Professional Documents
Culture Documents
Sebenta - Finance - Vasco Tamen
Sebenta - Finance - Vasco Tamen
FINANCE
2020/2021
2. Financial Ratios
Financial ratios are a convenient way to analyze financial statements. Ratios are not very helpful
by themselves: they need to be compared to something.
• Time-Trend Analysis: Used to see how the firm’s performance is changing through time
• Peer Group Analysis: Compare to similar companies or within industries
➢ Financial Leverage
Leverage Ratios measure how much a firm relies on financial debt (i.e., interest-bearing) rather
than equity.
➢ Profitability Ratios
Profitability Ratios measure the ability to sell the product for more than cost and return on
investment
Example:
A. Suppose one deposit 100€ at a 10% interest rate per year for the next 5 years. How much
will one have in five years?
𝐹𝑉 = 100€ × (1 + 10%)5 = 161,051€
B. How much an investor have to set aside today to have 5.000€ in 5 years at 10% per year?
𝐹𝑉 5000€
𝑃𝑉 = 𝑇
= = 3104,61€
(1 + 𝑟) (1 + 10%)5
C. If we deposit 5.000€ today in an account paying 10%, how long does it take to grow to
10,000€?
𝐹𝑉 = 𝑃𝑉 × (1 + 𝑟)𝑇 ⇔ 10,000€ = 5,000€ × (1 + 10%)𝑇 ⇔ T = log1,1 2 = 7,27 years
D. Assume the total cost of a college education will be 50.000€ in 12 years. You have 5.000€
to invest today. What rate of interest must you earn to cover the cost?
12
𝐹𝑉 = 𝑃𝑉 × (1 + 𝑟)𝑇 ⇔ 50,000€ = 5,000€ × (1 + 𝑟)12 ⇔ 𝑟 = √10 − 1 = 21.15%
Example:
If our initial balance is 100€, what is the end-of-year balance (no payments during the year)?
The APR for our credit card is 12% compounded:
• Monthly
𝐴𝑃𝑅 𝑚×𝑇 12% 12×1
𝐹𝑉 = 𝑃𝑉 × (1 + ) = 100€ × (1 + ) = 112,68€
𝑚 12
• Quarterly
𝐴𝑃𝑅 𝑚×𝑇 12% 4×1
𝐹𝑉 = 𝑃𝑉 × (1 + ) = 100€ × (1 + ) = 112,55€
𝑚 4
C. Now, consider another lender that offers loans that require monthly payments. This lender
would have to advertise an APR compounded monthly. What is the implied APR
compounded monthly (𝐴𝑃𝑅month )?
𝐴𝑃𝑅month = 12 × 𝐸𝐼𝑅month = 12 × 0.797% = 9.564%
Imagine that a lender offers a loan in which you need to make semi-annual payments. The
lender would have to advertise an APR compounded semi-annually. What would that be?
𝐴𝑃𝑅semester = 2 × 𝐸𝐼𝑅semester = 2 × 4.881% = 9.762%
3. Special Cases
Let´s recall the concept of Geometric Series which is the infinite sum of a geometric sequence.
𝑁 𝑁
𝑛
1 − 𝑓𝑁
𝑁
∑ 𝑎𝑛 = ∑ 𝑎 ∙ 𝑓 = 𝑎 + 𝑎 ∙ 𝑓 + ⋯ + 𝑎 ∙ 𝑓 = 𝑎 ×
1−𝑓
𝑛=0 𝑛=0
1−𝑓𝑁 1−0 𝒂
The sum of all terms (infinite geometric series) if |𝑓| < 1: lim (𝑎 × 1−𝑓
) = 𝑎 × 1−𝑓 = 𝟏−𝒇
𝑁∈ℕ
3.1. Perpetuity
𝐶 𝐶
𝐶 𝐶 (1 + 𝑟) (1 + 𝑟) 𝐶
𝑃𝑉 = + +⋯= = 𝑟 =
(1 + 𝑟) (1 + 𝑟)2 1 𝑟
1− (1 + 𝑟)
(1 + 𝑟)
𝐶 𝐶
𝐶 𝐶(1 + 𝑔) 𝐶(1 + 𝑔)2 (1 + 𝑟) (1 + 𝑟) 𝐶
𝑃𝑉 = + + +⋯= = 𝑟−𝑔 =
(1 + 𝑟) (1 + 𝑟) 2 (1 + 𝑟)3 1 + 𝑔 𝑟−𝑔
1− (1 + 𝑟)
(1 + 𝑟)
Example:
What is the present value of a perpetuity of 15€ if it grows forever at 5% and the interest rate is
10% per year? The first payment is made at the end of the first year.
15
𝑃𝑉 = = 150€
10% − 5%
3.3. Annuity
Stream of constant cash flows that lasts for
a fixed number of periods (maturity).
Annuity Factor
Present value of 1€ a year for 𝑇 years at interest rate of 𝑟.
Notes:
• The annuity formula provides the value of the annuity one period before the first cash
flow.
• You can use the function “PV” in excel.
𝑃𝑉𝑡=1 = 𝐶 × 𝐴10
6%/2
1 1
𝑃𝑉𝑡=0 = 𝑃𝑉𝑡=1 × = 𝐶 × 𝐴10
3% × ⇔ 𝐶 = 30.187€
1 + 6%/2 1 + 3%
𝑃𝑉𝑡=−1 = 𝐶 × 𝐴10
6%/2
What if the first installment paid at end of year 1 but at the end of semester 1 we pay interest?
250.000€ × 3% 1
250.000€ = + C × 𝐴10
3% × ⟺ 𝐶 = 29.308€
1 + 3% 1 + 3%
1+𝑔 𝑇
𝐶 𝐶(1 + 𝑔) 𝐶(1 + 𝑔) 2
𝐶(1 + 𝑔) 𝑇−1
𝐶 1 − (1 + 𝑟 )
𝑃𝑉 = + + + ⋯+ = ×
(1 + 𝑟) (1 + 𝑟)2 (1 + 𝑟)3 (1 + 𝑟)𝑇 (1 + 𝑟) 1+𝑔
1− 1+𝑟
𝐶 1+𝑔 𝑇
= × [1 − ( ) ]
𝑟−𝑔 1+𝑟
Example:
A defined-benefit retirement plan offers to pay 20.000€ per year for 20 years and increase the
annual payment by 3% each year. What is the present value at retirement if the interest rate is
10% per year?
𝐶 1+𝑔 𝑇 20.000€ 1 + 3% 20
𝑃𝑉 = × [1 − ( ) ]= [1 − ( ) ] = 209.010€
𝑟−𝑔 1+𝑟 10% − 3% 1 + 10%
The Bond´s discount rate is the Yield to Maturity (𝑦) which represents the yield of alternative
investments with comparable risk in the market.
• Sets the 𝑃𝑉 of Future Cash Flows equal to the current market price
• Required interest rate on the bond if all payments are met
• It is usually presented as an effective annual rate (EAR)
There are three different types of bonds that we are going to study:
1.1. Types of Bonds
1.1.1. Coupon Bonds
Bonds that pay periodic Coupons
Payments (C) until maturity and the
Face Value (F) at maturity.
Coupon Payments (C) = 𝐹 × 𝐶%
Therefore, these types of bonds are just a combination of an annuity of C and the 𝑃𝑉 of F.
𝑪 𝟏 𝑭
Bond Value = [𝟏 − (𝟏+𝒚)𝑻 ] + (𝟏+𝒚)𝑻
𝒚
Note: 𝐶, 𝑇, and 𝑦 need to be consistent with the frequency of payments. For semi-annual
payments, y as effective semi-annual rate, T in semesters, and C as the payment per semester
6.375 1 100
Bond Value = 5%
[1 − (1+5%)10 ] + (1+5%)10 = 110,617 €
Example:
Find the value of a 30-year zero-coupon bond with a 100 € par value and a YTM = 𝑦 of 6%.
100
Bond Value =
(1 + 6%)30
1.1.3. Consols
C Bonds that pay a fixed infinite
stream of coupons.
∞
Therefore, these types of bonds do not have final maturity, they are just a Perpetuity.
𝒄
𝐁𝐨𝐧𝐝 𝐕𝐚𝐥𝐮𝐞 =
𝒚
𝐶% < 𝑦 ⟹ Bond Price < Face Value ⟹ Traded at Discount, Losing Money
𝐶% > 𝑦 ⟹ Bond Price > Face Value ⟹ Traded at Premium, Raising Money
Example:
Consider a bond with a 6.375% coupon rate and face value is 100, coupons are paid annually.
𝐶% = 𝑦 ⇒ Par Value
Clean Price ⟹ Quoted Price Dirty Price ⟹ Discounted Cash Flows, actually paid
Now, let´s do the exercises above the other way around. Having the value of the Bond and all
future cash flows, try to compute the yield:
𝐶 1 𝐹
Bond Value = [1 − (1+𝑦)𝑇 ] + (1+𝑦)𝑇
𝑦
Since, in this case, solving the above equation for y is hard, we should use Excel - Yield Function,
What-If Analysis, or Goal Seek. [Class 25/02] [Excel File]
1
Annual yields in Europe are often reported as EAR´s but in the US, they report them as APRs. The Excel
yield function gives APR.
Example:
Let us assume that we have 4 zero-coupon bonds, each with a par value of 1000 €, as follows:
• 1-year bond with a market price of 925,93 €
1000
925,93 = ⟺ 𝑟1 = 8%
1 + 𝑟1
• 2-year bond with a market price of 841,75 €
1000
841,75 = ⟺ 𝑟2 = 9%
(1 + 𝑟2 )2
• 3-year bond with a market price of 758,33 €
1000
925,93 = ⟺ 𝑟3 = 9,66%
(1 + 𝑟3 )3
• 4-year bond with a market price of 683,18 €
1000
925,93 = ⟺ 𝑟4 = 9,99%
(1 + 𝑟4 )4
What are the implications of the yield curve for bond pricing?
If the term structure is not flat, then to compute the price of the coupon bond you need to use:
𝐶 𝐶 𝐶 𝐶+𝐹
𝑃= + 2
+ ⋯+ 𝑇−1
+
(1 + 𝑟1 ) (1 + 𝑟2 ) (1 + 𝑟𝑇−1 ) (1 + 𝑟𝑇 )𝑇
Bond value is 𝑃𝑉 of promised bond payments but interest rate (i.e., discount rate) needs to be
adjusted by default risk:
𝑦 = 𝑟 ∗ + 𝐼𝑃 + 𝑀𝑅𝑃 + 𝑫𝑹𝑷
Where,
- 𝐷𝑅𝑃 = Default Risk Premium
- 𝑟 ∗ = Real risk-free rate (Usually Gov. Bonds)
- 𝐼𝑃 = Expected Inflation Premium
- 𝑀𝑅𝑃 = Maturity Risk Premium
There is a different term structure of interest rates for different levels of default risk/rating.
The higher probability of default, the higher will be the yields. Because of this risk, bonds with
default risk trade in the market at a lower price/higher yield than the comparable government
bond.
Typically, they are similar, but sometimes the expectation of the future dividends may be different
between investments.
We can think of a stock value today (𝑃0 ) as the present value of future dividends and capital gains.
Therefore, we have three different types of stocks´ valuation:
𝐷1 1 + 𝑔1 𝑇 𝐷𝑇+1 1
Stock Price 𝑃0 = × [1 − ( ) ]+ ×
𝑅 − 𝑔1 1+𝑅 𝑅 − 𝑔2 (1 + 𝑅)𝑇
Example:
The dividend is expected to grow at 8% for two years (after year 1). Afterward, dividends are
expected to grow at 4% in perpetuity. The firm will pay a dividend of 2 € a year from today. What
is the price of the stock today if the discount rate is 12%?
Cash flows considering that after
year 4 dividends grow at a constant
rate (4%).
𝑅𝑒𝑡𝑒𝑛𝑡𝑖𝑜𝑛 𝑅𝑎𝑡𝑖𝑜 = share of profits are retained by the company and available to invest
𝐷𝑖𝑣𝑖𝑑𝑒𝑛𝑑𝑠 𝐷𝑖𝑣𝑖𝑑𝑒𝑛𝑑𝑠 𝑝𝑒𝑟 𝑠ℎ𝑎𝑟𝑒
𝑃𝑎𝑦𝑜𝑢𝑡 = Fraction of earnings that is paid out to stockholders = 𝑁𝑒𝑡 𝐼𝑛𝑐𝑜𝑚𝑒 = 𝐸𝑎𝑟𝑛𝑖𝑛𝑔𝑠 𝑝𝑒𝑟 𝑠ℎ𝑎𝑟𝑒
The value of a firm is then the sum of the value of a firm that pays out 100% of its earnings
(𝐸𝑃𝑆) as dividends and the net present value of the growth opportunities (𝑁𝑃𝑉𝐺𝑂).
𝑬𝑷𝑺𝟏
𝑷𝟎 = + 𝑵𝑷𝑽𝑮𝑶
𝑹
The key part is to assess the growth opportunities. This is not easy since it involves a lot of
uncertainty but under some assumptions, we can estimate the value. Growth opportunities are
opportunities to invest in positive NPV projects.
Bearing in mind that 𝑁𝑃𝑉1 will grow every year at a constant rate of 𝑔 = 14% we can then
consider NPV as a growing annuity:
𝑁𝑃𝑉1 0,875
𝑁𝑃𝑉𝐺𝑂0 = = = 43,75
𝑅 − 𝑔 16% − 14%
And so, finally, we get:
𝑃0 = 31,25 + 43,75 = 75
Conclusions:
If the 𝑅𝑂𝐸 < 𝑅 ⟹ NPVGO < 0
The firm’s investment opportunities are worse than the ones available to investors. In this case,
firms are better off paying out dividends to investors instead of retaining earnings to invest
internally.
The NPVGO is very useful in situations in which it is not a linear function of the 𝑅𝑂𝐸, that is,
the firm has new growth projects that are different from its current business. In this case, the DGM
will not work.
Price-Earnings Ratio
Analysts frequently relate earnings per share to price. The Price-earnings ratio is calculated as the
current stock price divided by next year (annual) EPS (i.e., number of years to recover investment)
𝑷 𝑷𝟎
=
𝑬 𝑬𝑷𝑺𝟏
Using Dividend Growth Model:
𝐷1 𝐸𝑃𝑆1 × 𝑃𝑎𝑦𝑜𝑢𝑡 𝑷 𝑷𝟎 𝑷𝒂𝒚𝒐𝒖𝒕
𝑃0 = = ⟹ = =
𝑅−𝑔 𝑅−𝑔 𝑬 𝑬𝑷𝑺𝟏 𝑹−𝒈
𝑃
If firms are comparable in terms of payout, growth, and risk we may compare 𝐸 across firms to
access:
𝑃
Low 𝐸 ⟹ Stock is cheap 𝑃
↑ 𝑃𝑎𝑦𝑜𝑢𝑡 𝑜𝑟 𝑔 ⟹ ↑
𝐸
𝑃
High 𝐸 ⟹ Stock is expensive 𝑃
↑𝑅 ⟹↓
𝐸
Relevant cost (we should consider) Not Relevant (we shouldn´t consider)
• Sunk Costs
They are irreversible. A decision should not evaluate what was spent but what will be spent.
• Opportunity Costs
If an existing asset is used in a new project, potential revenues from alternative uses of the asset
are lost. The lost revenues should be considered as a cost (opportunity cost) of the new project.
• Side Effects (Positive or Negative Externalities)
For example, Erosion and cannibalism. If our new product causes existing customers to demand
less (or more) of current products, we need to recognize that.
Terminal Value
It is the residual value of a fixed asset at the end of a project. We may compute it through one of
the following ways:
a) Salvage Value after taxes
At end of the project is to be liquidated or sold by the market value.
𝑆𝑎𝑙𝑣𝑎𝑔𝑒 𝑣𝑎𝑙𝑢𝑒 𝑎𝑓𝑡𝑒𝑟 𝑡𝑎𝑥 = 𝑆𝑎𝑙𝑣𝑎𝑔𝑒 𝑉𝑎𝑙𝑢𝑒 − 𝑇𝑎𝑥 = 𝑆𝑎𝑙𝑣𝑎𝑔𝑒 𝑉𝑎𝑙𝑢𝑒 − 𝐶𝑎𝑝𝑖𝑡𝑎𝑙 𝐺𝑎𝑖𝑛 × 𝑡
= 𝑆𝑎𝑙𝑣𝑎𝑔𝑒 𝑉𝑎𝑙𝑢𝑒 − (𝑆𝑎𝑙𝑣𝑎𝑔𝑒 𝑉𝑎𝑙𝑢𝑒 − 𝐵𝑜𝑜𝑘 𝑉𝑎𝑙𝑢𝑒) × 𝑡
Where,
- 𝑡 = marginal corporate tax rate
- 𝐵𝑜𝑜𝑘 𝑉𝑎𝑙𝑢𝑒 = Purchase Value − Acc. Depr.= Net of Depreciation
𝐶𝑎𝑝𝑖𝑡𝑎𝑙 𝐺𝑎𝑖𝑛 < 0 ⟹ Loss in the sale of Fixed Asset ⟹ No tax, but it may be Tax-Deductible
⇓
Tax Saving
b) Continuation Value
The firm will last forever (end of life Year 𝑇), and therefore we will use the Growing Perpetuity.
𝐹𝐶𝐹𝑇+1 𝐹𝐶𝐹𝑇 × (1 + 𝑔)
𝐶𝑉 = =
𝑅−𝑔 𝑅−𝑔
So, let´s organize the information and compute the Investment Cash Flows (ICF):
How to make the investment decision? Use one of the following rules!
Assumption: 𝑁𝑃𝑉 rule assumes that all cash flows can be reinvested at the discount rate
Perks:
• Accepting positive NPV projects benefits shareholders while maximizing shareholder
value (stock price)
• NPV uses cash flows
• NPV discounts the cash flows properly (need to find appropriate discount rate based on
the risk of cash flows)
Problems:
• Ignores the time value of money
• Ignores cash flows after the payback period (biased against long-term projects)
• Requires arbitrary acceptance criteria
To face these problems, we could evaluate the Discounted Payback Period, that is, considering
the time value of money.
Decision Rule: Accept the project if it pays back on a discounted basis within the specified time.
Assumption: 𝐼𝑅𝑅 rule assumes that all cash flows can be reinvested at the 𝐼𝑅𝑅
To find the 𝐼𝑅𝑅, in most cases, we need to solve a non-linear equation which is not easy!
Therefore, we will use Excel tools as IRR, solver, goal seek functions… Alternatively, we use a
graphical approach by plotting NPV versus the discount rate where the 𝐼𝑅𝑅 will be the 𝑥-axis
intercept.
Problems:
• Problem 1: Investing or financing?
The difference between them is when considering an investing project, the initial investment is
negative and future cash flows are positive, whereas in a financing project today we receive a
positive cash flow, and then future cash flows will be negative.
Let´s consider the following one-year examples:
130 130
100 − =0 −100 + =0
1 + 𝐼𝑅𝑅 1 + 𝐼𝑅𝑅
Example:
Consider two mutually exclusive projects (𝑟 = 8%): Project A (−1800, 3200); Project B
(−1200, 2300). Build the incremental cash flows and identify which project we should invest in.
900
Incremental Cash Flow: Project A – B: (−600,900) ⟹ −600 + 1+𝐼𝑅𝑅 ⇔ 𝐼𝑅𝑅 = 50% > 8%
So, we should invest in Project A since it is valuable to invest 600€ today with a discount of 8%
because the 𝑃𝑉 𝑜𝑓 𝐹𝑢𝑡𝑢𝑟𝑒 𝐶𝐹 > 𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝐼𝑛𝑣𝑒𝑠𝑡𝑚𝑒𝑛𝑡!
𝐼𝑅𝑅 and 𝑁𝑃𝑉 give different answers: IRR favors small scale project, which has lower NPV; but
we should pick large scale project!
Solution: Look at incremental cash flows!
In this example create a project where Large − Small: (−1000, 1500).
Since, 𝐼𝑅𝑅 = 50% > 10%, we should accept the Large Project!
Timing Problem
Let´s consider two mutually exclusive projects (𝑟 = 10%):
Note: Bear in mind, you should create an investment project and not a financing project!
50 100 150
+ +
(1 + 10%) (1 + 10%)2 (1 + 10%)3 240,8
𝑃𝐼 = = = 1,204 ⟹ Accept!
200 200
Example:
Consider the following mutually exclusive projects and assume 𝑟 = 10% and 𝑡 = 0.
• Machine A costs 4.000€, has annual operating costs of 100€ and lasts 10 years.
• Machine B costs 1.000€, has annual operating costs of 500€ and lasts 5 years.
Since we do not have revenues and assuming they are the same for the two machines we could
compute the present value of costs:
𝑃𝑉𝐴 = 4000 + 100 × 𝐴10
10% = 4614,5€ Using the 𝑁𝑃𝑉 rule the machine B will have
5 lower costs and so it would be the best.
𝑃𝑉𝐵 = 1000 + 500 × 𝐴10% = 2895,4€
But in this case, we are ignoring the fact that machine A has twice the lifetime expectance than
B. Therefore, we should use one of the following methods to evaluate which one is the best:
Now, using the 𝑁𝑃𝑉 rule the machine A will have lower costs and so it would be the best pick.
3. Risk Analysis
Allows us to look behind the 𝑁𝑃𝑉 number to see how stable our estimates are. We will only study
the following three types of risk analysis:
I. Sensitivity Analysis
Calculate 𝑁𝑃𝑉 (or 𝐼𝑅𝑅) with one input varied while keeping all other inputs constant.
1. Returns
Dollar Returns = 𝑃𝑡+1 + 𝐷𝑡+1 − 𝑃𝑡
Where,
- 𝑃𝑡 : Market Price at the beginning period
- 𝑃𝑡+1 : Market Price at the end period
- 𝐷𝑡+1 : Dividends (cash flow) paid during the period
Example:
Suppose you bought 100 shares of the firm one year ago at 25€. You received 20€ in dividends
(0,20 €/share × 100 shares). At the end of the year, stock sells for 30€. What was the total
percentage return?
Investment: 25€ × 100 = 2500€
At the end of the year, the stocks are worth 3000€ and dividends of 20€.
𝑃𝑡+1 +𝐷𝑡+1 −𝑃𝑡 𝑃𝑡+1 −𝑃𝑡 𝐷𝑡+1 3000−2500 20
Percentage Return: 𝑟𝑡+1 = = + = + = 20,08%
𝑃𝑡 𝑃𝑡 𝑃𝑡 2500 2500
Example:
Suppose your investment provides the following returns over a four-year period: Year 1 = 10%;
Year 2 = −5%; Year 3 = 20%; Year 4 = 15%.
Which is better?
Geometric average is an excellent measure Arithmetic average is the best estimate of
of past realized performance and a good the expected return in a single period in the
estimate of annual return to be obtained over future.
extended periods in the future.
∑𝑇 (𝑟𝑖 − 𝑟̅ )2
𝜎 = √ 𝑖=1
𝑇−1
2
Given by the return on Treasury Bills.
1. Individual Securities
To make this type of analysis we will need the following concepts.
Consider that there are 𝑆 states of the world (Recession, normal, and Boom). If you know the
probability of each state (𝑃𝑠 ) and the return in each state (𝑟𝑠 ), then you can compute all of the
following:
𝑆
Expected Return
𝐸(𝑟) = 𝜇 = ∑ 𝑃𝑠 ∙ 𝑟𝑠
(Average)
𝑠=1
𝑆 𝑆
Variance (Var) 𝑉𝑎𝑟(𝑟) = 𝜎 = ∑ 𝑃𝑠 ∙ (𝑟𝑠 − 𝜇) = ∑ 𝑃𝑠 ∙ (𝑟𝑠 )2 − 𝜇2
2 2
𝑠=1 𝑠=1
Standard Deviation
(Stdev) 𝜎 = +√𝜎 2
Covariance (Covar)
𝑆 𝑆
Measure the direction of 𝐶𝑜𝑣𝐴,𝐵 = 𝜎𝐴,𝐵 = ∑ 𝑃𝑠 ∙ (𝐴𝑠 − 𝐴̅)(𝐵𝑠 − 𝐵̅) = ∑ 𝑃𝑠 ∙ 𝐴𝑠 𝐵𝑠 − 𝐴̅𝐵̅
a linear relationship
𝑠=1 𝑠=1
between two variables.
Correlation Coefficient
𝐶𝑜𝑣𝐴,𝐵
(Correl) 𝜌𝐴,𝐵 = , 𝑟𝑥,𝑦 𝜖[−1,1]
𝜎𝐴 ∙𝜎𝐵
Measures co-movement 2
between two random Closer to 0 ⟹ Weak Relationship; |𝜌𝐴,𝐵 | ≥ ⟹ Significant
√𝑛
variables.
Example:
Consider the following two risky assets.
There is a 1/3 chance of each state of the
economy, and the only assets are stock 𝐴
and stock 𝐵.
1
𝐸(𝑟𝐴 ) = 𝜇 = × [−7% + 12% + 28%] = 11%
3
1
𝜎𝐴 2 = 3 × [(−7%)2 + (12%)2 + (28%)2 ] − (11%)2 = 0,0205 ⟹ 𝜎 = √0,0205 = 14,3%
1
𝐶𝑜𝑣𝐴,𝐵 = × [(−7%)(5%) + (12%)(25%) + (28%)(−5%)] − 11% × 8,33% = −0,005
3
−0,005
𝜌𝐴,𝐵 = = −0,28
14,3% ∙ 12,5%
Estimations
In real life, we do not know the probability of each state of the world and the corresponding return.
We need to use historical data to estimate average returns, variance, and covariance of returns.
∑𝑇𝑡=1 𝑟𝑡 2
∑𝑇𝑡=1(𝑟𝑡 − 𝑟̅ )2 ∑𝑆𝑠=1(𝐴𝑠 − 𝐴̅)(𝐵𝑠 − 𝐵̅)
𝑟̅ = 𝜎̂ = 𝜎̂
𝐴,𝐵 =
𝑇 𝑇−1 𝑇−1
We are implicitly assuming that the returns came from the same probability distribution in each
year of the sample.
The estimated mean and variance are themselves random variables since there is an estimation
error that depends on the particular sample of data used (sampling error).
We can calculate the standard error of our estimates and figure out a confidence interval for them.
This contrasts with the true (but unknown) mean and variance which are fixed numbers, not
random variables.
Annualizing Parameters
Annual return is approximately equal to the sum of the 12 monthly returns, assuming monthly
returns are independently distributed (a consequence of market efficiency) and have the same
variance. So, to annualize:
• Mean, Variance; Covariance: × 12
• Standard Deviation: × √12
2. Portfolios
Combination of many securities.
It is described by the weights (𝑤) of each security: fraction of wealth invested in different assets.
𝑤1 + 𝑤2 + ⋯ + 𝑤𝑘 = 1
Example:
Consider the following portfolio: $500 MSFT (buy), $200 in GE (short sell).
Total Investment = 500 − 200 = 300
500 5 −200 2
Portfolio weights: MSFT ⟹ 300 = 3 ; GE ⟹ 300
= −3
Observe the decrease in risk that diversification offers. A portfolio with 60% in stock 𝐴 and
40% in stock 𝐵 has less risk than either stock in isolation.
3. Mean-Variance Analysis
3.1. Two Stocks
The mean-variance space is a representation of all different
portfolios in terms of:
• 𝑦-axis: Expected return, 𝐸(𝑅)
• 𝑥-axis: Standard deviation, 𝜎(𝑅)
𝜎(𝑟𝑃 ) < 𝑤𝐴 ∙ 𝜎𝐴 + (1 − 𝑤𝐴 ) ∙ 𝜎𝐵
𝜎(𝑟𝑃 ) = ∓ 𝑤𝐴 ∙ 𝜎𝐴 ± (1 − 𝑤𝐴 ) ∙ 𝜎𝐵
Efficient Portfolio
Efficient Portfolios are the points of the frontier above the global minimum variance portfolio
(green line). Therefore, the best portfolios are the ones that maximize 𝐸(𝑅) [or minimum 𝜎(𝑅)]
given a level 𝜎(𝑅) [or 𝐸(𝑅)].
𝑬(𝒓𝑷 ) = ∑ 𝑤𝑖 ∙ 𝐸(𝑟𝑖 )
𝒊=𝟏
𝑵 𝑵 𝑵 𝑵 𝑵
𝟐 ]2
𝝈 (𝒓𝑷 ) = ∑ ∑ 𝑤𝑖 ∙ 𝑤𝑘 ∙ 𝜎𝑖,𝑘 = ∑[𝑤𝑖 𝜎𝑖 + 𝟐∑ ∑ 𝑤𝑖 ∙ 𝑤𝑘 ∙ 𝜎𝑖,𝑘
𝒊=𝟏 𝒌=𝟏 𝒊=𝟏 𝒊=𝟏 𝒌=𝟏 ; 𝒊<𝒌
3.3. Conclusions
• The Efficient Frontier is the set of mean-variance combinations from the minimum-variance
frontier where for a given risk no other portfolio offers a higher expected return.
• Investors should only pick portfolios from the Efficient Mean-Variance Frontier.
• Investors will never want to hold a portfolio below the minimum variance point. They will
always get higher returns along the positively sloped part of the minimum-variance frontier.
5. Risk-free Assets
Let´s consider expanding the Markowitz approach by considering investing not just in risky
assets but also in a risk-free asset.
The risk-free asset has a certain payoff. There is no uncertainty about the terminal value of this
type of asset.
- Risk of the risk-free asset is zero ⟹ 𝜎𝑓 = 0
- Covariance between the risk-free asset and any risky asset is zero ⟹ 𝜎𝑀,𝑓 = 0
Let assume a portfolio, 𝑃, which is well-diversified and combines Risky Assets, 𝑀, and Risk-free
Asset, 𝐹. Then the investor determines the weights of each type of assets in the portfolio:
• 𝑤𝑀 is the weight of Risky Assets
• 𝑤𝑓 = 1 − 𝑤𝑀 is the weight of a Risk-free asset
[𝑬(𝒓𝑴 ) − 𝒓𝒇 ]
𝑬(𝒓𝑷 ) = ∙ 𝝈(𝒓𝑷 ) + 𝒓𝒇
𝝈𝑴
This relationship between 𝐸(𝑟𝑃 ) and 𝜎(𝑟𝑃 ) is
called Capital Allocation Line!
Bear in mind that we are assuming 𝑟𝑀 as the best
portfolio risky asset.
𝐸(𝑟𝑀 )−𝑟𝑓
Slope: Sharpe ratio 𝜎𝑀
All investor has the same optimal risky portfolio. But the investor’s risk aversion will determine
how much of their wealth is invested in the optimal risky portfolio and how much is invested in
the risk-free asset (i.e., determine the actual position on the CAL).
To decide which of the efficient portfolio is the best we should use quadratic utility function since
the investor only cares about the mean and variance (risk) of returns:
𝝈𝟐𝑷
𝑼(𝒓𝑷 ) = 𝑬(𝒓𝑷 ) − 𝜸 ∙
𝟐
Where 𝛾 stands for Risk Aversion Coefficient. Usually determined using surveys and on average,
𝛾 ∈ [2,4]. A Higher 𝛾 ⟹ High-risk aversion, that is, the investor does not like risk.
To find optimal portfolio choice for a combination of a risky asset and a risk-free asset, we
have to maximize the utility function above subject to the CML we get that:
𝐸(𝑟𝑀 ) − 𝑟𝑓
𝑤𝑀 =
𝛾 × 𝜎𝑀 2
Examples:
A. You are advising two investors (Investor A and B) about portfolio allocation. For simplicity,
suppose the only three asset categories you focus on are the risk-free asset, the US stock
market, and venture capital (not part of the US stock market). You have estimated the
following based on annual data:
The riskless rate is 5% and the correlation
between US stocks and venture capital is
0,5.
Based on this the Tangency Portfolio is to invest 73,3% of risky asset holdings in US stocks
and 26,7% in venture capital.
1) What are the expected return and standard deviation of the Tangency Portfolio?
The Tangency Portfolio and Market portfolio are the same!
𝐸(𝑟𝑃 ) = 𝑤𝑈𝑆 × 𝐸(𝑟𝑈𝑆 ) + 𝑤𝑉𝐶 × 𝐸(𝑟𝑉𝐶 ) = 0,733 × 0,10 + 0,267 × 0,14 = 0,1107
𝜎 2 (𝑟𝑃 ) = [𝑤𝑈𝑆 ∙ 𝜎𝑈𝑆 ]2 + [𝑤𝑉𝐶 ∙ 𝜎𝑉𝐶 ]2 + 2𝑤𝑈𝑆 ∙ 𝑤𝑉𝐶 ∙ 𝐶𝑜𝑣𝑈𝑆,𝑉𝐶 =
= 0,7332 × 0,152 + 0,2672 × 0,302 + 2 × 0,733 × 0,267 × 0,5 = 0,0273
𝜎(𝑟𝑃 ) = √0,0273 = 0,1653
2) Investor A has $50 million and she is fairly risk-averse, she will only accept a standard
deviation of 12% in her portfolio returns. What dollar amounts should Investor A invest
(or short, if you get a negative amount for an asset category) in each of the three asset
categories (riskless asset, US stocks, venture capital)? What expected return (in %) can
Investor A expect on this portfolio?
How much to invest in a Risk-free asset and Tangency Portfolio?
𝜎(𝑟𝑃 ) = 𝑤𝑀 ∙ 𝜎𝑀 ⟺ 12% = 𝑤𝑀 ∙ 0,1653 ⟺ 𝑤𝑀 = 0,7261 = 72,61%
Invest 72,61% in tangency portfolio and 27,39% in risk-free asset!
B. Consider the Tangency Portfolio from the two-stock example: the expected return is 10,17%
and the standard deviation is 9,54%. Now assume that the investor wants to invest in an
optimal portfolio. The risk-free rate is 8%.
1) If the investor has a risk aversion coefficient of 𝛾 = 2, what is the fraction of wealth that
she should invest in the tangency portfolio?
The fraction of wealth invested in the tangency portfolio should be:
10,17% − 9,54%
𝑤𝑀 = = 1,19
2 × 9,54%2
2) And what is the expected return of the portfolio?
𝐸(𝑟𝑃 ) = 𝑤𝑀 ∙ 𝐸(𝑟𝑀 ) + (1 − 𝑤𝑀 ) ∙ 𝑟𝑓 = 1,19 ∙ 0,1017 + (1 − 1,19) ∙ 0,08 = 10,59%
Given this assumption, everyone has the same efficient frontier and holds the same tangency
portfolio (of stocks). Then, the tangency portfolio is the market portfolio.
Market Equilibrium
Every investor solves the mean-variance problem and holds a combination of risk-free assets and
a portfolio of risky assets (tangency). The sum of all investors’ risky portfolios will have the same
weights as tangency one.
In equilibrium, the sum of all investors’ desired portfolios must equal the aggregate supply and
demand of assets is the market portfolio.
[𝑬(𝒓𝑴 ) − 𝒓𝒇 ]
𝑬(𝒓𝑷 ) = ∙ 𝝈(𝒓𝑷 ) + 𝒓𝒇
𝝈𝑴
Investors choose a point along the line – Capital Market Line (CML). Efficient portfolios are a
combination of the risk-free asset and the market portfolio 𝑀.
Although investors have the same CML, depending on the risk aversion the investor
chooses along the line!
𝜕𝐸(𝑟𝑃 ) 𝜕𝜎 2 (𝑟𝑃 )
= 𝐸(𝑟𝑖 ) − 𝐸(𝑟𝑀 ) = 2 ∙ [𝑤𝑖 ∙ 𝜎𝑖2 − (1 − 𝑤𝑖 ) ∙ 𝜎𝑀
2
+ (1 − 2𝑤𝑖 ) ∙ 𝜎𝑖,𝑀 ]
𝜕𝑤𝑖 𝜕𝑤𝑖
In equilibrium (𝑀) excess demand for stock 𝑖 is zero; now evaluate change in risk and return for
𝑤 = 0. The risk-return trade-off (Sharpe ratio) in equilibrium (𝑀) needs to be equal to the CML
Sharpe ratio. So, we get the following:
𝜕 𝐸(𝑟𝑃 )
𝜕 𝑤𝑖 𝐸(𝑟𝑖 ) − 𝐸(𝑟𝑀 ) 𝐸(𝑟𝑖 ) − 𝐸(𝑟𝑀 ) 𝐸(𝑟𝑀 ) − 𝑟𝑓
= 2 ⟹ 2 =
𝜕 𝜎(𝑟𝑃 ) 𝜎𝑖,𝑀 − 𝜎𝑀 𝜎𝑖,𝑀 − 𝜎𝑀 𝜎𝑀
𝜕 𝑤𝑖 𝜎𝑀 𝜎𝑀
𝝈𝒊,𝑴
⟺ 𝑬(𝒓𝒊 ) = 𝒓𝒇 + 𝟐 [𝑬(𝒓𝑴 ) − 𝒓𝒇 ]
𝝈𝑴
𝜎𝑖,𝑀
Usually, the above equation is written with a 𝛽 = 2 which is a measure of the responsiveness
𝜎𝑀
of stock to movements in the market portfolio (i.e., systematic risk).
𝑬(𝒓𝒊 ) = 𝒓𝒇 + 𝜷 ∙ [𝑬(𝒓𝑴 ) − 𝒓𝒇 ]
Expected Return on Stock = Risk-free rate + Beta of Stock × Master Risk Premium
𝛽𝑃 = ∑ 𝑤𝑖 ∙ 𝛽𝑖
𝑖=1
CML plots the relationship between expected SML is the relationship between expected
returns and standard deviation for efficient returns and 𝛽.
portfolios.
All portfolios, whether efficient or not, must lie on the SML but only efficient portfolios are on
the CML.
𝐸(𝑟𝐴 ) = 𝐸(𝑟𝐵 ) ⇎ 𝜎𝐴 = 𝜎𝐵 In other words, the only relevant measure of risk for pricing
securities is 𝛽 (a measure of covariance or marginal variance).
⟺ 𝛽𝐴 = 𝛽𝐵
4. Estimations
To apply to real life, we have to discover reliable proxy/estimations for the following inputs.
➢ Market Portfolio
In CAPM, the market portfolio should include all assets in the world. However, that´s not possible
to do in real life so we will use a broad and value-weighted stock market index as a proxy (for
example, MSCI World, S&P 500).
➢ Risk-free Rate: 𝑟𝑓
CAPM says that should be riskless and match the horizon of the investment. People use 3 months
of Treasury Bills.
➢ Beta: 𝛽
Note that the CAPM formula is a relationship based on expectations. So, we usually use a time-
series regression based on realized returns:
𝑟𝑖,𝑡 − 𝑟𝑓,𝑡 = 𝛼𝑖 + 𝛽𝑖 (𝑟𝑀,𝑡 − 𝑟𝑓,𝑡 ) + 𝜖𝑖,𝑡 ⟺ 𝑌 = 𝛼𝑖 + 𝛽𝑖 ∙ 𝑋 + 𝜖𝑖,𝑡
Characteristics:
• It may change over time, 𝑡.
• Do not use data from too long ago.
• Five years of weekly or monthly data is reasonable.
➢ Risk: 𝜎
The standard deviation of stock returns can be broken down into systematic risk and idiosyncratic
risk.
𝑟𝑖 − 𝑟𝑓 = 𝛼𝑖 + 𝛽𝑖 (𝑟𝑀 − 𝑟𝑓 ) + 𝜖𝑖
➢ Jensen´s Alpha: 𝛼𝑖
Excess return over that predicted by CAPM, therefore, is the intercept in the above formula. It is
used as a measure of portfolio performance.
𝛼𝑖 = (𝑟̅𝑖 − 𝑟𝑓 ) − 𝛽𝑖 (𝑟̅̅̅
𝑀 − 𝑟𝑓 )
𝛼𝑖 > 0 ⟹ Security has earned a higher return on average than is required for its risk level.
1.2. Assumptions
Although not all assumption is reasonable, they are essential to start studying this subject.
• Perfect capital markets:
- no transaction costs and no taxes
- no bankruptcy costs.
- no agency costs.
- no asymmetric information
- no arbitrage
• Firms and individuals can borrow at the same rate.
𝐷𝐿 + 𝐸𝐿 > 𝐸𝑈
Warning: Correct tax rate to value interest tax shield is the expected marginal tax rate (or
expected increase in firm’s tax liability when its taxable income increases by $1). This may not
be the statutory rate3!
The firm may not always be taxable. Earnings may not be large enough to fully utilize the shield
Losses can be carried forward (and also carried back in some countries). Non-debt tax shields
may already suffice to offset earnings: Depreciation; Investment tax credit.
3
rate imposed by law on taxable income.
➢ Agency Costs
Conflicts of interest between shareholders and debt holders - 3 Selfish Strategies:
1) Incentive to take large risks (risk shifting) and overinvestment
Shareholders may take high-risk, negative NPV projects in the hope of realizing the
upside potential, leaving bondholders to bear the downside risk.
Example:
What happens if the firm is liquidated today? Bondholders get 200 (= Bond Market Value) and
shareholders get nothing.
Without gamble: PV of Bonds = 200; PV of Stocks = 0. Let´s consider the option of gambling
which has an initial investment of 200€ (all the firm’s cash). Required return is 50% (discount
rate). What is the 𝑁𝑃𝑉 of “The Gamble” project?
Expected cash flow gamble = 1000 × 0.10 + 0 = 100€
100
𝑁𝑃𝑉 = −200 + = −133
1 + 50%
Will the project be accepted by shareholders?
Expected cash flow from gamble
• To Bondholders = 300 × 0,10 + 0 = 30€
• To Shareholders = (1000 − 300) × 0,10 + 0 = 70€
30 70
𝑁𝑃𝑉Bond = = 20 𝑁𝑃𝑉Stock = = 47
1 + 50% 1 + 50%
So, this is a perfect example of Risk Shifting, shareholders gain with gamble at expenses of
bondholders and decide to do a project with negative NPV!
In terms of the Pie Theory talked earlier if we consider Taxes and bankruptcy costs can be viewed
as just another claim on the cash flows of the firm and that:
𝐺 ⟹ Payments to Government 𝐿 ⟹ Bankruptcy Cost
So, now the value of a firm is:
𝑽=𝑬+𝑫+𝑮+𝑳
• Stock price increases with leverage and vice-versa (consistent with M&M with taxes)
• Firms signal good news when they lever up (signaling)
There are differences in capital structure across industries. But evidence shows that firms behave
as if they had a target debt-to-equity ratio.
Determinant of 𝛽𝐸
We could divide it into two main risks: Business Risk (BR) and Financial Risk (FR).
• BR: Cyclicality of Revenues
High Cyclical Stocks ⟹ High 𝛽 Cyclicality ≠ Variability: High 𝜎 ⇏ High 𝛽
Revenues depend on the business cycle, for example, retailers.
Project Debt Capacity: 33% × (50 + 43) = 31, that is, the firm requires 31 in additional debt to
maintain its target leverage ratio if it undertakes the project.
We can now use 𝑊𝐴𝐶𝐶 to discount free cash flows to find the project’s 𝑁𝑃𝑉:
10
𝑁𝑃𝑉 = −50 + = 17,6
14,8%
3.2. Adjusted Present Value (APV)
Value of a project can be thought of as the value of the project to an unlevered firm (𝑁𝑃𝑉) plus
the present value of financing side effects (𝑁𝑃𝑉𝐹):
𝐴𝑃𝑉 = 𝑁𝑃𝑉 + 𝑁𝑃𝑉𝐹
There are four side effects of financing:
• Interest tax shield
• Costs of issuing new securities (flotation costs)
• Costs of financial distress
• Subsidies to debt financing
Example:
Consider a project with the following incremental
after-tax cash flows for an all-equity firm:
Unlevered cost of equity is 𝑟𝑈 = 9%.
A. Suppose firm finances the project with 600€ of debt at 𝑟𝐷 = 8% and the tax rate is 30%.
125 250 375 500
𝑁𝑃𝑉 = −1000 + (1+9%) + 1,092 + 1,093 + 1,094 = −31,1 ⟹ Unleverd firm do not Accept
𝐴𝑃𝑉 = 𝑁𝑃𝑉 + 𝑁𝑃𝑉𝐹 = −31,1 + 47,69 = 16,6 ⟹ Firm should Accept the project with Debt!
30%×6,06
Flotations Costs = 606.06 × 1% = 6.06 ⇒ Generates a Tax Shield = 4
= 0,45 each year
0,45 0,45 0,45 0,45 0,45 1
𝑁𝑃𝑉𝐹 = −6,06 + 1+8% + 1,082 + 1,083 + 1,084 = −6,06 + 8%
[1 − 1,084 ] = −4,57
𝐴𝑃𝑉 = 𝑁𝑃𝑉 + 𝑁𝑃𝑉𝐹 = −31,1 + 48,19 − 4,57 = 12,52 ⟹ Firm should Accept!
C. Suppose now firm finances the project with 600€ of debt at a below-market rate of 2%
(special line of credit, government subsidy)
After-tax Interest Tax Shield = 𝐷 × 𝑟𝐷 × (1 − 𝑡) = 600 × 2% × 70% = 8,4
8,4 1 600
𝑁𝑃𝑉𝐹 = 600 − [1 − 4 ]− = 131,2
8% (1 + 8%) (1 + 8%)4
𝑁𝑃𝑉𝐹 captures both the interest tax shield and the non-market rate effect; notice that we discount
at 8% (market rate).
𝐴𝑃𝑉 = 𝑁𝑃𝑉 + 𝑁𝑃𝑉𝐹 = −31,1 + 131,2 = 100,1 ⟹ Firm accept the project with Debt!
𝐷
➢ WACC and FTE: is constant over time.
𝐸
WACC is by far the most common method. FTE is a reasonable choice for highly levered
firms.
➢ APV: level of debt is known over time.
APV method is frequently used for special situations like interest rate subsidies, flotation costs,
LBOs.
1. Payout Decision
1.1. Dividends
The process for a cash dividend has 4 important dates:
• Declaration Date: Board of Directors declares payment of dividends.
• Ex-Dividend Date: If you purchase the stock on and after the ex-dividend date you are
not entitled to receive a dividend
• Record Date: Corporation prepares a list of all individuals believed to be stockholders
• Payment Date: Stockholders receive the dividend.
In a perfect world, the stock price will fall by the amount of the dividend on the ex-dividend date.
Taxes complicate things a bit. Empirically, the price drop is less than the dividend and occurs
within the first few minutes of the ex-date.
Example:
CashKing Inc. is a 42€ stock about to pay a 2€ cash dividend. Investor Bob owns 80 shares and
prefers a 3€ dividend. Bob’s homemade dividend strategy: Sell 2 shares on the ex-dividend date.
Recall that one of the assumptions underlying the dividend-irrelevance argument is: “The
investment policy of the firm is set ahead of time and is not altered by changes in dividend policy”.
1.3. In Practise
Corporations “smooth” dividends. Fewer companies are paying dividends.
Dividends provide information to the market: Positive reactions in Dividends Increases!
Firms should follow a sensible policy:
• Do not forgo positive NPV projects just to pay a dividend
• Avoid issuing stock to pay dividends
• Consider share repurchase when there are few better uses for the cash
This means that the only way you can get higher returns is by taking on more risk; and there is
no information out there that can be used to construct strategies that earn returns higher than
required for their risk, so money management is a waste of effort.
2. Types of Efficiency
➢ Semi-strong (Public)
Current prices fully reflect all past prices and all publicly available information.
Fundamental analysis - using economic and accounting information
- Sorting through income statements, talking to the company
- Studying industries and the macroeconomy
Some evidence for semi-strong efficiency
- No abnormal returns after public announcements
- Professional money managers do not outperform the market consistently
➢ Strong (Private)
Current prices fully reflect all information, public and private.
Insider trading will not produce profits. Knowing a merger is going to take place before it is
announced publicly will not produce profits. Although illegal, evidence that prices move before
public announcements, suggesting insider information
Mixed evidence: On the one hand, insider trading appears profitable, indicating markets are not
strong form efficient. But on the other hand, these profits are short-lived, suggesting the market
may be close to efficient.