Financial Economics and
Investments, Spring 25
Topic 2
Utility and Risk
Contents
1. Preference and Utility
2. Uncertainty and Risk Averse
3. Risk and Risk Premium
Preference and Utility
• Benefit of something is the gain or pleasure that it brings and
is determined by preferences - by what a person likes and
dislikes and the intensity of those feelings.
– Preference and Utility
• Cost is what you must give up to get something.
Preference and Utility
Households want to maximize utility
– The benefit or satisfaction of households is called utility.
– How do you gain utilities?
Preference and Utility
• An individual who reports “A is preferred to B” is taken to
mean that all things considered, he or she feels better off
under situation A than under situation B.
• Axioms of rational choice
– Completeness
– Transitivity
– Continuity
Preference and Utility
Given the axioms of rational choice, we can formally demonstrate that
people are able to rank all possible situations from the least desirable to
the most, a ranking that economists refer to as utility.
• Utility refers to overall satisfaction
• Nonuniqueness of utility measures:
– Ranked but not evenly spaced
• The ceteris paribus assumption
– A common practice is to devote attention exclusively to choices among
quantifiable options while holding constant the other things that affect behavior.
– This ceteris paribus (“other things being equal”) assumption is invoked in all
economic analyses of utility-maximizing choices so as to make the analysis of
choices manageable within a simplified setting.
The Law of Diminishing Marginal Utility Explained
in One Minute
Contents
1. Preference and Utility
2. Uncertainty and Risk Averse
3. Risk and Risk Premium
Uncertainty
• A random experiment: flip a fair coin.
• Basic outcomes: “𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻” and “𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇”
1
• PMF: 𝑃𝑃 𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 = 𝑃𝑃 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 =
2
• You pay $50 per round to play this game and win $100 if it lands on
“𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻”; otherwise, you win nothing.
• Let a random variable 𝑋𝑋 denote the dollar amount that you will gain in one
round, and 𝐸𝐸[𝑋𝑋] denote your expected gain.
1 1
𝐸𝐸 𝑋𝑋 = $100 + $0 − $50 = $0 fair game
2 2
• Will you pay $50 per round to play this game?
Uncertainty
Suppose you won a lottery and are presented with two options:
1. Receive $1,000 immediately and walk away.
2. Flip a fair coin. You will win $2,000 if it lands on “𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻” and
nothing otherwise.
1 1
– 𝐸𝐸 𝑋𝑋 = $2,000 + $0 = $1,000
2 2
• Which option would you choose?
Uncertainty
Suppose you won a lottery and are presented with two options:
1. Receive $𝑥𝑥 immediately and walk away.
2. Flip a fair coin. You will win $2,000 if it lands on “𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻” and
nothing otherwise.
1 1
– 𝐸𝐸 𝑋𝑋 = $2,000 + $0 = $1,000
2 2
• If you’re indifferent between these two options, then the $𝑥𝑥 is
called the certainty equivalent of game #2
– The certainty equivalent is the rate that a risk-free investment would
need to offer to provide the same utility as the risky portfolio.
Risk Averse
• Attitudes towards risk
– Risk averse: 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 < 𝐸𝐸[𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔]
– Risk neutral: 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 = 𝐸𝐸[𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔]
– Risk loving: 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 > 𝐸𝐸[𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔]
• Are you willing to participate in a fair game?
– Risk averse: no
– Risk neutral: I don’t care one way or the other
– Risk loving: absolutely!
• Empirical studies analyzing historical returns on various asset classes suggest
that risky assets command a risk premium, implying that most investors are risk-
averse.
Risk Averse
A utility function, adopted by the CFA Institute, assigns a portfolio
with expected return 𝐸𝐸 𝑥𝑥 and variance of returns 𝜎𝜎 2 the following
utility
1 2
𝑈𝑈 = 𝐸𝐸 𝑥𝑥 − 𝐴𝐴𝜎𝜎
2
– where 𝑈𝑈 is the utility and 𝐴𝐴 is an index of the investor’s risk aversion
– 𝜎𝜎 2 is the measure of risk
• Trade-off between risk and return: utility increases with expected
return but falls with volatility
• Investors who are more risk-averse (having higher values of 𝐴𝐴)
require a higher expected return from risky investments.
Contents
1. Preference and Utility
2. Uncertainty and Risk Averse
3. Risk and Risk Premium
Risk
Risk comes from uncertainty
• Two broad sources of uncertainty:
1. Risk that comes from conditions in the general economy, such as
the business cycle, inflation, interest rates, and exchange rates.
– None of these macroeconomic factors can be predicted with certainty.
– Market risk, or systematic risk
2. Firm-specific influences, such as the firm’s success in research
and development and personnel changes.
– These factors affect the firm without noticeably affecting other firms in
the economy.
– firm-specific risk, or nonsystematic risk
Risk Premium
Suppose you won a lottery and are presented with two options:
1. Receive $1,000 immediately and walk away. Risk-free
2. Flip a fair coin. You will win $2,000
$20,000 if it lands on “𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻” and
nothing otherwise.
11 1 1
– 𝐸𝐸 𝑋𝑋 = $2,000
$20,000+ 2+ $0 $0
= $1,000
= $10,000
2
2 2
• Assuming you are risk averse, which option would you choose?
Interest rate risk Risk Premium
Inflation risk
• Risk-free rate is the rate you would earn in risk-free assets
such as T-bills, money market funds, or the bank.
– All the money market instruments are virtually free of interest rate
risk because of their short maturities and usually pose little default or
credit risk.
– Even that bond would offer a guaranteed real return only if its
maturity matched the investor’s desired holding period.
• Even inflation-indexed bonds are subject to interest rate risk because real
interest rates change unpredictably through time.
Risk Premium
• Risk-free rate is the rate you would earn in risk-free assets such as
T-bills, money market funds, or the bank.
𝑟𝑟𝑓𝑓
• A risk premium is a measure of excess return that is required by an
individual to compensate being subjected to an increased level of
risk.
– A risk premium is the investment return an asset is expected to yield in
excess of the risk-free rate of return.
𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑚𝑚𝐴𝐴 = 𝐸𝐸 𝑟𝑟𝐴𝐴 − 𝑟𝑟𝑓𝑓
Measuring Risk
• We can directly observe neither expected return nor risk.
– We observe only realized rates of return. These provide noisy estimates of
the expected returns and risk that investors actually anticipated.
• While we have theories about the proper relationship between risk
and expected return, there is no theory about the levels of risk we
should find in the marketplace.
– We can at best estimate from historical experience the level of risk that
investors are likely to confront.
– No matter how extensive the historical record, learning from it can never
guarantee that it captures the worst (or best) that nature might present in
the future.
Measuring Risk
We assume normality of returns and use the standard deviation
of the excess return as the measure of portfolio risk.
• The standard deviation of the rate of return does not
distinguish between good or bad surprises; it treats both
simply as deviations from the mean.
• As long as the probability distribution is more or less
symmetric about the mean, 𝜎𝜎 is a reasonable measure of risk.
• Assume that the probability distribution is normal: 𝐸𝐸(𝑟𝑟) and 𝜎𝜎
completely characterize the distribution.
Measuring Risk
We measure the attraction of a portfolio by the ratio of its risk premium to
the SD of its excess returns.
• This reward-to-volatility measure was first used extensively by William
Sharpe and hence is commonly known as the Sharpe ratio.
• It is widely used to evaluate the performance of investment managers.
𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝
𝑆𝑆ℎ𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 =
𝑆𝑆𝑆𝑆 𝑜𝑜𝑜𝑜 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟
– Normality assures us that the standard deviation is a complete measure of risk
and, hence, the Sharpe ratio is a complete measure of portfolio performance.
Appendix
Return and Normality
Measuring Downside Exposure
Learning from Historical Returns
Not required for the exam
Return and Normality
We assume normality of returns and use the standard deviation of
the excess return as the measure of portfolio risk.
• It is not unreasonable to accept the simplification offered by
normality as an acceptable approximation
• However, not all returns are (close to) normally distributed!
– Unless returns are normally distributed, standard deviation is not
sufficient to measure risk. We also need to think about “tail risk,” that is,
our exposure to unlikely but very large outcomes in the left tail of the
probability distributions.
• Use caution when using the Sharpe ratio in real life.
Return and Normality
Historical returns on stocks exhibit somewhat more frequent
large deviations from the mean than would be predicted from a
normal distribution.
• However, the discrepancies from the normal distribution tend
to be minor and inconsistent across various measures of tail
risk, and have declined in recent years.
• The lower partial standard deviation (LPSD), skew, and kurtosis
of the actual distribution quantify the deviation from normality
Return and Normality
• Figure 5.3 is the frequency
distribution of monthly
returns for the broad
market index over the last
95 years, 1927–2021.
• Approximately normally
distributed.
It shows a frequency distribution of
monthly excess returns on the market
index since 1927.
• The solid bars show the historical
frequency of excess returns falling
within each range, while the
pattern bars show the frequencies
that we would observe if these
returns followed a normal
distribution with the same mean
and variance as the actual
empirical distribution.
• You can see here some evidence of
a fat-tailed distribution: The actual
frequencies of extreme returns,
both high and low, are higher than
would be predicted by the normal
distribution.
Return and Normality
• Summary: Historical returns on stocks exhibit somewhat more
frequent large deviations from the mean than would be
predicted from a normal distribution. However, the
discrepancies from the normal distribution tend to be minor
and inconsistent across various measures of tail risk, and have
declined in recent years. The lower partial standard deviation
(LPSD), skew, and kurtosis of the actual distribution quantify
the deviation from normality.
Measuring Downside Exposure
The use of standard deviation as a measure of risk when the
return distribution is non-normal presents two problems:
1. The asymmetry of the distribution suggests we should look at
negative outcomes separately and
2. Because an alternative to a risky portfolio is a risk-free
investment, we should look at deviations of returns from the
risk-free rate rather than from the sample average, that is, at
negative excess returns.
Measuring Downside Exposure
Widely used measures of tail risk are
• Value at risk (VaR)
– VaR measures the loss that will be exceeded with a specified probability such as 1% or
5%.
– VaR tells you the investment loss at the first percentile of the return distribution, but it
ignores the magnitudes of potential losses even further out in the tail.
• Expected shortfall (ES) or conditional tail expectation (CTE)
– measures the expected rate of return conditional on the portfolio falling below a certain
value.
– 1% ES is the expected value of the outcomes that lie in the bottom 1% of the
distribution.
– Using a sample of historical returns, we would estimate the 1% expected shortfall by
identifying the worst 1% of all observations and taking their average.
– [Link]
Measuring Downside Exposure
• The lower partial standard deviation (LPSD) of excess returns is computed
like the usual standard deviation, but using only “bad” returns.
– it uses only negative deviations from the risk-free rate (rather than negative
deviations from the sample average), squares those deviations to obtain an
analog to variance, and then takes the square root to obtain a “left-tail standard
deviation.”
• Notice that this measure ignores the frequency of negative excess returns;
that is, portfolios with the same average squared negative excess returns
will yield the same LPSD regardless of the relative frequency of negative
excess returns.
• Practitioners who replace standard deviation with this LPSD typically also
replace the Sharpe ratio with the ratio of average excess returns to LPSD.
This variant on the Sharpe ratio is called the Sortino ratio.
Learning from Historical Returns
• Observation frequency has no impact on the accuracy of estimates of expected
return. It is the duration of a sample time series (as opposed to the number of
observations) that improves accuracy
– Because the average monthly return must be consistent with the average annual
returns, the additional intra-year observations provide no additional information about
mean return.
• A longer sample, for example, a 100-year return, does provide a more accurate
estimate of the mean return than a 10-year return, provided the probability
distribution of returns remains unchanged over the 100 years.
– Are return data from the 19th century relevant to estimating expected returns in the
21st century?
• In contrast to the mean, the accuracy of estimates of the standard deviation can
be made more precise by increasing the number of observations.
