Risk Management IFT Notes

Risk Management
1. Introduction .............................................................................................................................................. 3
2. Risk Management as a Process ................................................................................................................. 3
3. Risk Governance........................................................................................................................................ 4
4. Identifying Risks ........................................................................................................................................ 5
5. Measuring Risk .......................................................................................................................................... 6
5.1. Measuring Market Risk ...................................................................................................................... 6
5.2. Value at Risk ....................................................................................................................................... 7
5.3. The Advantages and Limitations of VaR .......................................................................................... 11
5.4. Extensions and Supplements to VaR................................................................................................ 11
5.5. Stress Testing ................................................................................................................................... 11
5.6. Measuring Credit Risk ...................................................................................................................... 12
5.7. Liquidity Risk .................................................................................................................................... 14
5.8. Measuring Nonfinancial Risks .......................................................................................................... 14
6. Managing Risk ......................................................................................................................................... 14
6.1. Managing Market Risk ..................................................................................................................... 14
6.2. Managing Credit Risk ....................................................................................................................... 15
6.3. Performance Evaluation................................................................................................................... 15
6.4. Capital Allocation ............................................................................................................................. 16
6.5. Psychological and Behavioral Considerations .................................................................................. 17
Summary ..................................................................................................................................................... 17
Examples from the Curriculum ................................................................................................................... 23
Example 1. Some Risk Governance Concerns of Investment Firms ........................................................ 23
Example 2. An Analysis of Risk Exposures............................................................................................... 23
Example 3. An Operational Risk for Financial Services Companies: The Rogue Trader ......................... 24
Example 4. Accounting Risk: The Case of Derivative Contracts .............................................................. 24
Example 5. VaR with Different Probability Levels and Time Horizons .................................................... 25
Example 6. Calculating VaR Using the Historical Method....................................................................... 26
Example 7. Value at Risk and the Management of Market Risk at Goldman Sachs ............................... 27
Example 8. Calculating Credit Risk Exposures......................................................................................... 31
Example 9. Basel II—A Brief Overview .................................................................................................... 32
Example 10. A Fund Management Company and Risk Budgeting .......................................................... 33

IFT Notes for the Level III Exam [Link] Page 1

 Risk Management IFT Notes

Example 11. Repricing a Forward Contract............................................................................................. 34

This document should be read in conjunction with the corresponding reading in the 2019 Level III CFA®
Program curriculum. Some of the graphs, charts, tables, examples, and figures are copyright
2018, CFA Institute. Reproduced and republished with permission from CFA Institute. All rights reserved.

Required disclaimer: CFA Institute does not endorse, promote, or warrant the accuracy or quality of the
products or services offered by IFT. CFA Institute, CFA®, and Chartered Financial Analyst® are
trademarks owned by CFA Institute.

IFT Notes for the Level III Exam [Link] Page 2

 Risk Management IFT Notes

1. Introduction
A portfolio manager needs to understand risk management as it relates to his firm. He should also
understand the risk management process of firms where he invests.

LO.a: Discuss features of the risk management process, risk governance, risk reduction, and an
enterprise risk management system.
This LO is covered in sections 2 and 3.

2. Risk Management as a Process

Risk management is a process that involves:
• Recognising the exposures to risk.
• Creating appropriate ranges for exposures.
• Constantly measuring these exposures.
• Performing appropriate adjustments whenever exposure levels fall outside of target ranges.

This is a constant process and may need adjustments in any of these activities to reflect new policies.

Exhibit 1 shows the practical application of the process of risk management.

IFT Notes for the Level III Exam [Link] Page 3

 Risk Management IFT Notes

While applying the risk management process to portfolio management, managers need to measure and
price the risks of financial transactions or positions. Exhibit 2 demonstrates the process of pricing and
measuring risk.

3. Risk Governance
LO.b: Evaluate strengths and weaknesses of a company’s risk management process.

Risk governance is an element of corporate governance. It is the process of setting overall policies and
standards in risk management.

The two types of risk governance models are:

• Decentralized: Here each unit calculates and reports its risk exposures independently. The main
advantage is that this model allows people who understand risk better to directly manage it.
• Centralized: Here the risk management is moved closer to senior management. The main

IFT Notes for the Level III Exam [Link] Page 4

 Risk Management IFT Notes

advantage is that it allows offsetting of risks across units. For example, if one unit is exporting
goods to Japan and other unit is importing goods from Japan, then the Yen risk exposure can be
offset. Centralized risk management is now called enterprise risk management (ERM).

Refer to Example 1 from the curriculum.

LO.c: Describe steps in an effective enterprise risk management system

An effective ERM system usually includes the following steps:

1. Identify each risk factor to which the company is exposed.
2. Quantify each exposure’s size in money terms.
3. Map these inputs into a risk estimation calculation.
4. Identify overall risk exposures as well as the contribution to overall risk deriving from each risk
factor.
5. Set up a process to report on these risks periodically to senior management, who will set up a
committee of division heads and executives to determine capital allocations, risk limits, and risk
management policies.
6. Monitor compliance with policies and risk limits.

4. Identifying Risks
LO.d: Evaluate a company’s or a portfolio’s exposures to financial and nonfinancial risk factors

Exhibit 3 shows the main sources of risk. Financial risk is the risk derived from events in the external
financial markets. Non-financial risk refers to all other forms of risk.

Refer to Example 2 from the curriculum.

The various types of risks are:

Market Risk: Risk associated with interest rates, exchange rates, stock prices and commodity prices.

Credit Risk: Risk of loss caused by a counterparty or debtor’s failure to make a promised payment.

IFT Notes for the Level III Exam [Link] Page 5

 Risk Management IFT Notes

Liquidity Risk: Risk that a financial instrument cannot be purchased or sold without a significant
concession in price because of the market’s potential inability to efficiently accommodate the desired
trading size (large bid-ask spread).

Operational Risk: Risk of loss from failures in a company’s systems and procedures or from external
events.

Refer to Example 3 from the curriculum.

Model Risk: Risk that a model is incorrect or misapplied; in investments, it often refers to valuation
models.

Settlement (Herstatt) Risk: Risk that one party could be in the process of paying the counterparty while
the counterparty is declaring bankruptcy.

Regulatory Risk: Risk associated with how a transaction will be regulated or with the potential for
regulations to change.

Legal/Contract Risk: The possibility of loss arising from the legal system’s failure to enforce a contract in
which an enterprise has a financial stake.

Tax Risk: Risk associated with uncertainty in tax laws.

Accounting Risk: Risk associated with uncertainty about how a transaction should be recorded and the
potential for accounting rules and regulations to change.

Refer to Example 4 from the curriculum.

Sovereign Risk: A form of credit risk in which the borrower is the government of a sovereign nation.

Political Risk: Risk associated with changes in political environment.

ESG Risk: Risk to a company’s market valuation resulting from environmental, social and governance
factors.

Performance Netting Risk: Potential for loss resulting from the failure of fees based on net performance
to fully cover contractual payout obligations to individual portfolio managers that have positive
performance when other portfolio managers have losses and when there are asymmetric incentive fee
arrangements with the portfolio managers.

Settlement Netting Risk: Risk that a liquidator or a counterparty in default could challenge a netting
arrangement so that profitable transactions are realized for the benefit of creditors.

5. Measuring Risk
5.1. Measuring Market Risk
Several statistical tools are available to measure market risk:
• Volatility of the asset, measured by the standard deviation of asset prices.
• Volatility relative to a benchmark (active risk, tracking risk) measured by deviation of a
portfolio’s returns in excess of a benchmark.

IFT Notes for the Level III Exam [Link] Page 6

 Risk Management IFT Notes

• For stocks, beta measures the sensitivity to market movements.

• For bonds, duration and convexity measure sensitivity of a bond to a small parallel shift in the
yield curve.
• For options:
o delta measures an option’s sensitivity to a small change in the value of the underlying
o gamma measures the delta’s sensitivity to a change in the value of the underlying
o vega measures an option’s sensitivity to a change in the underlying’s volatility
o theta measures an option’s sensitivity to a change in the time to expiration

5.2. Value at Risk

LO.e: Calculate and interpret value at risk (VaR) and explain its role in measuring overall and
individual position market risk

VAR is a probability-based measure of loss potential. More precisely, VAR is an estimate of the loss (in
money terms) that we expect to exceed with a given level of probability over a specified time period.

Consider the following example of VaR for an investment portfolio: The VaR for a portfolio is $1.5 million
for one day with a probability of 0.05. This statement can be interpreted in the following two ways:
• There is a 5 percent chance that the portfolio will lose at least $1.5 million in a single day.
• We can say with 95 percent confidence, that the maximum loss will be $1.5 million for one day.

Elements of measuring VAR

VAR measures requires the user to make the following decisions about the calculation’s structure.

1. Picking a probability level:

• The probability chosen is typically either 0.05 or 0.01.
• Using 0.01 is more conservative and will give a higher VAR. It is recommended to use 0.01 if
returns are non-linear.
2. Selecting the time period over which to measure VAR:
• The choices are: day, week, two-week, month etc.
• If the portfolio has a high turnover it is recommended to use a shorter period.
• Using a longer period results in a higher VAR.
3. Choosing the specific approach to modeling the loss distribution.

Once these key parameters are selected, we can obtain the VAR estimate. Exhibit 4 shows the
probability distribution for the returns on a portfolio over a specified time period.

Return on Portfolio Probability

Less than –40% 0.010
–40% to –30% 0.010
–30% to –20% 0.030
–20% to –10% 0.050
–10% to –5% 0.100

IFT Notes for the Level III Exam [Link] Page 7

 Risk Management IFT Notes

–5% to –2.5% 0.125

–2.5% to 0% 0.175
0% to 2.5% 0.175
2.5% to 5% 0.125
5% to 10% 0.100
10% to 20% 0.050
20% to 30% 0.030
30% to 40% 0.010
Greater than 40% 0.010
1.000

The first row in this table tells us that there is a 0.01 probability of a loss of 40% or worse. The second
row tells us that there is a 0.01 probability of a loss between 30% and 40%. The third row tells us that
there is a 0.03 probability of a loss between 20% and 30%. Adding the three probability numbers we can
say there is 0.05 probability of a loss of 20% or worse.

Assuming a portfolio of $100 million, the VAR with a probability of 0.05 can be calculated as:

20% of $100 million = $20 million.

Methods to estimate VAR

The three standardized methods to estimate VAR are:

1. Analytical variance-covariance method

2. Historical method
3. Monte Carlo simulation method

Analytical variance-covariance method:

Here we assume that the portfolio returns are normally distributed. Exhibit 5 shows the expected annual
returns and standard deviation of a portfolio obtained by combining the S&P 500 and NASDAQ.

S&P 500 NASDAQ Combined Portfolio

Percentage invested (w) 0.75 0.25 1.00
Expected annual return (μ) 0.12 0.18 0.135
Standard deviation (σ) 0.20 0.40 0.244
Correlation (ρ) 0.90

Exhibit 6 demonstrates how we can calculate the annual VAR of this portfolio with a probability of 0.05.
We start by determining the z-value associated with a cumulative probability of 0.05. This z-value is -
1.65. The distribution we are working with has a mean of 0.135 and a standard deviation of 0.244.
Given these numbers it can be shown that 5% of the area under the curve is to the left of -0.268. The

IFT Notes for the Level III Exam [Link] Page 8

 Risk Management IFT Notes

calculations are outlined in the figure.

If the portfolio is worth $50 million, we can express VaR as $50,000,000(0.268) = $13.4 million. In other
words there is a 5% chance that the loss will be $13.4 million or worse in a year.

To calculate the daily VAR, we can adjust the expected return to its daily average of approximately
0.135/250 = 0.00054 and the standard deviation to its daily value of 0.244 sqrt(250) = 0.01543. This is
based on the assumption of 250 trading days in a year and statistical independence between days. With
these numbers the daily VaR is 0.00054 – 1.65(0.01543) = –0.0249. On a dollar basis, the daily VaR is
$50,000,000(0.0249) = $1.245 million.

Refer to Example 5 from the curriculum.

Historical method:

In this method we calculate the returns for a given portfolio using actual prices from a user-specified
period in recent past. The advantage of using this method is that the user does not have to make any
assumptions about the type of probability distributions that generates returns. The disadvantage is that
it relies on data from the past and past conditions might not hold in the future.

Refer to Example 6 from the curriculum.

Refer to Example 7 from the curriculum.

Monte Carlo simulation method:

In this method we produce random outcomes according to an assumed probability distribution and a set
of input parameters.

For example, consider a $50 million portfolio invested 75% in S&P500 and 25% in NASDAQ. The annual
expected return is 13.5% and the standard deviation is 24.4%. We use a random number generator to

IFT Notes for the Level III Exam [Link] Page 9

 Risk Management IFT Notes

produce a series of 300 random values. The results are shown in Exhibit 9.

To obtain the point in the lower tail with a 5 percent significance, we rank order the data and find the
15th-lowest outcome. We use the 15th-lowest outcome because there are 300 iterations and 5 percent of
300 = 15. This corresponds to a portfolio value of $34.25 million – a loss of $15.75 million. Based on this
information we can say: there is a 5% chance that the loss in one year will be $15.75 million or worse.

LO.f: Compare the analytical (variance–covariance), historical, and Monte Carlo methods for
estimating VaR and discuss the advantages and disadvantages of each

The table below compares the three methods

Variance-Covariance Method Historical Method Monte Carlo Simulation

Advantages: Advantages: Advantages:

Simple method. Non-parametric (minimal Widely used.
probability-distribution We can assume an
Disadvantages: assumptions). appropriate distribution of
Relies on simplifying assumptions input data.
such as normality of returns. Disadvantages:
If returns are not normally Relies on past events (which Disadvantages:
distributed then we cannot rely might not be good predictors Output only as good as our
totally on standard deviation as a of the future). input assumption.
measure of risk. Monte Carlo simulation
Consider skewness and kurtosis. software is expensive and
Portfolios containing options don’t requires a lot of computing
have normal distributions. power.

IFT Notes for the Level III Exam [Link] Page 10

 Risk Management IFT Notes

LO.g: Discuss advantages and limitations of VaR and its extensions, including cash flow at risk,
earnings at risk, and tail value at risk
This LO is covered in sections 5.3 and 5.4

5.3. The Advantages and Limitations of VaR

Advantages Limitations

Quantifies potential loss in simple terms. One sided. Entire focus is on the left tail.

Most regulatory authorities accept VAR as an Lulls into false sense of security. Gives the
acceptable risk measure. incorrect impression that risk is properly
understood and quantified.
Some companies use VAR as a measure of
capital at risk across different lines of business. Often underestimates magnitude and frequency
of worst returns.

Extremely difficult to calculate VAR for large

organizations.

Key points to remember are:

• No risk measure can precisely predict future losses; VAR is no exception.
• Back testing allows you to determine accuracy level of VAR estimates.
• Do not use VAR in isolation.
• Ensure that inputs to VAR calculation are as reliable as possible and relevant to the current
investment mix.

5.4. Extensions and Supplements to VaR

Incremental VAR (IVAR) measures the incremental effect of an asset on the VAR of a portfolio by
measuring the difference between the portfolio's VAR while including a specified asset and the
portfolio's VAR with that asset eliminated.

Cash flow at risk (CFAR) is the minimum cash flow loss that we expect to be exceeded with a given
probability over a specified time.

Earnings at risk (EAR) is the minimum earnings loss that we expect to be exceeded with a given
probability over a specified time.

Tail value at risk (TVAR) is VAR plus the expected loss in excess of VAR when such excess loss occurs.

5.5. Stress Testing

LO.h: Compare alternative types of stress testing and discuss advantages and disadvantages of each

Managers use stress testing to supplement VAR as a risk measure. VAR quantifies potential losses under
normal market conditions. Stress testing seeks to identify unusual circumstances that could lead to
losses in excess of those typically expected.

IFT Notes for the Level III Exam [Link] Page 11

 Risk Management IFT Notes

The two broad approaches are: 1) scenario analysis and 2) stress modeling.

Scenario analysis is the process of evaluating a portfolio under different states of the world. Stylized
scenarios (one type of scenario analysis) involves simulating a movement in at least one interest rate,
exchange rate, stock price, or commodity price relevant to the portfolio.

Derivatives Policy Group recommends the following scenarios:

▪ parallel yield curve shifting by ±100 basis points (1 percentage point);
▪ yield curve twisting by ±25 basis points;
▪ each of the four combinations of the above shifts and twists;
▪ implied volatilities changing by ±20 percent from current levels;
▪ equity index levels changing by ±10 percent;
▪ major currencies moving by ±6 percent and other currencies by ±20 percent;
▪ swap spread changing by ±20 basis points.

Another approach to scenario analysis is to use actual extreme events that have occurred in the past.
We can also create scenarios based on hypothetical events – i.e. events that have never happened but
might happen.

Stress modeling: It is difficult to estimate sensitivity of a portfolio to scenarios we design; so another

approach is to use an existing model and apply shocks and perturbations to the model inputs in some
mechanical way.

Stressing models can take several forms:

• Factor push: Here we push the prices and risk factors of an underlying model in the most
disadvantageous way to calculate the combined effect on the portfolio’s value.
• Maximum loss optimization: Here we try to optimize mathematically the risk variables that will
produce the maximum loss.
• Worst case scenario analysis: Here we examine the worst case that we actually expect to occur.

5.6. Measuring Credit Risk

LO.i: Evaluate the credit risk of an investment position, including forward contract, swap, and option
positions

Credit losses have two dimensions:

• Likelihood of loss.
• Associated amount of loss.

In risk management, credit risk exposure should be viewed from two different time perspectives:
• Current credit risk is the risk associated with events in the near future.
• Potential credit risk is the risk associated with events at a later date.

Cross-default provision: A owes B with a due date in 6 months and also owes C now. Say A defaults on
the payment to C. If there is credit-default provision then A has technically also defaulted to B.

Credit VAR measures credit risk. It reflects the minimum loss with a given probability during a period of

IFT Notes for the Level III Exam [Link] Page 12

 Risk Management IFT Notes

time.

Option Pricing Theory and Credit Risk

The stock of a company with leverage can be viewed as a call option on assets. All upside potential
above the value of debt belongs to the shareholders, if the asset value is less than debt value then
shareholders will receive nothing.

Bond with credit risk can be viewed as a default free bond plus an implicit short put option written by
bondholders for the stockholders. In other words, the bondholders have implicitly written the
stockholders a put on the assets. From the stockholders’ perspective, this put is their right to fully
discharge their liability by turning over the assets to the bondholders, even though those assets could be
worth less than the bondholders’ claim.

The Credit Risk of Forward Contracts

This is the risk that the party that owes the larger amount could default.

Credit risk exposure is based on the market value which is calculated as:

PV or amount owed – PV of amount to be paid.

For example: At T = 0, S = 100, I = 5%, → F = 105

Three months later asset price = 102.

We can determine that the long forward contract’s value at that time is $102 – $105/(1.05)0.75 =
$0.7728. This is the value to the long because the contract is a claim on the asset, which is currently
worth $102, and an obligation to pay $105 for it in nine months. To the holder of the long position, this
contract is worth $0.7728, and to the holder of the short position, it is worth –$0.7728.

The long’s claim is positive; the short’s claim is negative. Therefore, the long currently bears the credit
risk.

The Credit Risk of Swaps

Swap’s market value is the PV of amounted owed – PV of amount to be paid.

For interest rate and equity swaps the potential credit risk is the largest during the middle period of the
swap’s life:
1. Risk is low at the start because both counterparties will have performed sufficient current credit
analysis.
2. Risk is low at the end because few payments remain.

Currency swaps have greatest credit risk closer to the end of swap’s life. This is because the notional
principal needs to be swapped at the end of the currency swap.

Credit Risk of Options

Forward contracts and swaps have bilateral default risk. This means that either party could face credit
risk. Options have unilateral credit risk. Only the long party faces credit risk.

IFT Notes for the Level III Exam [Link] Page 13

 Risk Management IFT Notes

Refer to Example 8 from the curriculum.

5.7. Liquidity Risk

VAR assumes liquidity; however, some assets are not very liquid.

Bid-ask spread relative to price is a measure of liquidity. Assets which are traded very infrequently
present a larger problem because they provide a statistical illusion of low volatility.

The selling price of illiquid assets might be much lower than expected. A famous case of
underestimating liquidity risk is the LTCM failure.

5.8. Measuring Nonfinancial Risks

Non-financial risks include the risk of fire, flood, tornadoes, terrorist attacks, and so on. These risk are
very difficult to measure. Most of these risks are more suitable for insurance.

Refer to Example 9 from the curriculum.

6. Managing Risk
6.1. Managing Market Risk
LO.j: Demonstrate the use of risk budgeting, position limits, and other methods for managing
market risk

We need to:
• identify sources of market risk
• define how these risks will be measured
• set appropriate risk tolerance levels
• identify corrective action if actual risk is outside tolerance levels

Risk budgeting focuses on where to take risk and how to efficiently allocate risk. For example, consider a
bank with an FX unit and a fixed income unit. The allocation of capital and risk budget for each unit is
shown below:
• FX desk: allocated capital = 100 mm and permitted daily VAR = 5 mm.
• Fixed income desk: allocated capital = 200 mm and permitted daily VAR = 5 mm.
• Permitted daily VAR for both desks combined = 9 mm.

The FX desk has an allocated capital of 100 mm and a daily VAR limit of 5 mm. The fixed income desk
has twice the allocated capital (200 mm) but the same daily VAR limit of 5 mm. Clearly in percentage
terms the FX desk is allowed to take more risk. This simple example also illustrates how the risk is
budgeted across the two desks.

Note that the sum of risk budgets for individual units (5 mm + 5 mm = 10 mm) exceeds the risk budget
for the organization (9 mm) because of the diversification effect.

Refer to Example 10 from the curriculum.

IFT Notes for the Level III Exam [Link] Page 14

 Risk Management IFT Notes

6.2. Managing Credit Risk

LO.k: Demonstrate the use of exposure limits, marking to market, collateral, netting arrangements,
credit standards, and credit derivatives to manage credit risk

Credit risk is one sided and returns are not symmetric. Hence this risk is not easily measured using
standard deviation and VAR.

The various methods to reduce credit risk are:

Reducing credit risk by limiting exposure:

• Limiting exposure to a given party is the primary means of managing credit risk.
• Banks have regulatory constraints on the amount of credit risk they can assume.

Reducing credit risk by marking to market:

• Futures contracts typically require daily mark to market.
• The same concept can also be applied in the OTC market.
• Refer to Example 11 from the curriculum.

Reducing credit risk with collateral:

• Futures markets require that all participants post collateral.
• The same concept can be also be applied in OTC market.

Reducing credit risk with netting:

• Payment netting reduces amount of money that must be paid.
• Hence it reduces credit risk.

Reducing credit risk with minimum credit standards and enhanced derivative product companies:
• Enhanced derivatives products companies (EDPCs) are separate from the parent organization
and are not liable for the parent’s debts.
• They are also heavily capitalized.
• The objective is to obtain good credit quality rating from the rating agencies.
• EDPCs often have higher credit ratings than the parent company.

Transferring credit risk with credit derivatives:

• Credit derivatives can be used to transfer risk to another party.
• Examples include: credit default swap, total return swap, credit spread option, credit spread
forwards.

6.3. Performance Evaluation

LO.l: Discuss the Sharpe ratio, risk-adjusted return on capital, return over maximum drawdown, and
the Sortino ratio as measures of risk-adjusted performance

In order to maximize the risk-adjusted return, we must measure performance against risks taken.

The following measures can be used:

IFT Notes for the Level III Exam [Link] Page 15

 Risk Management IFT Notes

Sharpe ratio:
𝑀𝑒𝑎𝑛 𝑝𝑜𝑟𝑡𝑓𝑜𝑙𝑖𝑜 𝑟𝑒𝑡𝑢𝑟𝑛 − 𝑅𝑖𝑠𝑘 𝑓𝑟𝑒𝑒 𝑟𝑎𝑡𝑒
𝑆ℎ𝑎𝑟𝑝𝑒 𝑟𝑎𝑡𝑖𝑜 =
𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑝𝑜𝑟𝑡𝑓𝑜𝑙𝑖𝑜 𝑟𝑒𝑡𝑢𝑟𝑛
Sharpe ratio is inaccurate when applied to portfolios with significant nonlinear risks such as option
positions.

Risk-Adjusted Return on Capital (RAROC):

Here we divide the expected return on an investment by a measure of capital at risk. Risk measurement
can be calculated in various ways. A company may require that an investment’s expected RAROC exceed
a RAROC benchmark level for capital to be allocated to it.

Return over Maximum Drawdown (RoMAD):

Drawdown is the difference between a portfolio’s maximum point of return (known as high water mark)
and any subsequent low point of performance. Return over maximum drawdown is simply the average
return in a given year that a portfolio generates, expressed as a percentage of this drawdown figure.

Sortino ratio:

Sortino ratio = (Mean portfolio return – MAR)/Downside deviation

Unlike Sharpe ratio, Sortino ratio only measures downside deviation below the minimum acceptable
rate (MAR). Thus it does not penalize portfolio managers for volatility due to extreme positive
performance.

6.4. Capital Allocation

LO.m: Demonstrate the use of VaR and stress testing in setting capital requirements

Risk management is an important component in the process of allocating capital across units of a risk
taking enterprise. Different capital allocation methods are shown in the table below:

Methodology Comment

Nominal, Notional or Enterprise defines capital for each business unit.

Monetary Position Limits
Simple and allows us to calculate percentage return on capital
allocated.

Does not capture effects of correlation and offsetting risks.

Individual may be able to work around limits.

VAR-Based Position Limits Use VAR limit as alternative or supplement to notional limit.

Limit regime only as effective as VAR calculation.

Relation between overall VAR and individual VARs is complex.

IFT Notes for the Level III Exam [Link] Page 16

 Risk Management IFT Notes

Maximum Loss Limits Establish maximum loss limit for each risk-taking unit.

Internal Capital Specify level of capital that will be appropriate for the firm.
Requirements
Example: Enough capital such that probability of insolvency over 1-year
is 0.01.

Regulatory Capital Many institutions such as banks and security firms must calculate and
Requirements meet regulatory capital requirements.

6.5. Psychological and Behavioral Considerations

From a risk management perspective it is important to establish a risk governance framework when the
incentives of risk takers diverge from those of risk capital allocators.

For example, when portfolio managers who are paid a percentage of profits in a given year fall in a
negative performance situation any incremental loss will not affect the manager’s compensation;
however, the organization will suffer from incremental losses. In such situations a portfolio manager will
have an incentive to take a high level of risk to increase profits. This high level of risk might not be
appropriate for clients who are the providers of capital.

Summary
a. discuss features of the risk management process, risk governance, risk reduction, and an enterprise
risk management system;

Risk management process: Includes:

• Recognising exposures to risk
• Creating appropriate ranges for exposures
• Constantly measuring these exposures.

Risk governance (subset of corporate governance): process of setting overall policies and standards in
risk management.

b. evaluate strengths and weaknesses of a company’s risk management process;

The two types of risk governance models are:

▪ Decentralized: Each unit calculates and reports its risk exposures independently; allows people who
understand risk better to directly manage it.
▪ Centralized: Risk management is moved closer to senior management. The main advantage is that it
allows offsetting of risks across units. Centralized risk management is now called enterprise risk
management (ERM).

c. describe steps in an effective enterprise risk management system;

An effective ERM system usually includes the following steps:

1. Identify each risk factor to which the company is exposed.
2. Quantify each exposure’s size in money terms.

IFT Notes for the Level III Exam [Link] Page 17

 Risk Management IFT Notes

3. Map these inputs into a risk estimation calculation.

4. Identify overall risk exposures as well as the contribution to overall risk deriving from each risk
factor.
5. Set up a process to report on these risks periodically to senior management, who will set up a
committee of division heads and executives to determine capital allocations, risk limits, and risk
management policies.
6. Monitor compliance with policies and risk limits.

Those responsible for risk management should be independent of the trading function; back office
should be independent of front office.

d. evaluate a company’s or a portfolio’s exposures to financial and nonfinancial risk factors;

Financial risks:
▪ Market Risk: Risk associated with interest rates, exchange rates, stock prices and commodity prices.
▪ Credit Risk: Risk of loss caused by a counterparty or debtor’s failure to make a promised payment.
▪ Liquidity Risk: Risk that a financial instrument cannot be purchased or sold without a significant
concession in price because of the market’s potential inability to efficiently accommodate the
desired trading size (large bid-ask spread).
▪ Sovereign Risk: A form of credit risk in which the borrower is the government of a sovereign nation.

Non-financial risks:
▪ Operational Risk: Risk of loss from failures in a company’s systems and procedures or from external
events.
▪ Model Risk: Risk that a model is incorrect or misapplied; in investments, it often refers to valuation
models.
▪ Settlement (Herstatt) Risk: Risk that one party could be in the process of paying the counterparty
while the counterparty is declaring bankruptcy.
▪ Regulatory Risk: Risk associated with how a transaction will be regulated or with the potential for
regulations to change.
▪ Legal/Contract Risk: The possibility of loss arising from the legal system’s failure to enforce a
contract in which an enterprise has a financial stake.
▪ Tax Risk: Risk associated with uncertainty in tax laws.
▪ Accounting Risk: Risk associated with uncertainty about how a transaction should be recorded and
the potential for accounting rules and regulations to change.
▪ Political Risk: Risk associated with changes in political environment.
▪ ESG Risk: Risk to a company’s market valuation resulting from environmental, social and governance
factors.
▪ Performance Netting Risk: Potential for loss resulting from the failure of fees based on net
performance to fully cover contractual payout obligations to individual portfolio managers that have
positive performance when other portfolio managers have losses and when there are asymmetric
incentive fee arrangements with the portfolio managers.
▪ Settlement Netting Risk: Risk that a liquidator or a counterparty in default could challenge a netting
arrangement so that profitable transactions are realized for the benefit of creditors.

IFT Notes for the Level III Exam [Link] Page 18

 Risk Management IFT Notes

e. calculate and interpret value at risk (VaR) and explain its role in measuring overall and individual
position market risk;

VAR is an estimate of the loss (in money terms) that we expect to exceed with a given level of
probability over a specified time period.
Example: The VaR for a portfolio is $1.5 million for one day with a probability of 0.05. This statement
can be interpreted in the following two ways:
• 5 percent probability that the portfolio will lose at least $1.5 million in a single day.
• 95 percent probability, that the maximum loss will be $1.5 million for one day.

VAR measures requires the user to make the following decisions about the calculation’s structure.
1. Picking a probability level: Using 0.01 is more conservative and will give a higher VAR.
2. Selecting the time period over which to measure VAR:
• If the portfolio has a high turnover it is recommended to use a shorter period.
• Using a longer period results in a higher VAR.
3. Choosing the specific approach to modeling the loss distribution.

f. compare the analytical (variance–covariance), historical, and Monte Carlo methods for estimating VaR
and discuss the advantages and disadvantages of each;

Analytical variance-covariance method:

Consider a portfolio with 65% allocation to stocks (μ=12%, σ=22%) and 35% allocation to bonds (μ=5%,
σ=7%).
𝜇𝑝 = 𝑤𝑠 𝜇𝑠 + 𝑤𝐵 𝜇𝐵 = 0.65(0.12) + 0.35(0.05) = 0.0955
σ2P = wS2 σ2S + wB2 σ2B + 2ρwS wB σS σB =
(0.65)2(0.22)2+(0.35)2(0.07)2+2(0.15)(0.65)(0.35)(0.22)(0.07) = 0.0221
𝜎𝑃 = √0.0221 = 0.1487
For a 5 percent yearly VaR, we have μP – 1.65σP = 0.0955 – 1.65(0.1487) = –0.1499
5% yearly VaR is $150,000,000(0.1499) = $22.485 million
For weekly VaR, we adjust the expected return to 0.0955/52 and the standard deviation
to 0.1487/Sqrt(52)

Historical method: In this method we calculate the returns for a given portfolio using actual prices from
a user-specified period in the recent past.

Monte Carlo simulation method: In this method we produce random outcomes according to an
assumed probability distribution and a set of input parameters.

IFT Notes for the Level III Exam [Link] Page 19

 Risk Management IFT Notes

Variance-Covariance Method Historical Method Monte Carlo Simulation

Advantages: Simple. Advantages: Non-parametric Advantages:
(does not rely on probability- Widely used.
Disadvantages: distribution assumptions). We can assume an appropriate
Relies on normality of returns; distribution of input data.
doesn’t consider skewness and Disadvantages: Relies on
kurtosis. past events which might not Disadvantages:
If returns are not normally be good predictors of the Output only as good as our input
distributed then we cannot rely future. assumption.
totally on standard deviation as a Monte Carlo simulation
measure of risk. Portfolios software is expensive and
containing options don’t have requires a lot of computing
normal distributions. power.

g. discuss advantages and limitations of VaR and its extensions, including cash flow at risk, earnings at
risk, and tail value at risk;

Advantages Limitations
Quantifies potential loss in simple terms. One sided. Entire focus is on the left tail.
Most regulatory authorities accept VAR as Lulls into false sense of security. Gives the incorrect impression
an acceptable risk measure. that risk is properly understood and quantified.
Some companies use VAR as a measure of Often underestimates magnitude and frequency of worst
capital at risk across different lines of returns.
business. Extremely difficult to calculate VAR for large organizations.

Extensions of VaR:
A. Incremental VAR (IVAR) measures the incremental effect of an asset on the VAR of a portfolio by
measuring the difference between the portfolio's VAR while including a specified asset and the
portfolio's VAR with that asset eliminated.
B. Cash flow at risk (CFAR) is the minimum cash flow loss that we expect to be exceeded with a given
probability over a specified time.
C. Earnings at risk (EAR) is the minimum earnings loss that we expect to be exceeded with a given
probability over a specified time.
D. Tail value at risk (TVAR) is VAR plus the expected loss in excess of VAR when such excess loss occurs.

h. compare alternative types of stress testing and discuss advantages and disadvantages of each;

▪ Stress testing supplements VaR by identifying unusual circumstances that could lead to losses in
excess of those typically expected. The two broad approaches are: 1) scenario analysis and 2) stress
modeling.
o Scenario analysis is the process of evaluating a portfolio under different states of the world.
Stylized scenarios involves simulating a movement in at least one interest rate, exchange rate,
stock price, or commodity price relevant to the portfolio. One approach to scenario analysis is to
use actual extreme events that have occurred in the past. We can also create scenarios based
on hypothetical events – i.e. events that have never happened but might happen.

IFT Notes for the Level III Exam [Link] Page 20

 Risk Management IFT Notes

o Stress modeling: It is difficult to estimate sensitivity of a portfolio to scenarios we design; so

another approach is to use an existing model and apply shocks and perturbations to the model
inputs in some mechanical way.
Stressing models can take several forms:
➢ Factor push: Here we push the prices and risk factors of an underlying model in the most
disadvantageous way to calculate the combined effect on the portfolio’s value.
➢ Maximum loss optimization: Here we try to optimize mathematically the risk variables that
will produce the maximum loss.
➢ Worst case scenario analysis: Here we examine the worst case that we actually expect to
occur.

i. evaluate the credit risk of an investment position, including forward contract, swap, and option
positions;

Credit risk: risk that the party that owes the larger amount could default.
▪ Credit losses have two dimensions: likelihood of loss and associated amount of loss.
▪ Credit risk exposure has two time perspectives: current credit risk and potential credit risk.

Forward contracts
Credit risk exposure is based on the market value: PV of amount to be received – PV of amount to be
paid.

Swaps
Swap’s market value is the PV of amounted to be received – PV of amount to be paid.
For interest rate and equity swaps the potential credit risk is the largest during the middle period of the
swap’s life:
1. Risk is low at the start because both counterparties will have performed sufficient current credit
analysis.
2. Risk is low at the end because few payments remain.

Currency swaps have greatest credit risk closer to the end of swap’s life. This is because the notional
principal needs to be swapped at the end of the currency swap.

Options
Forward contracts and swaps have bilateral default risk. This means that either party could face credit
risk. Options have unilateral credit risk. Only the long party faces credit risk.

j. demonstrate the use of risk budgeting, position limits, and other methods for managing market risk;

Risk Budgeting
▪ To manage risk, we need to: identify sources of market risk, define how these risks will be
measured, set appropriate risk tolerance levels, identify corrective action if actual risk is outside
tolerance levels.
▪ Risk budgeting focuses on where to take risk and how to efficiently allocate risk.

k. demonstrate the use of exposure limits, marking to market, collateral, netting arrangements, credit

IFT Notes for the Level III Exam [Link] Page 21

 Risk Management IFT Notes

standards, and credit derivatives to manage credit risk;

Credit risk is one sided and returns are not symmetric; hence, not easily measured using standard
deviation and VAR.
The various methods to reduce credit risk are:
▪ using exchange traded rather than OTC derivatives
▪ limiting exposure
▪ marking to market
▪ use of collateral
▪ netting
▪ minimum credit standards
▪ credit derivatives such as credit default swaps, total return swaps, credit spread options and
credit spread forwards.

l. discuss the Sharpe ratio, risk-adjusted return on capital, return over maximum drawdown, and the
Sortino ratio as measures of risk-adjusted performance;
Mean portfolio return − Risk free rate
𝐒𝐡𝐚𝐫𝐩𝐞 𝐫𝐚𝐭𝐢𝐨 =
Standard deviation of portfolio return
Sharpe ratio is inaccurate when applied to portfolios with significant nonlinear risks such as option
positions.

Risk-Adjusted Return on Capital (RAROC) = Expected return on an investment / measure of capital at

risk
A company may require that an investment’s expected RAROC exceed a RAROC benchmark level for
capital to be allocated to it.
Return over Maximum Drawdown (RoMAD) = average annual return / drawdown
Drawdown is the difference between a portfolio’s maximum point (known as high water mark) and any
subsequent low point of performance.
Sortino ratio = (Mean portfolio return – MAR)/Downside deviation
Unlike Sharpe ratio, Sortino ratio only measures downside deviation below the minimum acceptable
rate (MAR). Thus it does not penalize portfolio managers for volatility due to extreme positive
performance.

m. demonstrate the use of VaR and stress testing in setting capital requirements.

Methodology Comment
Nominal, Notional or Enterprise defines capital for each business unit.
Monetary Position Limits Simple and allows us to calculate percentage return on capital allocated.
Does not capture effects of correlation and offsetting risks.
VAR-Based Position Use VAR limit as alternative or supplement to notional limit.
Limits Limit regime only as effective as VAR calculation.
Relation between overall VAR and individual VARs is complex.
Maximum Loss Limits Establish maximum loss limit for each risk-taking unit.

IFT Notes for the Level III Exam [Link] Page 22

 Risk Management IFT Notes

Internal Capital Specify level of capital that will be appropriate for the firm.
Requirements Example: Enough capital such that probability of insolvency over 1-year is
0.01.
Regulatory Capital Many institutions such as banks and security firms must calculate and meet
Requirements regulatory capital requirements.

Examples from the Curriculum

Example 1. Some Risk Governance Concerns of Investment Firms
Regardless of the risk governance approach chosen, effective risk governance for investment firms
demands that the trading function be separated from the risk management function. An individual or
group that is independent of the trading function must monitor the positions taken by the traders or risk
takers and price them independently. The risk manager has the responsibility for monitoring risk levels
for all portfolio positions (as well as for portfolios as a whole) and executing any strategies necessary to
control the level of risk. To do this, the risk manager must have timely and accurate information,
authority, and independence from the trading function. That is not to say that the trading function will
not need its own risk management expertise in order to allocate capital in an optimal fashion and
maximize risk-adjusted profit. Ideally, the risk manager will work with the trading desks in the
development of risk management specifications, such that everyone in the organization is working from
a common point of reference in terms of measuring and controlling exposures.

Effective risk governance for an investment firm also requires that the back office be fully independent
from the front office, so as to provide a check on the accuracy of information and to forestall collusion.
(The back office is concerned with transaction processing, record keeping, regulatory compliance, and
other administrative functions; the front office is concerned with trading and sales.) Besides being
independent, the back office of an investment firm must have a high level of competence, training, and
knowledge because failed trades, errors, and over-sights can lead to significant losses that may be
amplified by leverage. The back office must effectively coordinate with external service suppliers, such
as the firm’s global custodian. The global custodian effects trade settlement (completion of a trade
wherein purchased financial instruments are transferred to the buyer and the buyer transfers money to
the seller), safekeeping of assets, and the allocation of trades to individual custody accounts.
Increasingly, financial institutions are seeking risk reduction with cost efficiencies through straight-
through processing (STP) systems that obviate manual and/or duplicative intervention in the process
from trade placement to settlement.

Back to Notes.

Example 2. An Analysis of Risk Exposures

Liam McNulty is the risk manager for a large multinational agricultural concern, Agripure. The company
grows its own corn, wheat, and soybeans but pays large sums to third parties for pesticides, fertilizer,
and other supplies. For this, it must borrow heavily to finance its purchases. Customers typically
purchase Agripure’s goods on credit. Moreover, Agripure buys and sells its products and raw materials
worldwide, often transacting in the domestic currency of its customers and suppliers. Finally, to finance
its own expansion, Agripure intends to issue stock.

IFT Notes for the Level III Exam [Link] Page 23

 Risk Management IFT Notes

Recommend and justify the risk exposures that McNulty should report as part of an enterprise risk
management system for Agripure.

Solution:

McNulty should report on the following risk exposures:

▪ Market risk, including these subtypes:

• Commodity price risk, because Agripure has exposures in raw materials and finished
products.

• Foreign exchange risk, because it buys and sells products world-wide, often transacting
in the home currency of the entity on the other side of the transaction.

• Equity market risk, because Agripure’s expansion financing is affected by the price it
receives for its share issuance.

• Interest rate risk, because Agripure has exposures in financing its raw material
purchases and because its customers typically purchase their goods on credit.

▪ Credit risk, because Agripure’s customers typically purchase their goods on credit.

▪ Operational risk, because as an agricultural producer Agripure is subject to weather-related risk

(an external event).

Back to Notes.

Example 3. An Operational Risk for Financial Services Companies: The Rogue Trader
Among the more prominent examples of operational risk for financial service companies is that of the
so-called rogue trader: an individual who has either assumed an irresponsibly high level of risk, engaged
in unauthorized transactions, or some combination of the two. The risks associated with this type of
activity increase the longer it goes undetected, and often the very lack of controls that creates the
opportunity for a rogue trader in the first place renders it difficult to quickly determine that a problem
exists. In some extreme cases, such as an incident that occurred in the Singapore office of Barings Bank,
a rogue trader can cause an entire organization to fold. The incidence of high-profile rogue trading
episodes has multiplied since the early 1990s, but in nearly all of these episodes, the problem’s major
source was a lack of rudimentary corporate controls and oversight.

Back to Notes.

Example 4. Accounting Risk: The Case of Derivative Contracts

Accounting for derivative contracts has raised considerable confusion. When confusion occurs,
companies run the risk that the accounting treatment for transactions could require adjustment, which
could possibly lead to a need to restate earnings. Earnings restatements are almost always embarrassing
for a company, because they suggest either a desire to hide information, the company’s failure to fully
understand material elements of its business, or some combination of the two. Restatements are very
detrimental to corporate valuations because they cause investors to lose confidence in the accuracy of
corporate financial disclosures. Beyond that, if negligence or intent to mislead was involved, the

IFT Notes for the Level III Exam [Link] Page 24

 Risk Management IFT Notes

company could face civil and criminal liabilities as well.

Confusion over the proper accounting for derivatives gives rise to accounting as a source of risk. As with
regulatory and tax risk, sometimes equivalent combinations of derivatives are not accounted for
uniformly. The accounting profession typically moves to close such loopholes, but it does not move
quickly and certainly does not keep pace with the pace of innovation in financial engineering, so
problems nearly always remain.

The IASB in IAS 39 (International Accounting Standard No. 39) requires the inclusion of derivatives and
their associated gains and losses on financial statements, as does the FASB in SFAS 133 (Statement of
Financial Accounting Standard No. 133). These rulings contain some areas of confusion and
inconsistency, however, affording considerable room for interpretation.

Back to Notes.

Example 5. VaR with Different Probability Levels and Time Horizons

Consider a portfolio consisting of stocks as one asset class and bonds as another. The expected return
on the portfolio’s stock portion is 12 percent, and the standard deviation is 22 percent. The expected
return on the bond portion is 5 percent, and the standard deviation is 7 percent. All of these figures are
annual. The correlation between the two asset classes is 0.15. The portfolio’s market value is $150
million and is allocated 65 percent to stocks and 35 percent to bonds. Determine the VaR using the
analytical method for the following cases:

1. a 5 percent yearly VaR.

2. a 1 percent yearly VaR.

3. a 5 percent weekly VaR.

4. a 1 percent weekly VaR.

Solutions:

First, we must calculate the annual portfolio expected return and standard deviation. Using S to indicate
stocks and B to indicate bonds, we have

𝜇𝑝 = 𝑤𝑠 𝜇𝑠 + 𝑤𝐵 𝜇𝐵 = 0.65(0.12) + 0.35(0.05) = 0.0955

𝜎𝑃2 = 𝑤𝑆2 𝜎𝑆2 + 𝑤𝐵2 𝜎𝐵2 + 2𝜌𝑤𝑆 𝑤𝐵 𝜎𝑆 𝜎𝐵

(0.65)2(0.22)2+(0.35)2(0.07)2+2(0.15)(0.65)(0.35)(0.22)(0.07)

=0.0221

𝜎𝑃 = √0.0221 = 0.1487
Solution to 1:

For a 5 percent yearly VaR, we have μP – 1.65σP = 0.0955 – 1.65(0.1487) = –0.1499. Then the VaR is
$150,000,000(0.1499) = $22.485 million.

IFT Notes for the Level III Exam [Link] Page 25

 Risk Management IFT Notes

Solution to 2:

For a 1 percent yearly VaR, we have μP – 2.33σP = 0.0955 – 2.33(0.1487) = –0.251. Then the VaR is
$150,000,000(0.251) = $37.65 million.

Solution to 3:

For weekly VaR, we adjust the expected return to 0.0955/52 = 0.00184 and the standard deviation
to 0.1487/Sqrt(52)=0.02062 .

The 5 percent weekly VaR is then μP – 1.65σ = 0.00184 – 1.65(0.02062) = –0.03218. Then the VaR is
$150,000,000(0.03218) = $4.827 million.

Solution to 4:

The 1 percent weekly VaR is μP – 2.33σP = 0.00184 – 2.33(0.02062) = –0.0462. Then the VaR is
$150,000,000(0.0462) = $6.93 million.

Back to Notes.

Example 6. Calculating VaR Using the Historical Method

For simplicity, we use a one-stock portfolio. Exhibit 8 shows the 40 worst monthly returns on IBM stock
during the last 20 years, in descending order, as of 2011 (minus signs omitted):

Exhibit 8. IBM Stock: Worst Monthly Returns

0.26190 0.11692 0.09077 0.07537

0.22645 0.11553 0.08926 0.07298
0.20511 0.10838 0.08585 0.07260
0.19462 0.10805 0.08481 0.07247
0.18802 0.10687 0.08422 0.07075
0.17183 0.10503 0.08356 0.06894
0.16415 0.09873 0.08234 0.06782
0.14834 0.09550 0.08197 0.06746
0.14773 0.09276 0.08143 0.06501
0.12444 0.09091 0.07547 0.06437

For both calculations below, assume the portfolio value is $100,000.

1. Calculate a 5 percent monthly VaR using the historical method.

2. Calculate a 1 percent monthly VaR using the historical method.

Solutions:

First, we note that during the last 20 years, there were 240 monthly returns. We see here only the worst
40 returns. Therefore, although we lack the entire distribution of returns, we do have enough to

IFT Notes for the Level III Exam [Link] Page 26

 Risk Management IFT Notes

calculate the VaR.

Solution to 1:

Out of 240 returns, the 5 percent worst are the 12 worst returns. Therefore, the historical VaR would be
about the 12th-worst return. From the exhibit, we see that this return is –0.11553. So, the one-month
VaR is 0.11553($100,000) = $11,553.

Solution to 2:

The 1 percent worst returns include 2.4 returns. We would probably use the second-worst return, which
is –0.22645. The VaR is 0.22645($100,000) = $22,645. Alternatively, we might average the second- and
third-worst returns to obtain (–0.22645 + –0.20511)/2 = –0.21578. Then the one-month VaR would be
0.21578($100,000) = $21,578.

Back to Notes.

Example 7. Value at Risk and the Management of Market Risk at Goldman Sachs
The following excerpt is from the 2010 Form 10-K of Goldman Sachs:

Value-at-Risk

VaR is the potential loss in value of inventory positions due to adverse market movements over a
defined time horizon with a specified confidence level. We typically employ a one-day time horizon with
a 95% confidence level. Thus, we would expect to see reductions in the fair value of inventory positions
at least as large as the reported VaR once per month. The VaR model captures risks including interest
rates, equity prices, currency rates and commodity prices. As such, VaR facilitates comparison across
portfolios of different risk characteristics. VaR also captures the diversification of aggregated risk at the
firmwide level.

Inherent limitations to VaR include:

▪ VaR does not estimate potential losses over longer time horizons where moves may be extreme.

▪ VaR does not take account of the relative liquidity of different risk positions.

▪ Previous moves in market risk factors may not produce accurate predictions of all future market
moves.

The historical data used in our VaR calculation is weighted to give greater importance to more recent
observations and reflect current asset volatilities. This improves the accuracy of our estimates of
potential loss. As a result, even if our inventory positions were unchanged, our VaR would increase with
increasing market volatility and vice versa.

Given its reliance on historical data, VaR is most effective in estimating risk exposures in markets in
which there are no sudden fundamental changes or shifts in market conditions.

We evaluate the accuracy of our VaR model through daily backtesting (i.e., comparing daily trading net
revenues to the VaR measure calculated as of the prior business day) at the firmwide level and for each

IFT Notes for the Level III Exam [Link] Page 27

 Risk Management IFT Notes

of our businesses and major regulated subsidiaries.

VaR does not include:

▪ positions that are best measured and monitored using sensitivity measures; and

▪ the impact of changes in counterparty and our own credit spreads on derivatives as well as
changes in our own credit spreads on unsecured borrowings for which the fair value option was
elected.

Stress Testing

We use stress testing to examine risks of specific portfolios as well as the potential impact of significant
risk exposures across the firm. We use a variety of scenarios to calculate the potential loss from a wide
range of market moves on the firm's portfolios. These scenarios include the default of single corporate
or sovereign entities, the impact of a move in a single risk factor across all positions (e.g., equity prices
or credit spreads) or a combination of two or more risk factors.

Unlike VaR measures, which have an implied probability because they are calculated at a specified
confidence level, there is generally no implied probability that our stress test scenarios will occur.
Instead, stress tests are used to model both moderate and more extreme moves in underlying market
factors. When estimating potential loss, we generally assume that our positions cannot be reduced or
hedged (although experience demonstrates that we are generally able to do so).

Stress test scenarios are conducted on a regular basis as part of the firm's routine risk management
process and on an ad hoc basis in response to market events or concerns. Stress testing is an important
part of the firm's risk management process because it allows us to highlight potential loss
concentrations, undertake risk/reward analysis, and assess and mitigate our risk positions.

Limits

We use risk limits at various levels in the firm (including firmwide, product and business) to govern risk
appetite by controlling the size of our exposures to market risk. Limits are reviewed frequently and
amended on a permanent or temporary basis to reflect changing market conditions, business conditions
or tolerance for risk.

The Firmwide Risk Committee sets market risk limits at firmwide and product levels and our Securities
Division Risk Committee sets sub-limits for market-making and investing activities at a business level.
The purpose of the firmwide limits is to assist senior management in controlling the firm's overall risk
profile. Sub-limits set the desired maximum amount of exposure that may be managed by any particular
business on a day-to-day basis without additional levels of senior management approval, effectively
leaving day-to-day trading decisions to individual desk managers and traders. Accordingly, sub-limits are
a management tool designed to ensure appropriate escalation rather than to establish maximum risk
tolerance. Sub-limits also distribute risk among various businesses in a manner that is consistent with
their level of activity and client demand, taking into account the relative performance of each area.

Our market risk limits are monitored daily by Market Risk Management, which is responsible for
identifying and escalating, on a timely basis, instances where limits have been exceeded. The business-
level limits that are set by the Securities Division Risk Committee are subject to the same scrutiny and

IFT Notes for the Level III Exam [Link] Page 28

 Risk Management IFT Notes

limit escalation policy as the firmwide limits.

When a risk limit has been exceeded (e.g., due to changes in market conditions, such as increased
volatilities or changes in correlations), it is reported to the appropriate risk committee and a discussion
takes place with the relevant desk managers, after which either the risk position is reduced or the risk
limit is temporarily or permanently increased.

Metrics

We analyze VaR at the firmwide level and a variety of more detailed levels, including by risk category,
business, and region. The tables below present average daily VaR and year-end VaR by risk category.

Average Daily VaR (in millions)

Year Ended
Risk Categories December 2010 December 2009 November 2008
Interest rates $ 93 $176 $ 142
Equity prices 68 66 72
Currency rates 32 36 30
Commodity prices 33 36 44
Diversification effecta (92) (96) (108)
Total $134 $218 $ 180
a
Equals the difference between total VaR and the sum of the VaRs for the four risk categories. This
effect arises because the four market risk categories are not perfectly correlated.

Our average daily VaR decreased to $134 million in 2010 from $218 million in 2009, principally due to a
decrease in the interest rates category which was primarily due to reduced exposures, lower levels of
volatility and tighter spreads.

Our average daily VaR increased to $218 million in 2009 from $180 million in 2008, principally due to an
increase in the interest rates category and a reduction in the diversification benefit across risk
categories, partially offset by a decrease in the commodity prices category. The increase in the interest
rates category was primarily due to wider spreads. The decrease in the commodity prices category was
primarily due to lower energy prices.

Year-End VaR and High and Low VaR (in millions)

As of December Year Ended December 2010

Risk Categories 2010 2009 High Low
Interest rates $ 78 $ 122 $123 $ 76
Equity prices 51 99 186 39
Currency rates 27 21 62 14
Commodity prices 25 33 62 18

IFT Notes for the Level III Exam [Link] Page 29

 Risk Management IFT Notes

As of December Year Ended December 2010

Risk Categories 2010 2009 High Low
a
Diversification effect (70) (122)
Total $111 $ 153 $223 $105
a
Equals the difference between total VaR and the sum of the VaRs for the four risk categories. This
effect arises because the four market risk categories are not perfectly correlated.

Our daily VaR decreased to $111 million as of December 2010 from $153 million as of December 2009,
principally due to a decrease in the equity prices and interest rates categories, partially offset by a
decrease in the diversification benefit across risk categories. The decreases in the equity prices and
interest rates categories were primarily due to reduced exposures and lower levels of volatility.

During the year ended December 2010, the firmwide VaR risk limit was exceeded on one occasion in
order to facilitate a client transaction and was resolved by a reduction in the risk position on the
following day. Separately, during the year ended December 2010, the firmwide VaR risk limit was
reduced on one occasion reflecting lower risk utilization.

During the year ended December 2009, the firmwide VaR risk limit was exceeded on two successive
days. It was resolved by a reduction in the risk position without a permanent or temporary VaR limit
increase. Separately, during the year ended December 2009, the firmwide VaR risk limit was raised on
one occasion and reduced on two occasions as a result of changes in the risk utilization and the market
environment.

The chart below reflects the VaR over the last four quarters.

Daily VaR ($ in millions)

The chart below presents the frequency distribution of our daily trading net revenues for substantially
all inventory positions included in VaR for the year ended December 2010.

IFT Notes for the Level III Exam [Link] Page 30

 Risk Management IFT Notes

Daily Trading Net Revenues ($ in millions)

As noted above, daily trading net revenues are compared with VaR calculated as of the end of the prior
business day. Trading losses incurred on a single day exceeded our 95% one-day VaR on two occasions
during 2010. Trading losses incurred on a single day did not exceed our 95% one-day VaR during 2009.

Source: Goldman Sachs 2010 Form 10-K, pp. 85-87. The Goldman Sachs Group, Inc. All rights reserved.

Back to Notes.

Example 8. Calculating Credit Risk Exposures

Calculate the amount at risk of a credit loss in the following situations:

1. A US party goes long a forward contract on €1 denominated in dollars in which the underlying is
the euro. The original term of the contract was two years, and the forward rate was $0.90. The
contract now has 18 months or 1.5 years to maturity. The spot or current exchange rate is
$0.862. The US interest rate is 6 percent, and the euro interest rate is 5 percent. The interest
rates are based on discrete compounding/discounting. At the point when the contract has 1.5
years remaining, the value of the contract to the long per $1 notional principal equals the spot
exchange rate, $0.862, discounted at the international interest rate for 1.5 years, minus the
forward rate, $0.90, discounted at the domestic interest rate for 1.5 years:

$0.862/(1.05)1.5−$0.90/(1.06)1.5=−$0.0235

Evaluate the credit risk characteristics of this situation.

2. Consider a plain vanilla interest rate swap with two months to go before the next payment. Six
months after that, the swap will have its final payment. The swap fixed rate is 7 percent, and the
upcoming floating payment is 6.9 percent. All payments are based on 30 days in a month and
360 days in a year. Two-month Libor is 7.250 percent, and eight-month Libor is 7.375 percent.
The present value factors for two and eight months can be calculated as follows:

1/(1+0.0725(60/360))=0.9881

1/(1+0.07375(240/360))=0.9531

IFT Notes for the Level III Exam [Link] Page 31

 Risk Management IFT Notes

The next floating payment will be 0.069(180/360) = 0.0345. The present value of the floating payments
(plus hypothetical notional principal) is 1.0345(0.9881) = 1.0222. Given an annual rate of 7 percent, the
fixed payments will be 0.07(180/360) = 0.035.

The present value of the fixed payments (plus hypothetical notional principal) is, therefore,
0.035(0.9881) + 1.035(0.9531) = 1.0210. Determine the amount at risk of a credit loss and state which
party currently bears the risk. Assume a $1 notional principal.

3. A dealer has sold a call option on a stock for $35 to an investor. The option is currently worth
$46, as quoted in the market. Determine the amount at risk of a credit loss and state which
party currently bears the risk.

Solution to 1:

The position has a negative value to the long, so the credit risk is currently borne by the short. From the
short’s point of view, the contract has a value of $0.0235 per $1 notional principal. No payments are due
for 18 months, but the short’s claim on the long is worth $0.0235 more than the long’s claim on the
short. Therefore, this amount is the current value of the amount at risk for a credit loss. Of course, the
amount could, and probably will, change over the life of the contract. The credit risk exposure might
even shift to the other party.

Solution to 2:

The market value of the swap to the party paying fixed and receiving floating is 1.0222 – 1.0210 =
0.0012. This value is positive to the party paying fixed and receiving floating; thus this party currently
assumes the credit risk. Of course, the value will change over the life of the swap and may turn negative,
meaning that the credit risk is then assumed by the party paying floating and receiving fixed.

Solution to 3:

All of the credit risk is borne by the investor (the owner of the call), because he will look to the dealer
(the seller) for the payoff if the owner exercises the option. The current value of the amount at risk is
the market price of $46.

Back to Notes.

Example 9. Basel II—A Brief Overview

The Basel banking regulations apply only to large international banks, but national governments use
them as a guideline in formulating their own financial laws and regulations, so the regulations have
much more widespread importance. In January 2001, the Basel Committee on Banking Supervision
issued a proposal for a New Basel Capital Accord that would replace the 1988 Capital Accord. This first
accord, “Basel I,” was widely criticized for being too inflexible in applying an across-the-board 8 percent
capital adequacy ratio that made no discrimination between a well risk-managed bank and one that was
not.

The Basel II proposal incorporates three mutually reinforcing pillars that allow banks and supervisors to
evaluate properly the various risks that banks face:

IFT Notes for the Level III Exam [Link] Page 32

 Risk Management IFT Notes

▪ Pillar 1: Capital Requirements;

▪ Pillar 2: Supervisory Review;

▪ Pillar 3: Market Discipline.

The first pillar of Basel II moves away from a blanket, one-size-fits-all approach and allows banks to
develop their own mathematically based financial models. Once these internally developed techniques
have been successfully demonstrated to the regulators, banks are able to progress to higher levels of
risk management that within the accord are offset by reduced regulatory capital charges. Key to these
higher levels of risk management are advanced systems for managing credit risk and operational risk.

The second pillar, supervisory review, requires banks to meet Basel-recommended operational risk
requirements that have been tailored by their host country. “Risky” banks, whose risk management
systems score lowly in the areas of market risk and operational risk, face penalties. Better-risk-managed
banks will have major competitive advantages over rivals, in that, all else equal, they are likely to be
subject to reduced capital requirements per unit of risk.

The third pillar says that banks must fulfill the Basel requirements for transparency and disclosing
company data. A key point here is that banks must reveal more detail about their profits and losses,
which may lead to a supervisory authority reviewing risk systems and changing the capital allocation
under the first pillar.

Back to Notes.

Example 10. A Fund Management Company and Risk Budgeting

We can readily illustrate the methodology and underlying economics of risk budgeting with the example
of a fund management company. We choose, for this example, a multistrategy hedge fund, because
although mutual funds and other types of institutional money managers certainly face similar risk
management issues, they are often bound by strict guidelines that tie their risk budgeting to factors
such as the performance of a benchmark index and other mandated fund management protocols. For
example, the Vanguard family of mutual funds offers a wide range of indexed mutual funds. These
funds’ associated risk budgets are very narrowly defined, as the managers are called on at all times to
track the underlying index very closely in terms of securities held, associated portfolio weightings, and
so forth. As investor funds flow in and out of these securities, portfolio managers execute trades that do
little more than reestablish this replication balance. Of course, many institutional fund products allow
for much broader deviations from market benchmarks; in most cases, however, risk budgeting will be
constrained by certain principles associated with benchmarking.

Hedge funds with multiple portfolio managers (as well as, in some cases, the proprietary trading
divisions of banks and broker/dealers) have many fewer risk constraints than indexed mutual funds;
they have more freedom, therefore, in establishing risk budgets. Because of the absolute return (as
opposed to benchmark-driven) nature of their performance, and because of issues such as performance
netting risk covered earlier in this reading, it is very much in their interest to ensure that each portfolio
in the enterprise operates within a well-conceived risk budget framework. Included among the critical
components of such a program might be the following:

IFT Notes for the Level III Exam [Link] Page 33

 Risk Management IFT Notes

▪ Performance Stopouts A performance stopout is the maximum amount that a given portfolio is
allowed to lose in a period (e.g., a month or a year).

▪ Working Capital Allocations Most funds will allocate a specific amount of working capital to
each portfolio manager, both as a means of enforcing risk disciplines and also to ensure the
ability to fund all operations.

▪ VaR Limits Discussed above.

▪ Scenario Analysis Limits The risk manager of the fund company may establish risk limits based
on the scenario analysis discussed in the preceding section. Under such an approach, the
portfolio manager would be compelled to construct a portfolio such that under specified
scenarios, it did not produce losses greater than certain predetermined amounts.

▪ Risk Factor Limits Portfolio managers may be subject to limits on individual risk factors, as
generated by a VaR analysis (e.g., VaR exposure to a certain risk cannot exceed, say, $X or X%) or
driven by linear (e.g., duration, beta) or nonlinear (e.g., convexity, gamma) risk estimation
methodologies.

▪ Position Concentration Limits Many risk managers seek to enforce diversification by mandating
a specific maximum amount for individual positions.

▪ Leverage Limits A maximum amount of leverage in the portfolio may be specified.

▪ Liquidity Limits To help manage liquidity exposure, large funds will often also set position limits
as a specified maximum percentage of daily volume, float, or open interest.

Of course, other types of limits are imposed on portfolio managers in a multistrategy environment, and
by the same token, the risk-budgeting strategy of a given enterprise may include only a subset of the
examples provided immediately above. Nevertheless, some subset of these limit structures is present in
nearly every multistrategy fund vehicle, and it is difficult to imagine an effective risk control system that
does not set limits.

Back to Notes.

Example 11. Repricing a Forward Contract

Consider a one-year forward contract established at a rate of $105. The contract is four months into its
life. The spot price is $108, the risk-free rate is 4.25 percent, and the underlying makes no cash
payments. The two parties decided at the start that they will mark the contract to market every four
months. The market value of the contract is $108 – $105/(1.0425)8/12 = $5.873. Determine how the cash
flows and resets would work under these circumstances.

Solution:

The contract is positive to the long, so the short pays the long $5.873. The parties then reprice the
contract. The new price is $108(1.0425)8/12 = $111.04. At this point, the forward price is reset to
$111.04. The parties will then mark to market again at the eight-month point and reset the forward
price. This price will then stay in force until expiration.

IFT Notes for the Level III Exam [Link] Page 34

 Risk Management IFT Notes

Back to Notes.

IFT Notes for the Level III Exam [Link] Page 35