independent of
the function they review. Key elements of the control process
include internal review and audit and an effective risk limit
structure.
Auditing the Interest Rate Risk Measurement
Process
Banks need to review and validate each step of the interest rate
risk measurement process for integrity and reasonableness. This
review is often performed by a number of different units in the
organization, including ALCO or treasury staff (regularly and
routinely), and a risk control unit that has oversight
11D.571.3
Among the items that an audit should review and validate are:
The appropriateness of the banks risk measurement system(s)
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
123
Risk Limits
The banks board of directors should set the banks tolerance
for interest rate risk and communicate that tolerance to senior
management. Based on these tolerances, senior management
should establish appropriate risk limits that maintain a banks
exposure within the boards risk tolerances over a range of
possible changes in interest rates. Limit controls should ensure
that positions that exceed predetermined levels receive prompt
management attention. A banks limits should be consistent
with its overall approach to measuring interest rate risk and
should be based on its capital levels, earnings performance, and
risk tolerance. The limits should be appropriate to the size,
complexity, and capital adequacy of the bank and address the
potential impact of changes in market interest rates on both
reported earnings and the banks economic value of equity
(EVE).
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Pricing policies may also control the creation of interest rate risk
exposures and internal funds transfer pricing systems. Funds
transfer systems typically require line units to obtain funding prices
from the banks treasury unit for large transactions. Those funding
prices generally reflect the cost that the bank would incur to hedge
or match-fund the transaction.
Examiners should identify and evaluate the types of limits the
bank uses to control the risk to earnings and capital from changes
in interest rates. In particular, the examiner should determine
whether the risk limits are effective methods for controlling the
banks exposure and complying with the boards expressed risk
tolerances. The examiner also should assess the appropriateness
of the level of risk allowed under the banks risk limits in view of
the banks financial condition, the quality of its risk management
practices and managerial expertise, and its capital base.
Earnings-At-Risk Limits
Earnings-at-risk limits are designed to control the exposure of a
banks projected future reported earnings in specified rate
scenarios. A limit is usually expressed as a change in projected
124
11D.571.3
LESSON 23:
INTEREST RATE RISK MODELS
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Gap Reports
Gap reports are commonly used to assess and manage interest
rate risk exposure specifically, a banks repricing and maturity
imbalances. However, as explained later in this appendix, a basic
gap report can be an unreliable indicator of a banks overall
interest rate risk exposure. Although a simple gap report does
not identify and quantify basis risk, yield curve risk, and option
risk, bankers have modified gap reports to do so. Gap reports
stratify all of a banks assets, liabilities, and off-balance-sheet
instruments into maturity segments (time bands) based on the
instruments next repricing or maturity date. Balances within a
time band are then summed (assets are reported as positive
amounts and liabilities as negative amounts) to produce a net
gap position for each time band. Risk is measured by the size
of the gap (the amount of net imbalance within a time band)
and the length of time the gap is open.
al
Do
cu
11D.571.3
125
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Time bands for distant time periods, say, beyond 10 years, may be
relatively wide five years, for example. These wider time frames
are justified because the change in interest rate sensitivity is small
for maturities beyond 10 years. In other words, a banks use of
wide time bands beyond 10 years will not usually cause it to
misestimate its interest rate risk exposure for items in those time
bands.
Do
cu
126
The first method either recognizes that the cap is in full effect for
the remaining life of the product or ignores it for that same period.
The following example illustrates this all-or-nothing approach to
a cap on a floating rate loan: The bank has a 10-year $100,000
floating rate loan that reprices every six months but is subject to a
12 percent lifetime cap (the rate on the loan cannot exceed 12
percent). The all-or-nothing approach would consider the loan a
six-month floating rate loan when rates are below 12 percent. If
rates equal or exceed 12 percent, the loan becomes a fixed rate loan
with a 10-year repricing maturity. This approach has several
weaknesses. First, the method does not correctly reflect the
exposure of net interest income to future changes in interest rates.
For example, when the loan is slotted as a six-month repricing
asset and funded with a six-month CD, the gap report would not
indicate any interest rate risk. If interest rates were to rise above 12
percent, however, the loan could not reprice further but the funding
costs on the CD could continue to rise, and interest rate margins
would decline. Second, this treatment does not suggest how this
exposure may be hedged. Neither hedging the asset as a six-month
floating rate asset nor hedging it as a 10-year fixed rate asset would
be appropriate.
A better approach would be for the bank to prepare two gap
reports, one for a high-rate scenario and the other for a low-rate
scenario. Under the high-rate scenario, the cap would be binding
and the gap report would show the capped loans as fixed rate
assets. Under the low-rate scenario, the gap report would show
the loan as a floating rate asset.
A bank could use similar approaches to measure prepayment
option risks associated with fixed rate residential mortgage loans.
Under the high-rate scenario, the weighted average lives of the
fixed rate mortgages would be extended in the gap report, reflecting
the effect of slower prepayments.
Under the low-rate scenario, the weighted lives would be
shortened, reflecting faster prepayments. Comparing the gaps
between the two schedules provides an indication of the amount
of option risk the bank faces. Although this second method
provides a way to assess how embedded options may alter a banks
repricing imbalances under alternative interest rate scenarios, it
also has limitations. Like the all-or-nothing approach, this method
suggests that an option has value only when it becomes binding
11D.571.3
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
11D.571.3
rates.
127
Intra-Period Gaps
Although gap reports rely on stratifying balances into broad
time bands, they do not detect imbalances within those bands.
Some bankers have partly overcome this weakness by reporting
the weighted average repricing maturity within each time band.
Another method is to reduce the width of the bands.
New Business
Many gap reports used by banks consider only the banks
current financial positions. These reports are called static
reports because they capture only the risk that arises from the
banks existing balance sheet structure and do not incorporate
any assumptions about new business. Some banks may also
prepare dynamic gap reports. Typically, these reports are
generated from the banks earnings simulation models and
show how the banks gap would appear at some point in the
future, after new business assumptions are incorporated into
the risk measure.
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
11D.571.3
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
A bank might have risk limits that restrict losses in the account at
risk for a defined interest rate scenario over a certain period of
time. For example, the bank in the table above might limit losses
in annual net interest income from a 200 basis point change in
rates to 10 percent of its base net interest income.
Do
cu
Measurement of Risk
The greater the interest rate risk, the greater the change in the
value of a targeted account under different interest rate
scenarios. The target account is usually net interest income or
net income. Many simulation models also are capable of
measuring changes in the market value of equity. Several
business mix and rate scenarios usually are run. Rate scenarios
often include rising, flat, and declining rates, as well as a most
probable scenario.
shapes.
Test for internal consistency among assumptions.
11D.571.3
129
al
income.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Duration
Many economic sensitivity models also compute the duration
of a banks financial instruments. Duration is a measure of the
sensitivity of market values to small changes in interest rates. If
interest rates increase, the market value of a fixed income
instrument will decline. Duration indicates by how much. The
duration of a fixed income instrument that has no option
features is the percentage change in the market value of the
instrument from a change in market rates. For instance, the
market value of a bond with duration of five will decline by
roughly 0.5 percent if interest rates increase by 10 basis points.
Before advances in computing technology made simulations of
net present values under multiple interest rate scenarios feasible,
some bankers used duration as a proxy for estimating the net
economic value of their institution. Duration is still used by
many bank managers as a basis for evaluating the relative risks
of different financial instruments, portfolios, or investment
strategies.
Duration incorporates an instruments remaining time to maturity,
the level of interest rates, and intermediate cash flows. If a fixed
income instrument has only one cash flow, as a zero coupon
bond does, duration will equal the maturity of the instrument: a
zero coupon bond with five years remaining to maturity has a
duration of five years. If coupon payments are received before
maturity, the duration of the bond declines, reflecting the fact that
some cash is received before final maturity. For example, a fiveyear 10 percent coupon bond has duration of 4.2 years in a 10
percent interest rate environment. Duration is calculated by
weighting the present value of an instruments cash flows by the
time to receipt of those cash flows. Table 5 illustrates the calculation
of the Macaulay and modified durations of a $100,000 two-year
note that pays interest semiannually, has a 7.5 percent coupon,
and was purchased at par to yield 7.5 percent. This note has a
modified duration of 1.82. If rates were to increase 100 basis
11D.571.3
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
The higher the duration, the greater the price sensitivity of the
Do
cu
11D.571.3
131
Limitations of Duration
Duration as a measure of the sensitivity of economic value also
has limitations:
Macaulay and modified duration accurately measure changes in
The present value at 8 percent (PV +) is $94. Then the bank estimates
the present value at 6 percent (PV -), taking into account the decrease
in cash flows because the rate of prepayment is higher. The present
value at 6 percent (PV -) is $104. The bonds effective duration
[($104 - $94 / 2 (100) - (.01)] is 5. In other words, the bonds value
will decline by approximately 5 percent for the 100-basis-point
increase in interest rates.
Monte Carlo Simulation
Monte Carlo simulation measures the probable outcomes of
events, such as a movement in interest rates that have a random
or stochastic element. The simulation models discussed
previously measure the value of the bank under a limited
number of interest rate scenarios. Such approaches are
deterministic because the possible interest rate paths are
predetermined and controlled by the model user. Although
deterministic models are valuable, their outcomes depend on
the interest rate scenarios. If actual interest rates differ from
assumptions, the risk to the bank may be substantially different
from the measured risk. The outcome of a Monte Carlo
simulation is less preordained than that of a deterministic
simulation because its statistical modeling technique generates
thousands of randomly determined interest rate paths. These
interest rate paths result in a distribution of possible interest
rate scenarios. The value of the bank or the banks portfolios is
then evaluated for each of the possible interest rate paths,
yielding a range of possible values or outcomes.
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
132
11D.571.3
al
One can estimate how much, in percent, convexity can change the
price of an option-free instrument:
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Notes:
Model Exposure
Regardless of the type of model used, banks should take care
to minimize model exposure. Financial models fall into error
for many reasons. Users may make incorrect assumptions about
deposit behavior or about changes in the spread between
interest rates. They may select a model that is not appropriate
for all parameters. A model that provides reasonable results for
a certain range of inputs may fail to do so for extreme
assumptions. Some model users misuse good models; for
11D.571.3
133
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
LESSON 24:
MEASURING CREDIT RISK
Do
cu
134
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
11D.571.3
135
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Portfolio transactions
Just as CDS can be used to unbundle credit risk, they can also
be combined to create new portfolio instruments with risk and
return characteristics designed to meet the demands of
particular protection buyers and sellers. This use of CDS to
construct portfolio instruments is part of the evolution of the
market in collateralised debt obligations (CDOs). In its simplest
form, a CDO is a debt security issued by a special purpose
vehicle (SPV) and backed by a diversified loan or bond portfolio
(see Diagram 2). The diversification of the portfolio
distinguishes CDO transactions from asset-backed
securitisation (ABS) of homogenous pools of assets such as
mortgages or credit card receivables, a more established
technique. The economics of CDOs is that the aggregate
cashflows on a diversified portfolio have a lower variance than
the cashflows on each individual credit; the lower risk enabling
CDOs to be issued at a lower average yield. Because these are
structured deals, they do not have standardised features in the
potential claims on the SPV that are at least partly backed by the
136
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
The CDOs can be structured so that they are high yielding but
(see below) are said to be important sellers of protection on supersenior tranches, often via back-to-back transactions with another
bank or securities firm in order to obtain a reduced capital
requirement for the bank protection buyer5. Super-senior tranches
are intended to be almost free of credit risk they rank higher
than senior tranches, which are often AAA-rated. Annual premia
are correspondingly low, ranging between 6-12 basis points,
depending on market conditions. But the notional value of the
exposures can be very large. For example, super-senior tranches
on large diversified portfolios of investment grade credits may
cover the last 90% of losses on transactions of US$ billions in
size.
Do
cu
11D.571.3
137
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
138
11D.571.3
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
Non-financial companies
Judging from the Banks regular contacts with UK companies
and market intermediaries, corporate involvement in the credit
derivatives market remains limited to a handful of large
multinationals. Intermediaries do, however, see potential for a
139
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Insurance companies
Insurance companies are net sellers of protection and their
participation in the market seems to be increasing. An insurance
company can sell protection both through investment in
securities such as CDOs or credit-linked notes on the asset side
of its balance sheet and, on the liabilities side of its balance
sheet, by entering into single-name or portfolio default swaps,
writing credit insurance or providing guarantees.
Do
cu
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Intermediaries
Most of the large global investment banks and securities
houses have developed the capacity to buy and sell protection in
the single name CDS market in order to provide liquidity to
customers and trade for their own account. Many are bringing
together their CDS and corporate bond trading desks with a
view to encouraging traders to identify arbitrage opportunities
between the two markets. This parallels moves to integrate, to a
greater or lesser degree, government bond, swap and repo desks
during the 1990s. Intermediaries also use CDSs to manage
credit risk in their other activities. In particular, they buy
protection against counterparty risk arising in other OTC
derivative transactions, such as interest rate swaps (line
buying). In this context, CDSs are now established as an
alternative to collateralisation. For example, an intermediary may
prefer to buy protection from a third party than request
collateral from a counterparty if it is a valuable corporate
customer. The first collateralised debt obligation with credit
events linked to payments by counterparties on a portfolio of
OTC transactions was issued at the end of 2000. One role of
the intermediaries is to bridge the different needs of protection
sellers and buyers. An example is the legal or regulatory
restriction in a number of countries against insurance
companies using derivatives (except to hedge insurance
business), so that these insurers cannot sell protection directly
using ISDA documentation. They can, however, sell insurance
to other insurance companies against their credit exposures on
nearly identical terms. Some intermediaries have therefore
established captive insurance companies (known as
transformers) in financial centres such as Bermuda that do
allow insurers to enter into derivatives. The transformers
typically sell protection to banks using
al
Do
cu
11D.571.3
141
equity tranches of the CBOs that they manage. The track record of
the collateral manager is said to be a key consideration in
funds) may not always have ready access to financing and prefer
to take credit risk though an unfunded CDS than by purchasing
a bond. Financing a bond position exposes the investor to
some liquidity risk if its source of funding becomes more
expensive or dries up. Demand to sell protection by such
investors may reduce CDS premia relative to credit spreads on
bonds.
CDS may expose protection sellers to a little more risk than
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
for corporate bonds can mean CDS premia move higher relative
to credit spreads on bonds if demand to buy protection
increases. This reflects the cost of taking a short position in
bonds in order to arbitrage the two markets. Box 3 shows that
this seemed to happen in the telecom sector in the second half
of 2000.
al
Although the CDS market remains smaller than the bond and
The CDS market may also have greater liquidity for those looking
to take a short position in a particular credit. In the bond market
this means selling the bond short and borrowing it through reverse
repo or stock borrowing. Especially in Europe, liquidity in the
term stock borrowing (or repo) market for corporate bonds can
be unpredictable, partly because not all holders are willing or able
to lend securities. Taking a short position by buying protection
using CDS can be more straightforward. Market participants say
that the CDS market has had greater two-way liquidity than the
bond market in some recent cases when a companys
creditworthiness deteriorated sharply, such as Xerox and Pacific
Gas and Electric. Certainly market participants have been sufficiently
confident in market liquidity that they have used CDS to take
views on changes in creditworthiness, expecting to be able to close
out the position and realise any mark-to-market profit by entering
into an opposite trade in the future. A typical trade might be to
take a view on the shape of the term structure of credit spreads.
For example, a speculator may believe that the forward credit
spreads implied by current premia on term CDS are too high or
low. Such trading increases market liquidity for those buying
protection to hedge credit exposures or selling protection as part
of an investment portfolio.
Notes:
Do
cu
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
Note that, for simplicity, a Vasicek process has been used. The
model can be specified with any model for the short rate. This
allows the use a model that can be calibrated to the term structure
of interest rates such as a Hull-White model. Also note that the
stochastic short rate r is found in the drift for the value process.
This allows for the interplay of market and credit risk beyond
correlation effects.
143
LESSON 25:
INTEGRATING MARKET AND CREDIT RISK MANAGEMENT
al
where d(t) is the recovery rate, C is the coupon and B(r,t;c) is the
value of an comparable Treasury. Additional boundary conditions
will be applied based on the contingent claim being valued. This
emphasizes another advantage of this framework which is the
ability to incorporate a wide range of products by simply changing
the boundary conditions.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
With these processes, the following PDE can be written for the
valuation of a contingent claim, P.
Do
cu
Figure 2 - Bond Value Sensitivity to Firm Value Volatility and Net Cash
Payout Rate
As second level of analysis, the bond valuation was performed
for a range of firm net cash outflow payments and firm value
volatilities. Noting that V* changes as beta, the payout rate,
changes, we find the expected results. Joon (1993) note that firms
with higher net cash flows are more likely to meet their coupon
obligations and should, therefore, have lower credit yield spreads,
as is highlighted in this valuation analysis. As shown in figure 2,
when the payout rate is low, the bond valuation is highly sensitive
144
11D.571.3
Do
cu
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Notes:
11D.571.3
145
Do
cu
Risk factors, like a people risk event raised by a rogue trader cannot
be traced to a directly quantifiable exposure and / or probability
of loss.
4. Quantification: There are two schools of thought concerning
quantification of Operational Risk. One believes that
Operational Risk is an inherently subjective domain and
quantification would rely too much on human judgement to
be reliable. The other school believes that quantitative
measurement presents our only hope of managing operational
risk and hence advocates building up of loss event databases
to provide the requisite statistical foundations for quantification.
However the fact that Operational Risk events do not represent
any notional amounts, contract values or payoffs, makes
quantification of exposure particularly tricky.
al
Introduction
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
LESSON 26:
OPERATIONAL RISK
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Under the AMA, the regulatory capital requirement will equal the
risk measure generated by the banks internal operational risk
measurement system using the quantitative and qualitative criteria
specified in the Capital Accord. Use of AMA is subject to
supervisory approval by the central bank. Banks adopting the
AMA will be required to calculate their capital requirement using
the Advanced measurement Approach as well as as per the existing
Accord for a year prior to implementation of the New Accord at
year-end 2006.
Do
cu
11D.571.3
147
Risk Audit
Employing the services of external (or internal) auditors to
review the business processes of a business unit is another
approach. This process not only helps identify risks but also
helps put in place the oversight organisation for Operational
Risk.
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Qualitative Approaches
Qualitative approaches involve audits, self-assesments and
expert / collective judgement.
Do
cu
Quantitative Approaches
11D.571.3
management.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
Modelling losses
In the sections that follow, we show how to combine
distributional assumptions for event frequency and severity to
derive loss estimates, using the familiar example of transaction
processing errors.
Although it would be possible to model total transaction
handling losses as a single distribution, it is preferable to
combine separate distributions for the mishandling event
process and the severity. This has a number of advantages.
These include better drill-down into the causes and effects of
losses, and the improved ability to set trigger thresholds for
implementing dynamic control processes as part of the
workflow and to see the effects of those controls.
11D.571.3
149
E(S(t))=tm1
and
V(s(t))=tm2
where m1 is the mean of the Weibull distribution and m2 its
variance.
Thus, we can calculate the mean and variances of the total losses
due to mishandling.
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
The final step is to take the overall loss distribution and use it to
attribute risk capital to the overall transaction workflow. The
distribution must be scaled from the daily total severity distribution
to the appropriate horizon and confidence interval dictated by the
firms capital allocation policy. Simulation can be used to aggregate
loss distributions across multiple operational risk categories.
A similar methodology can be applied to address the important
area of model risk. Model risk is a function of input data quality
and inherent model applicability and accuracy. The valuation model
risk resulting from data quality problems should be considered.
The lifecycle of a transaction may be characterised in terms of
canonical events and the types of data quality problems typically
associated with those events, including errors in market data; failure
to capture initially all relevant trade attributes; and failure to capture
lifecycle events such as resets, changes in collateral values,
dividends or corporate actions correctly.
Do
cu
150
11D.571.3
Do
cu
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Notes:
11D.571.3
151
Do
cu
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
LESSON 27:
CASE STUDY ON CREDIT RISK
Some of the big macro hedge funds had encountered this problem
and reduced their size by giving money back to their investors.
When, in the last quarter of 1997 LTCM returned $2.7 billion to
investors, it was assumed to be for the same reason: a prudent
reduction in its positions relative to the market.
But it seems the positions werent reduced relative to the capital
reduction, so the leverage increased. Moreover, other risks had
been added to the equation. LTCM played the credit spread between
mortgage-backed securities (including Danish mortgages) or
double-A corporate bonds and the government bond markets.
Then it ventured into equity trades. It sold equity index options,
taking big premium in 1997. It took speculative positions in
takeover stocks, according to press reports. One such was Tellabs
whose share price fell over 40% when it failed to take over Ciena,
says one account. A filing with the SEC for June 30 1998 showed
that LTCM had equity stakes in 77 companies, worth $541 million.
It also got into emerging markets, including Russia. One report
said Russia was 8% of its book which would come to $10
billion!
Some of LTCMs biggest competitors, the investment banks,
had been clamouring to buy into the fund. Meriwether applied a
formula which brought in new investment, as well as providing
him and his partners with a virtual put option on the performance
of the fund. During 1997, under this formula [see separate section
below, titled UBS Fiasco], UBS put in $800 million in the form of
a loan and $266 million in straight equity. Credit Suisse Financial
Products put in a $100 million loan and $33 million in equity.
Other loans may have been secured in this way, but they havent
been made public. Investors in LTCM were pledged to keep in
their money for at least two years.
LTCM entered 1998 with its capital reduced to $4.8 billion.
A New York Sunday Times article says the big trouble for LTCM
started on July 17 when Salomon Smith Barney announced it was
liquidating its dollar interest arbitrage positions: For the rest of
the that month, the fund dropped about 10% because Salomon
Brothers was selling all the things that Long-Term owned. [The
article was written by Michael Lewis, former Salomon bond trader
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Perhaps they were right. But several factors were against LTCM.
Do
cu
11D.571.3
The off balance sheet contracts were mostly nettable under bilateral
Isda (International Swaps & Derivatives Association) master
agreements. Most of them were also collateralized. Unfortunately
the value of the collateral had taken a dive since August 17.
Surely LTCM, with two of the original masters of derivatives and
option valuation among its partners, would have put its portfolio
through stress tests to match recent market turmoil. But, like
many other value-at-risk (Var) modellers on the street, their worstcase scenarios had been outplayed by the horribly correlated
behaviour of the market since August 17. Such a flight to quality
hadnt been predicted, probably because it was so clearly irrational.
According to LTCM managers their stress tests had involved
looking at the 12 biggest deals with each of their top 20
counterparties. That produced a worst-case loss of around $3
billion. But on that Sunday evening it seemed the mark-to-market
loss, just on those 240-or-so deals, might reach $5 billion. And
that was ignoring all the other trades, some of them in highly
speculative and illiquid instruments.
The next day, Monday September 21, 1998, bankers from Merrill,
Goldman and JP Morgan continued to review the problem. It
was still hoped that a single buyer for the portfolio could be
found - the cleanest solution.
According to Lewiss article LTCMs portfolio had its second
biggest loss that day, of $500 million. Half of that, says Lewis,
was lost on a short position in five-year equity options. Lewis
records brokers opinion that AIG had intervened in thin markets
to drive up the option price to profit from LTCMs weakness. At
that time, as was learned later, AIG was part of a consortium
negotiating to buy LTCMs portfolio. By this time LTCMs capital
base had dwindled to a mere $600 million. That evening, UBS,
with its particular exposure on a $800 million credit, with $266
153
But any action had to be taken swiftly. The danger was a single
default by LTCM would trigger cross-default clauses in its Isda
master agreements precipitating a mass close-out in the over-thecounter derivatives markets. Banks terminating their positions
with LTCM would have to rebalance any hedge they might have
on the other side. The market would quickly get wind of their
need to rebalance and move against them. Mark-to-market values
would descend in a vicious spiral. In the case of the French equity
index, the CAC 40, LTCM had apparently sold short up to 30%
of the volatility of the entire underlying market. The Banque de
France was worried that a rapid close-out would severely hit French
equities. There was a wider concern that an unknown number of
market players had convergence positions similar or identical to
those of LTCM. In such a one-way market there could be a panic
rush for the door.
al
The Feds Peter Fischer invited those three banks and UBS to
breakfast at the Fed headquarters in Liberty Street the following
day. The bankers decided to form working groups to study possible
market solutions to the problem, given the absence of a single
buyer. Proposals included buying LTCMs fixed income positions,
and lifting the equity positions (which were a mixture of index
spread trades and total return swaps, and the takeover bets). During
the day a third option emerged as the most promising: seeking
recapitalization of the portfolio by a consortium of creditors.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
The four core banks met at 7pm and reviewed a term sheet which
had been drafted by Merrill Lynch. Then at 8.30 bankers from nine
more institutions showed. They represented: Bankers Trust,
Barclays, Bear Stearns, Chase, Credit Suisse First Boston, Deutsche
Bank, Lehman Brothers, Morgan Stanley, Credit Agricole, Banque
Paribas, Salomon Smith Barney, Societe Generale. David Pflug,
head of global credit risk at Chase warned that nothing would be
gained a) by raking over the mistakes that had got them in this
room, and b) by arguing about who had the biggest exposure:
they were all in this equally and together.
The delicate question was how to preserve value in the LTCM
portfolio, given that banks around the room would be equity
investors, and yet, at the same time, they would be seeking to
liquidate their own positions with LTCM to maximum advantage.
It was clear that John Meriwether and his partners would have to
be involved in keeping such a complex portfolio a going concern.
154
11D.571.3
They believed that the first-class collateral they held was sufficient
to mitigate their loss if LTCM disappeared. It may have been over
time, but their margin calls to top up deteriorating positions
simply pushed LTCM further towards the brink.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Post mortem
The working group also noted that banks covenants with LTCM
did not require the posting of, or increase in, initial margin as the
risk profile of the counterparty changed, for instance as leverage
increased.(Forfullreports,seSound
e
Practices for Banks
Interactions with Highly Leveraged Institutions, and Banks
Interactions with Highly Leveraged Institutions.) Another report
in June, 1999 by the Counterparty Risk Management Policy Group,
a group of 12 leading investment banks, suggested many ways in
which information-sharing and transparency could be improved.
It noted the importance of measuring liquidity risk, and
improving market conventions and market practices, such as
charging initial margin.
3. Supervision
Do
cu
11D.571.3
155
By June 30, 1999 the fund was up 14.1%, net of fees, from last
September. Meriwethers plan approved by the consortium, was
apparently to redeem the fund, now valued at around $4.7 billion,
and to start another fund concentrating on buyouts and mortgages.
On July 6, 1999, LTCM repaid $300 million to its original investors
who had a residual stake in the fund of around 9%. It also paid
out $1 billion to the 14 consortium members. It seemed
Meriwether was bouncing back.
al
The true test of moral hazard is whether the Fed would be expected
to intervene in the same way next time. Greenspan pointed to a
unique set of circumstances which made an LTCM solution
particularly pressing. It seems questionable whether the Fed would
act as broker for another fund bailout unless there were also such
wide systemic uncertainties.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
The sad truth revealed by this testimony is that the SEC and the
NYSE were concerned only with the risk ratios of their registered
firms and were ignorant and unconcerned, as were the firms
themselves, about the markets aggregate exposure to LTCM.
Bank of England experts note the absence of any covenant
between LTCM and its counterparties that would have obliged
LTCM to disclose its overall gearing. UK banks have long been in
the habit of demanding covenants from non-bank counterparties
concerning their overall gearing, the Bank of England says.
Do
cu
156
It seems that in the developed world, since the early 1990s, financial
firms have built up enough capital to meet most disasters the
world can throw at them. Their mistakes in emerging markets
were costly both for them and for the countries concerned, but
they havent threatened the life of the world financial system. It
seems the mechanisms for restructuring and acquisition are so
swift that the demise of a financial firm simply means it will be
stripped of the trash and carved up. In a down-cycle, however, the
outcome could be very different. Moreover, the social costs of this
financial overreach, followed by cannibalism, could be considerable.
Systemic, no; ripe for concerted private and public intervention,
yes.
On September 29, 1999, six days after the LTCM bailout, US
Federal Reserve chairman Alan Greenspan cut Fed fund rates by
25 basis points to 5.25%. On October 15, 1999 he cut them by
another quarter. His critics associate these cuts directly with the
bail-out of LTCM: it was an extra dose of medicine to make sure
the recovery worked. Some sources attribute the cut to rumours
that another hedge fund was in trouble.
The more generous view is that, if the financial markets were in
disarray, we aint seen nothing yet. Bruce Jacobs, who has followed
the systemic implications of the 1929, 1987 and subsequent minicrashes, fearful of the dangers of globally traded derivatives, writes
in a new book: Had LTC not been bailed out, the immediate
liquidation of its highly leveraged bond, equity, and derivatives
positions may have had effects, particularly on the bond market,
rivaling the effects on the equity market of the forced liquidations
of insured stocks in 1987 and margined stocks in 1929. Given the
links between LTC and investment and commercial banks, and
between its positions in different asset markets and different
countries markets, the systemic risk much talked about in
connection with the growth of derivatives markets may have
become a reality. [Capital ideas and market realities, Blackwell,
1999, page 293]
11D.571.3
al
The deal was calculated so that the $300 million premium was
equivalent to a coupon of Libor plus 50 basis points over the
seven years.
Assuming that LTCM performed well the deal provided UBS
with steady, tax-efficient, return plus a share in the upside, through
its $266 million stake.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
The losers
Among the investors who lost their capital in LTCM (according
to press reports) were:
But it is clear now that UBS risk managers never faced the possibility
of a collapse of LTCM which would have left them with $766
million exposure ($800 million hedge, $266 million investment,
less $300 million option premium). That is, they didnt wake up
to it, apparently, until around April 1998, in a post-merger review,
when it was too late to do much about it. Credit Suisse Financial
Products, which did a similar deal for $100 million, set that as the
maximum it was prepared to lose.
An interesting aspect of the UBS deal is to consider it from LTCMs
point of view. LTCM secured $800 million new investment capital
at Libor plus 50 basis points. It had a call on all returns above that
level. UBSs obligation, to convert any shares it wanted to sell into
a loan, provided LTCM with a synthetic seven-year put on its own
performance. Was this an added incentive to roll the dice? It was a
cheap gambling stake
Notes:
Do
cu
UBS fiasco
The biggest single loser in the LTCM debacle was UBS, which
was forced to write off Sfr950 million ($682 million) of its
exposure. The UBS involvement with LTCM pre-dated the
merger of Union Bank of Switzerland and Swiss Bank
Corporation in December 1998. Various heads rolled, including
that of chairman Mathis Cabiallavetta (formerly chief executive
of Union Bank of Switzerland), Werner Bonadurer, chief
operating officer, Felix Fischer, chief risk officer, and Andy
Siciliano, head of fixed income (who had been with SBC).
UBSs deal with LTCM was a variation on other attempts to
turn hedge funds into a securitized asset class with a protected
downside. However in this case UBS was protecting the
downside and LTCM was taking a good deal of the upside. The
11D.571.3
157
Corrective response
The Basle Committee on Banking Supervisions report on
highly leveraged institutions (HLIs) in January 1999 suggests
that supervisors demand higher capital charges for exposure to
highly leveraged institutions where there is no limit to overall
leverage: Possibly all exposures to all counterparties not
covered by covenants on leverage should carry a higher weight.
It further considers the possibility of extending a credit register
for bank loans in the context of HLIs. The register would
entail collecting, in a centralized place, information on the
exposures of international financial intermediaries to single
counterparties that have the potential to create systemic risk (ie
major HLIs). Exposures would cover both on and off-balancesheet positions. Counterparties, supervisors and central banks
could then obtain information about the overall indebtedness
of the single counterparty.
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
LESSON 28:
CASE STUDY
The build-up of the Nikkei positions took off after the Kobe
earthquake of January 17. This is reflected in Figure 10.1 - the
chart shows that Lessons positions went in the opposite direction
to the Nikkei - as the Japanese stock market fell, Leesons position
increased. Before the Kobe earthquake, with the Nikkei trading in
a range of 19,000 to 19,500, Leeson had long futures positions of
approximately 3,000 contracts on the Osaka Stock Exchange. (The
equivalent number of contracts on the Singapore International
Monetary Exchange is 6000 because SIMEX contracts are half the
size of the OSE.) A few days after the earthquake Leeson started
158
11D.571.3
Actual4
Actual
position in
terms of
open
interest of
relevant
contract 2
Futures
JGB
15940
$8980
million
601
Euroyen $26.5
million
85% of
March
1995
short 28034
contract
$19650
and 88%
million
of June
1995
contract.
short 6845
$350
million
Nikkei
225
Nil
5% of
June 1995
contract,
1% of
September
1995
contract
and 1% of
December
1995
contract.
Figure 10.2
37925 calls
$3580
million
32967 puts
$3100
million
Do
cu
Options
long 61039
$7000
million
al
30112
$2809
million
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Nikkei
225
49% of
March
1995
contract
and 24%
of June
1995
contract.
11D.571.3
159
Table
10.1
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Price per
SIMEX
Average
Value per
Price per
SIMEX
CONTACT JPY
millions
Value per
Profit/(Loss)
CONTACT to '92000'
JPY millions JPY millions
20
6984
January
18950
19019
66173
66413
240
23
3000
January
17810
18815
26715
28223
1508
Table
10.3
No. of
contracts in
account
'88888' 2
Buy
23
January
Period
Reported (milion)
Actual (milion)
Cumulativeactual1 (milion)
1Jan1993to31Dec1993
+GBP8.83
-GBP21
-GBP23
1Jan1994to31Dec1994
+GBP28.529
-GBP185
-GBP208
1Jan1995to31Dec1995
+GBP18.567
-GBP619
-GBP827
Do
cu
Table10.2
The cross-trade
How was Leeson able to deceive everyone around him? How
was he able to post profits on his switching activity when he
was actually losing? How was he able to show a flat book when
he was taking huge long positions on the Nikkei and short
positions on Japanese interest rates? The Board of Banking
Supervision (BoBS) of the Bank of England which conducted
an investigation into the collapse of Barings believes that the
vehicle used to effect this deception was the cross trade.1 A
cross trade is a transaction executed on the floor of an Exchange
by just one Member who is both buyer and seller. If a Member
has matching buy and sell orders from two different customer
accounts for the same contract and at the same price, he is
allowed to cross the transaction (execute the deal) by matching
160
Sell
8082 17810
18147
(71970)
(73332)
(1362)
25
10047
January
18220
18318
91528
92020
492
26
16276
January
18210
18378
148193
149560
1367
2245
11D.571.3
Do
cu
Figure 10.3
Graph to show the Nikkei Position of Account 92000.
Reproduced by permission from the Report of the Board of
Banking Supervision Inquiry into the Circumstances of the
Collapse of Barings.
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
The bottom line of all these cross-trades was that Barings was
counterparty to many of its own trades. Leeson bought from one
hand and sold to the other, and in so doing did not lay off any of
the firms market risk. Barings was thus not arbitraging between
SIMEX and the Japanese exchanges but taking open (and very
substantial) positions, which were buried in account 88888. It
was the profit and loss statement of this account which correctly
represented the revenue earned (or not earned) by Leeson. Details
of this account were never transmitted to the treasury or risk
control offices in London, an omission which ultimately had
catastrophic consequences for Barings shareholders and
bondholders.
e. Lack of supervision
161
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Do
cu
162
c. Adequate capital
There are two aspects to this issue - an institution must have
sufficient capital to withstand the impact of adverse market
moves on its outstanding positions as well as enough money
to keep these positions going. Barings management thought
that Leesons positions were market neutral and were thus quite
happy to fund margin requirements till the contracts expired. In
the end, these collateral calls from SIMEX and OSE proved too
much to bear (as was pointed out earlier, they were larger than
Barings capital base) and the 200-year old institution was forced
to call in the receivers. It was funding risk that seriously
wounded Barings but the terminal shot came from the
discovery that the enormous positions were unhedged.
Funding risk also nearly sank Metallgesellschaft, a German
industrial company, in 1993. In that year alone,
Metallgesellschafts US subsidiary paid out $900 million in
margins for its crude oil hedges on NYMEX. When the
American subsidiary asked for a cash infusion to meet further
11D.571.3
Do
cu
i. Funding
Barings control procedures were sloppy. No where is this point
better illustrated than in the way it funded BFS (or more
accurately Leesons unauthorised positions). Barings did not
require Leeson, to distinguish between variation margin needed
to cover proprietary and customer trades; neither did it have a
system to reconcile the funds Leeson requested to his reported
positions and/or that of its client positions. (The London
office for example could have used the Standard Portfolio
Analysis of Risk (SPAN) margining programme to calculate
margins and would then have realised that the amount of
money Leeson was requesting was significantly more than that
called for under Simexs margining rules.) London simply,
automatically, remitted to Leeson the sum of money he asked
for, despite misgivings felt by many senior operational staff
about the accuracy of his data. The fact that no one even asked
Leeson to justify his requests is all the more astounding given
the size of his demands. At the end of Dec 1994, the
cumulative funding of BFS by Barings London and Tokyo
stood at US$354 million. In the first two months of 1995, this
figure increased by US$835 million to US$1.2 billion. The BoBS
11D.571.3
al
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
163
al
e. Lack of supervision
Theoretically Leeson had lots of supervisors; in reality none
exercised any real control over him. Barings operated a matrix
management system, where managers who are based overseas
report to local administrators and to a product head (usually
based at head office or the regional headquarters).Leesons
Singapore supervisors were James Bax, regional manager South
Asia and a director of BFS, and Simon Jones, regional
operations manager South Asia, also a director of BFS and
chief operating officer of Barings Securities Singapore. Jones
and the heads of the support functions in Singapore also had
reporting lines to the Group-wide support functions in
London. Yet both Bax and Jones told the BoBS inquiry that
they did not feel operationally responsible for Leeson. Bax felt
Leeson reported directly to Baker or Walz on trading matters
and to Settlements/Treasury in London for backoffice matters.
Jones felt his role in BFS was limited only to administrative
matters and concentrated on the securities side of Barings
activities in South Asia. Leesons reporting lines for product
profitability are not clear cut since his supervisors have disputed
who was directly responsible for him from January 1, 1994. His
ultimate boss was Ron Baker, head of the financial products
group. But who had day-to-day control over him? Mary Walz,
global head of equity financial products, insists that she
thought Fernando Gueler, head of equity derivatives proprietary
trading in Tokyo was in charge of Leesons intra-day activities
since the latters switching activities were booked in Tokyo.
However, Gueler insists that in October 1994, Baker told him
that Leeson would report to London and not Tokyo. He thus
assumed that Walz would be in charge of Leeson. Walz herself
still disputes this claim. Tapes of telephone conversations show
that Leeson spoke frequently to both Gueler and Walz. (The
bottom line however is that Gueler reported to Walz.) Two
important incidents vividly illustrate the cavalier attitude Barings
had towards supervising Leeson. The first involves two letters
to BFS from SIMEX. In a letter dated 11 January, 1995; SIMEX
senior vice-president for audit and compliance Yu Chuan Soo,
complained about a margin shortfall of about US$116 million
in account 88888 and that Barings had appeared to break
SIMEX rule 822 by previously financing the margin
requirements of this account, (which appeared in SIMEXs
system as a customer account.) SIMEX also noted that the
initial margin requirement of this account was in excess of
US$342 million. BFS was asked to provide a written
explanation of the margin difference on account 88888 and of
its inability to account for the problem in the absence of
Leeson. No warning lights went off in Singapore. No one
investigated who this customer really was and why he was
having difficulties in meeting margin payments or why he had
such a huge position; or the credit risk Barings faced if this
customer defaulted on the margins that Barings had paid on
its behalf. A copy of the letter was not sent to operational heads
in London. Simon Jones did not press Leeson for an
explanation; indeed he dealt with the matter by allowing Leeson
to draft Barings response to SIMEX. The second incident did
come to the attention of London but again was dealt with
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
Figure 10.4
Top-up Funding from BSGT to BSL and Margin Balances from
BFS from 1 January 1995. Reproduced be permission from the
Report of the Board of Banking Supervision Inquiry into the
Circumstances of the Collapse of Barings
Do
cu
out far too much money just to cover the margin calls of clients.
iv. No limits
Barings did not impose any gross position limits on Leesons
proprietary trading activities because it felt that there was little
market risk attached to arbitrage trades since at the close of
business, the position must be flat. But the Barings collapse has
shown that placing gross position limits on each side of an
arbitrage book is perhaps not such a bad idea after all. While it is
true that an arbitrage book has little price (directional ) risk, it has
basis and settlement risk. The former arises because prices in two
markets do not always move in tandem and the latter because
164
11D.571.3
ww Co
w.p m
dfw P
iza D
rd. F
com Tr
i
al
Do
cu
Conclusion
The Nikkei 225 and JGB futures contracts traded by Leeson
were the simplest of derivative instruments. They were also the
most transparent - since they were listed contracts, Leeson was
required to pay (or receive) daily margins and so needed funds
from London. In January and February 1995 alone, he asked for
US$835 million. His could not hide his build-up of positions
on the OSE because the exchange publishes weekly numbers.
All his rivals could see his enormous positions, and many
assumed that the positions were hedged because such naked
positions were out of all proportion to the firms capital base or
even those of other players. His senior managers also assumed
Leesons were hedged. But unlike outsiders who had to assume
that these positions were hedged, Barings management did
not. They could have done something about it - they could
have probed Leeson, they could have tried to obtain more
information from their internal information systems, and most
of all they could have heeded the warning signals available in
late 1994 and throughout January and February of 1995. But
although Barings fate was only sealed in the final weeks of
February, the seeds of its destruction were sown when senior
management entered new businesses without ensuring
adequate support and control systems. The collapse of Britains
oldest merchant bank was an extreme example of operations
risk, i.e. the risk that deficiencies in information systems or
internal controls result in unexpected loss. Will it happen again?
Certainly, if senior managers of firms continue to disregard
rules and recommendations which have been drawn up to
ensure prudent risk-taking.
Notes:
11D.571.3
165