Notes 1: Introduction to Financial Risk Analytics
Guangwu Liu
Professor, MS Department
1 / 25
What is Risk?
Two ways to define risk:
1. Risk is a measure of uncertainty.
2. Risk is a measure to capture the potential of sustaining a loss.
In our context of risk management, we adopt the second definition.
I We focus exclusively on the negative e↵ects (loss) of the
uncertainty.
2 / 25
Scope of the Course
Both financial and non-financial institutions face the issue of risk
management. However, in the course we focus mainly on financial
institutions. Why?
1. Financial sector is a typical sector where risk management is
of extremely high priority.
2. Governments and regulators impose restrictive regulations on
financial institutions.
Nevertheless, knowledge in this context can also be applied to
many traditional industries, e.g., airlines, supply chain and
logistics,...
3 / 25
Risk Categories in Financial Institutions
Risk categories that are of importance in financial institutions.
1. Market risk
I e.g., movement in stock prices or equity value due to
movement in interest rate. Example: Accumulator (I’ll kill you
later!)
2. Credit risk CDO
I e.g., default of a counterparty. Example: Lehman Brothers
Mini-Bond.
3. Operational risk
I e.g., loss due to rogue traders or external frauds. Examples:
Barings Bank Event, and Mado↵ Fraud.
4 / 25
Importance of Risk Management
Failures of large financial institutions happen again and again.
I Barings Bank in 1995, Societe Generale in 2007, UBS in
2011,...
I The recent financial tsunami: Lehman Brothers, Merrill
Lynch, etc.
I In 2007-08, a large number of financial institutions may have
already bankrupted without the rescue of the governments.
I Recent trends: financial products and markets are becoming
more and more complicated, and hence risk management are
becoming more and more important.
5 / 25
Financial Risk Management
How to deal with financial risk?
I Avoid?
I But you cannot make profit without taking risks
I Reserves: especially for banks
I Diversification
I But only for non-systematic risk
I Hedging
I Usually, using derivatives
6 / 25
Financial Risk Management (cont.)
I Insurance: for exogenous low probability events
I Evaluation based on risk-adjusted performance
I Strategic risk management: enterprise-wide policy towards
risks
1. Identification
2. Measurement
3. Management
4. Monitoring
7 / 25
General Approaches to Risk Management
1. Risk Decomposition
I Identify and manage the risks one by one, e.g., interest rate
risk, exchange risk, etc.
I Usually done in front office
2. Risk Aggregation
I Aggregate di↵erent risks of the firm, evaluation the
diversification e↵ect, identify and measure the systematic risks,
and propose methods to remove unacceptable risks
I Usually done in middle office.
In practice, financial institutions use both approaches.
8 / 25
Risk-Return Trade-O↵
What is risk management about?
1. Risk management is essentially about the risk-return trade-o↵.
I To totally avoid risks is impossible.
I Higher returns usually imply higher risks.
I Trade-o↵ between return and risk is the key to survival and
success of a financial institution.
2. Early attempts to understand this trade-o↵ date back to
Markowitz in 1952. However, somehow the recent financial
tsunami reminds us that we still have a long way to go.
9 / 25
Risk and Return
I Risk is measured using the standard deviation of return.
I We illustrate the idea using examples.
I Example:
Table: Return in one year from investing $100,000 in an equity
Probability Return
0.05 +50%
0.25 +30%
0.40 +10%
0.25 -10%
0.05 -30%
10 / 25
Example (cont.)
I What is the expected return per annum E(R)?
I What is the standard deviation of the return?
11 / 25
Investment Opportunities
I Suppose that there are two investments R1 and R2 with
expected returns µ1 and µ2 , and standard deviation 1 and 2
respectively.
I We construct a portfolio, that puts a proportion w1 of our
money in the first investment and a proportion w2 = 1 w1 in
the second investment.
I Payo↵ of the portfolio:
w1 R 1 + w2 R 2 .
12 / 25
Investment Opportunities (cont.)
I Expected return of the portfolio
µ P = w1 µ 1 + w2 µ 2 .
I Standard deviation of the portfolio return
q
P = w12 12 + w22 22 + 2⇢w1 w2 1 2,
where ⇢ is the coefficient of correlation between the two
investments.
13 / 25
Investment Opportunities (cont.)
I We let µ1 = 10%, µ2 = 15%, 1 = 16%, 2 = 24% and
⇢ = 0.2.
I For di↵erent (w1 , w2 ), the expected return and standard
deviation of the portfolio are di↵erent.
w1 w2 µP P
0.0 1.0 15% 24.00%
0.2 0.8 14% 20.09%
0.4 0.6 13% 16.89%
0.6 0.4 12% 14.87%
0.8 0.2 11% 14.54%
1.0 0.0 10% 16.00%
14 / 25
Investment Opportunities (cont.)
Which combination of (w1 , w2 ) provides a better portfolio?
15 / 25
The Efficient Frontier from Risky Investments
16 / 25
The Efficient Frontier of All Investments
A key observation: the trade-o↵ between return and risk is linear.
17 / 25
Capital Asset Pricing Model (CAPM)
I The portfolio M in the previous slide is called market portfolio.
I From the linearity of trade-o↵ between return and risk, we
expect the following relationship between any investment and
the market portfolio:
R = a + RM + ✏,
where R is the return from the investment, RM is the return
from the market portfolio, a and are constants, and ✏ is a
random variable representing the regression error.
18 / 25
CAPM (cont.)
I Two components of the risk in the investment’s return:
1. The component RM : a multiple of the return from the
market portfolio, called systematic risk.
2. The component ✏: unrelated to the return from the market
portfolio, called non-systematic risk.
I If we assume that ✏’s are independent of each other, then the
non-systematic risk can almost be diversified in a large
portfolio.
I However, even in a large portfolio, systematic risk does not
disappear. i.e., it is not diversifiable.
19 / 25
CAPM (cont.)
I CAPM: in general, we have
E(R) = RF + [E(RM RF )],
where RF is the risk-free return rate.
I If = 0, there is no systematic risk and the expected return is
RF .
I If = 1, we have the same systematic risk as the market
portfolio.
I The excess expected return over the risk-free rate required on
the investment is times the excess expected return on the
market portfolio.
20 / 25
Example
I Suppose that the risk-free rate is 5%, and the return on the
market portfolio is 10%.
I An investment with = 0 should have an expected return of
5%.
I An investment with = 0.5 should have an expected return of
0.05 + 0.5 ⇥ (0.1 0.05) = 7.5%.
I An investment with = 1.2 should have an expected return of
0.05 + 1.2 ⇥ (0.1 0.05) = 11%.
21 / 25
Alpha (↵) in CAPM
I Intuitively, CAPM provides a fair expected return on bearing
certain systematic risk.
I However, all the analysis are based on some ideal
assumptions, which may not hold in reality.
I In reality, a portfolio manager may create a portfolio with
return RP ,
RP > RF + (RM RF ).
I Then ↵ is the extra return:
↵ = RP RF (RM RF ).
I Portfolio managers are continually searching for ways of
producing a positive alpha.
22 / 25
Assumptions of CAPM
Actually the above analysis of CAPM made a number of
assumptions that are too ideal to hold in practice.
1. We assumed that every investor considers standard deviation
of the return as the risk. However, di↵erent investors may
have di↵erent views on the attractiveness of the risky
opportunities: risk-taking, risk-neutral, and risk-averse.
2. We assumed that ✏’s are independent.
3. We assumed that investors can borrow and lend at the same
risk-free rate.
4. We didn’t consider tax and transaction cost.
5. ...
However, CAPM has proved to be a useful tool for portfolio
managers. It is commonly used as a benchmark for assessing the
performance of the portfolio manager.
23 / 25
Theory vs. Practice
The previous theoretical analysis of CAPM implies that an investor
does not have to care about non-systematic risk, since it can be
diversified away. However, in practice we cannot ignore it. Several
reasons:
I Bankruptcy costs are important
I Diversification may fail since assets are not very liquid and
divisible
I Market frictions: financial distress costs, taxes, external
financing costs, etc.
24 / 25
Portfolio Optimization
Rationale in portfolio optimization.
Essential idea of Markowitz model (1952):
Practical implications.
25 / 25