Rates University 2007

Options 1

Greg Knudsen
International Rate Structuring & MTNs
 1

Contents
This module is an introduction to the theory and practice of Options.

We will look at the following theoretical concepts ...

• Option definitions
• Call and Put payoffs
• Volatility
• Black Scholes pricing
• Put / Call Parity
• Option sensitivities (the Greeks)

We will then consider at some specific option examples ...

• FX Options
• Caps & Floors
 2

Option Definitions
An Option is the right, but not the obligation, to buy (Call) or sell (Put) an asset at a specified date
at a specified price.

Note the difference to a Forward which is the obligation to buy (or sell) an asset at a specified future date
at a specified price.

A Forward obliges the participants to transact the agreed exchange on the agreed date at the agreed price.

An Option allows the Option Buyer to choose to exercise the exchange (a purchase or a sale) or not,
depending upon the real or perceived value of that purchase or sale.

We define some common terms …

Call Option the buyer of a Call Option has the right to buy the underlying asset
on a specified date at a specified price (the strike)

Put Option the buyer of a Put Option has the right to sell the underlying asset
on a specified date at a specified price (the strike)

Strike the price at which the asset is bought or sold if the option is exercised

Premium the price of the Option contract

the buyer of the option pays the seller of the option this price at inception

Expiry the Maturity Date of the Option

 3

Option Definitions
European Option a European Option can only be exercised on the specified Option Maturity Date

American Option an American Option can be exercised at any time prior to the Option
Maturity Date

Bermudan Option a Bermudan Option can be exercised (once) on one of a specified set
of exercise dates up to and including the Option Maturity Date
These often co-incide with coupon payment dates on structured bonds and swaps

Option Value the value for which the option can be bought or sold in the market
At inception this is the premium paid to buy the option

Intrinsic Value the value (if positive) of immediate exercise of the Option
For a Call this is Max (0, Asset Value – Strike)
For a Put this is Max (0, Strike – Asset Value)

Time Value the residual value of the option after accounting for the Intrinsic Value
It captures the potential additional payoff from the underlying moving
(deeper) into the money before expiry
By definition … Time Value = Option Value – Intrinsic Value

In-the-money an option is In-the-money if the Intrinsic Value is positive

At-the-money an option is At-the-money if the current Asset Value = Strike

Out-of-the-money an option is Out-of-the-money if immediate exercise would result in a loss

 4

European Options

European Options have a well-established analytical valuation methodology.

This theory is called Black-Scholes Theory

Black-Scholes Theory was developed in the 1970s and was the first truly comprehensive solution to the
problem of European option valuation.

Under a set of assumptions about the distribution of the underlying asset (interest rates, FX rates etc), about
the volatility of the asset, etc, the theory uses arbitrage arguments to derive analytical formulas for calls and
puts.

In the Black-Scholes world, Geometric Brownian Motion is assumed as the dynamic model for the evolution
of the underlying asset.

The variable could be for example

• a short term interest rate
• a spot FX rate
• an observable Index
• a commodity price, etc.
 5

Payoffs and Volatilities

The Payoffs for Calls and Puts are given by Call Payoff X

Call Payoff = Max (0, ST – X) where …

ST = the asset price at Maturity (time T) X ST
X = the strike
X
Put Payoff = Max (0, X – ST) Put Payoff

A key component of Options pricing is the assumed volatility of the underlying

variable or asset. We look at the definitions …
X ST
Volatility a measure of the variability of the price of the underlying Asset or Index
In Black-Scholes, Lognormal Volatility is technically the standard deviation of the
Log of the Asset Price in 1 year’s time.
Lognormal Vol is also effectively the standard deviation of the percentage change
in the asset price over 1 year. For interest rate options, Basis Point Vol is the
standard deviation of the absolute change in the interest rate Index over 1 year.

Historical Vol the historical vol is the observed volatility of the asset price over a specified period.

Implied Vol the implied vol is the volatility implied by the option price traded in the market.
The implied vol may be higher or lower than the historical vol.
Note that implied vol is in a sense forward looking, historical vol backwards looking
Both Puts and Calls become more valuable when the implied volatility increases.
This is because higher volatility implies a higher probability of finishing deep in the money
 6

Calls and Puts … Long & Short

There are 4 basic payoff diagrams for Calls and Puts …

Payoff Payoff

X Asset X Asset
Long Call Long Put

Payoff X L = Libor
Payoff

X
X
Asset
Asset
Short Call Short Put
 7

Black-Scholes Methodology

Black and Scholes proved that, under certain assumptions, an underlying asset could perfectly hedge the
profits and losses of a standard European option by following a self-financing dynamic replication portfolio
strategy.

The Black-Scholes European option price is the discounted expected payoff of the option at Maturity and is
derivable analytically since the distribution of the underlying variable at the sole exercise date is available.

This is of course intuitively appealing. Under Geometric Brownian motion the distribution of the underlying at
Maturity is Lognormal. This distribution can be used to calculate the expected value of the option payoff.
Since this payoff occurs at Maturity it must be discounted in order to calculate the option value (premium)

Call Price vs Maturity Distribution

4.50% The graph shows the Call Price and Payoff for
a Call with as functions of the Forward.
4.00% Lognormal Dist
Also shown are the Lognormal Distributions
Vol = 10%
3.50% corresponding to 10% and 20% volatilities.
3.00%
Lognormal Dist The Call is priced using a volatility of 20%
2.50% Call Price
Price

Vol = 20%
2.00% If the Call were repriced using only 10% vol
then the Call price would fall.
1.50%

1.00% Payoff This is because the payoff is asymmetric.

0.50%
A higher vol implies a greater chance of
extreme positive payoffs.
0.00%
3.00% 3.40% 3.80% 4.20% 4.60% 5.00% 5.40% 5.80% 6.20% 6.60% 7.00%

Forward
 8

Intrinsic Value and Time Value

Intrinsic Value is the gain that could be realised through immediate exercise of the option.
For a Call this is Max (0, Asset – Strike)

Time Value is the remaining value of the option.

It captures the potential additional payoff from the underlying moving (deeper) into the money before expiry.

The graph shows both the Call Price

and Payoff for a Call with Strike X, as
functions of the Forward Rate. Call Price - Intrinsic Value & Time Value

The Call Price is divided into …

- Intrinsic Value
- Time Value
Call Price
For a Call … Payoff
Price

• if the Call is Out-of-the-money

(Forward less than Strike)
Time Value
there is only Time Value
(Intrinsic = 0) Time Value
Intrinsic Value

• if the Call is In-the-money X

(Forward greater than Strike) Forward
there is both Time Value and
Intrinsic Value
Out-of-the-money In-the-money
 9

Puts: Intrinsic Value and Time Value

Intrinsic Value
Option Value

Time Value

X Forward

In-the-money Out-of-the-money
Source: Citigroup

For a Put …
The above graph shows the Price and Payoff for
a Put with Strike X.
• if the Put is Out-of-the-money
(Forward greater than Strike)
The Put Price is again divided into … there is only Time Value
- Intrinsic Value (Intrinsic = 0)
- Time Value
• if the Put is In-the-money
(Forward less than Strike)
there is both Time Value and
Intrinsic Value
 10

Black-Scholes Call Price

Suppose the distribution of an Index at time T in the future is assumed to be lognormal.

A Call on an Index with strike X, maturing at time T, is defined to have the Payoff …

Call Payoff = Max (0, Index – X) Call Payoff

X

The Black-Scholes equation states that the value of the Call is given by

Call Pr ice = e - r ( T - t ) { F N (d1) - X N (d2 ) } X Index

Standard Normal
Normal Distribution
Distribution
where …

F = Index Fwd Rate to time T N(x) = Prob (Z<x)

X = Strike Z is a N(0,1) variable
σ = Volatility of the Index (standard Normal distribution)

N (x)
ln(F / X) + (1/ 2)σ2 (T - t ) d2 = d1 - σ (T - t )
d1 =
σ (T - t )
x
1 2
x
e - r(T- t )
is the discount factor for time T N (x) = ∫ 2π e -(1 2)z dz
-∞
 11

Call Price – evolution over time

We graph the Call price as a function of the Forward Rate F for a selection of Terms to Maturity

Note that the graph of the Call Price approaches the payoff Max (0, Index – X) as the Term to Maturity falls.
The time value falls to 0 and the Call Price converges to the Payoff (intrinsic value)

Call Price
Call Pr ice = e - r ( T - t ) { F N (d1) - X N (d2 ) }
2.50%

2.00%
F = Index Fwd Rate to time T Terms to Maturity
X = Strike 24 months
σ = Volatility of the Index 1.50% 12 months
6 months

Price
Maturity
2 1.00%
ln(F / X) + (1/ 2)σ (T - t )
d1 =
σ (T - t )
0.50%

d2 = d1 - σ (T - t )
0.00%
X
e- r ( T - t ) is the discount factor for time T
Forward
N(x) = Prob (Z<x)
Z is a N(0,1) variable
(standard Normal distribution)
 12

Black-Scholes Put Price

Again we suppose the distribution of an Index at time T in the future is lognormal.

A Put on an Index with strike X, maturing at time T, is defined to have the Payoff …

Call Payoff = Max (0, X – Index ) Put Payoff

X

The Black-Scholes equation states that the value of the Put is given by

Put Pr ice = e- r ( T - t ) { X N (-d2 ) - F N (-d1) } X Index

Standard Normal
Normal Distribution
Distribution
where …

F = Index Fwd Rate to time T N(x) = Prob (Z<x)

X = Strike Z is a N(0,1) variable
σ = Volatility of the Index (standard Normal distribution)

N (x)
ln(F / X) + (1/ 2)σ2 (T - t ) d2 = d1 - σ (T - t )
d1 =
σ (T - t )
x
e- r ( T - t ) is the discount factor for time T
 13

Put Price – evolution over time

We also graph the Put price as a function of the Forward Rate F for a selection of Terms to Maturity

Note that the graph of the Put Price approaches the payoff Max (0, X – Index) as the Term to Maturity falls.
The time value falls to 0 and the Call Price converges to the Payoff (intrinsic value)

Put Price
Put Pr ice = e- r ( T - t ) { X N (-d2 ) - F N (-d1) }
2.50%

F = Index Fwd Rate to time T 2.00% Terms to Maturity

X = Strike 24 months
σ = Volatility of the Index 1.50% 12 months
6 months

Price
Maturity
2
ln(F / X) + (1/ 2)σ (T - t ) 1.00%
d1 =
σ (T - t )
0.50%

d2 = d1 - σ (T - t )
0.00%

e- r ( T - t ) is the discount factor for time T X

Forward
N(x) = Prob (Z<x)
Z is a N(0,1) variable
(standard Normal distribution)
 14

Put - Call Parity

Put - Call Parity for European options implies that for identical Calls and Puts …

Call – Put = Long Forward

Payoff

Put – Call = Short Forward Long Call

To see this we look at the Payoffs at expiry … X Index

+
Call Payoff = Max (0, Index – X)
Payoff
Put Payoff = Max (0, X– Index )
Short Put

X Index
Since Max (0, Index – X) – Max (0, X – Index) = (Index – X)
=
it follows that …
Payoff

Call Payoff – Put Payoff = Long Forward Payoff Long

Forward
and
X Index
Put Payoff – Call Payoff = Short Forward Payoff
 15

Put - Call Parity

We can also check Put-Call Parity against the prices of European Calls and Puts

Call Pr ice = e - r ( T - t ) { F N (d1) - X N (d2 ) } (1)

Put Pr ice = e - r ( T - t ) { X N (-d2 ) - F N (-d1) } (2)

where … Normal Distribution
Standard Normal Distribution
N(x) = Prob (Z<x)
F = Index Fwd Rate at time T
X = Strike Z is a N(0,1) variable
σ = Volatility of the swap rate (standard Normal distribution)
ln(F / X) + (1/ 2)σ2 (T - t ) d2 = d1 - σ (T - t )
d1 =
σ (T - t )
N (–x) 1 – N (x)
Since Z is symmetric distribution around 0, we have for any x
Prob (Z< -x) = 1 – Prob (Z> -x) = 1 – Prob (Z< x)
–x 0 x
ie N(–x) = 1 – N(x)

Then … Call Pr ice - Put Pr ice = e - r ( T - t ) { [F N (d1) - X N (d2 )] - [ X N (-d2 ) - F N (-d1)] }

= e - r ( T - t ) { F [ N (d1) + N (-d1)] - X [ N (d2 ) + N (-d2 )] }

= e- r(T - t ) { F - X }

= Price of entering a long Forward on the Index struck at X

 16

Put - Call Parity

Similarly, (2) – (1) gives …

Put Pr ice - Call Pr ice = e - r ( T - t ) { X - F }

= Price of entering a short Forward on the Index struck at X

Now, If the Call and Put are at-the-money (ATM), so that X = F = Index Forward Price

then Call (F) – Put (F) = Buy Index forward at F

or
Call (F) – Put (F) = e- r(T - t ) { F - F } = 0

and

Price (at the money Call) = Price (at the money Put)
 17

The Greeks
In pricing and trading options, it is important to consider one or more risk parameters, collectively called
the Greeks.

These parameters measure the sensitivity of the option price to a variety of variables.

The most common greeks are

• Delta sensitivity of price to change in the underlying Index (or the Forward)
• Vega sensitivity of price to change in the volatility
• Theta sensitivity of price to change in time
• Gamma sensitivity of Delta to change in the underlying Index (a 2nd order effect)

Standard Normal
Normal Distribution
Distribution
For example, we know the Black-Scholes value of a European Call is …

Call Pr ice = e - r ( T - t ) { F N (d1) - X N (d2 ) }

where …

T = Option Maturity N(x) = Prob (Z<x)

N (x)
t = today (valuation date) Z is a N(0,1) variable
F = Index Fwd Rate to time T (standard Normal distribution)
X = Strike
ln(F / X) + (1 / 2)σ 2 (T - t ) x
σ = Volatility of the Index d1 =
σ (T - t )
d2 = d1 - σ (T - t )
 18

The Greeks
We can use Calculus to derive the Greeks. To do this we differentiate the Call Price with respect to the
variable for which we require sensitivity (keeping other variables constant)

Sparing you the details (see the Appendix) we have for Calls ...
∂c
Delta = = e - r ( T - t ) N (d1)
∂F
∂c 1 2
Vega = = e- r(T - t ) F e -(1 2)d1 (T - t )
∂σ 2π
∂2c 1 2 1
Gamma = 2 = e- r(T - t ) e - (1 2)d1
∂F 2π Fσ ( T - t )

Looking now at Puts we differentiate the Put Price with respect to the variable for which we require
sensitivity (keeping other variables constant)

Thus for Puts …

Note that the Vega and Gamma for a Put are identical
∂p to the Vega and Gamma for a Call.
Delta = = e - r ( T - t ) N (d1) - 1
∂F This follows from Put – Call Parity which states …
∂p 1 2 Call – Put = Long Forward
Vega = = e- r(T - t ) F e -(1 2)d1 (T - t )
∂σ 2π
2
∂ p 1 2 1
Gamma = 2 =e
- r(T - t )
e - (1 2)d1
∂F 2π Fσ ( T - t )
 19

Call Delta
The Delta of a Call measures the sensitivity of the Call Price with respect to the underlying.

∂c Delta is the rate of change of the Call Price

Delta =
∂F with respect to the underlying forward rate F

The Delta will therefore be the slope of the graph of Call Price as a function of the forward rate F

Call … Price, Payoff and Delta

The graph on the right shows … 2.50%

• Call Price as a function of F

2.00% Delta is the
• Call Payoff as a function of F rate of change
• Call Delta as the slope of the Call Price (slope) of Price
1.50%
Price
where F = Forward Rate
1.00%
Call Price
The Strike on this Call is 5.00%.
0.50% Call Payoff
At the Money (when the Forward Rate = 5.00%)
the Delta is approximately 0.50 0.00%
3.00% 3.40% 3.80% 4.20% 4.60% 5.00% 5.40% 5.80% 6.20% 6.60% 7.00%

Deep In the Money (Forward Rate >> 5.00%) Forward

out of the money in the money
the Delta approaches 1 as the Call behaves
like an outright long Forward Strike = 5.00%
 20

Call Delta
The Delta of a Call measures the sensitivity of the Call Price with respect to the underlying.

Here we show the Delta of an at-the-money 2yr forward Caplet with strike X = 4.84%

The Delta is calculated and graphed as a function of the forward rate F

Gamma is the
∂c rate of change
Delta = Caplet Delta
∂F (slope) of Delta

1.00

Delta = e - r ( T - t ) N (d1)
0.90
at the money
0.80 Delta ~ 0.50
0.70

N(x) = Prob (Z<x) 0.60

in the money
Z is a N(0,1) variable 0.50 Delta approaches 1
out of the money
(standard Normal distribution) 0.40
Delta approaches 0
0.30
2
ln(F / X) + (1 / 2)σ (T - t ) 0.20
d1 =
σ (T - t ) 0.10

0.00

e- r ( T - t ) is the discount factor for time T

0.00% 1.00% 2.00% 3.00% 4.00% 5.00% 6.00% 7.00% 8.00% 9.00% 10.00%

out of the money Forward Rate in the money

 21

Put Delta
The Delta of a Put measures the sensitivity of the Put Price with respect to the underlying.

∂p Delta is the rate of change of the Put Price

Delta =
∂F with respect to the underlying forward rate F

The Delta will therefore be the slope of the graph of Put Price as a function of the forward rate F

Put … Price, Payoff and Delta

The graph on the right shows … 2.50%

• Put Price as a function of F

2.00%
• Put Payoff as a function of F
Delta is the
• Put Delta as the slope of the Put Price 1.50% rate of change
where F = Forward Rate Price (slope) of Price
1.00%
Put Price
The Strike on this Put is 5.00%.
0.50%
At the Money (when the Forward Rate = 5.00%) Put Payoff
the Delta is approximately 0.50 0.00%
3.00% 3.40% 3.80% 4.20% 4.60% 5.00% 5.40% 5.80% 6.20% 6.60% 7.00%

Deep In the Money (Forward Rate << 5.00%)

in the money Forward out of the money
the Delta approaches 1 as the Put behaves
like an outright short Forward Strike = 5.00%
 22

Put Delta
The Delta of a Put measures the sensitivity of the Put Price with respect to the underlying.

Here I show the Delta of an at-the-money 2yr forward Floorlet with strike X = 4.84%

The Delta is calculated and graphed as a function of the forward rate F

Gamma is the
∂p rate of change
Delta = Floorlet Delta
∂F (slope) of Delta
0.00

-0.10
- r(T - t )
Delta = e N (d1) - 1 -0.20 at the money
Delta ~ - 0.50
-0.30

-0.40
N(x) = Prob (Z<x) out of the money
Z is a N(0,1) variable
-0.50 Delta approaches 0
-0.60
(standard Normal distribution) in the money
-0.70 Delta approaches -1
ln(F / X) + (1 / 2)σ 2 (T - t ) -0.80
d1 = -0.90
σ (T - t )
-1.00
0.00% 1.00% 2.00% 3.00% 4.00% 5.00% 6.00% 7.00% 8.00% 9.00% 10.00%
e- r ( T - t ) is the discount factor for time T
in the money Forward Rate out of the money
 23

Vega
The Vega of an option measures the sensitivity of the option with respect to the volatility.

Here I show the Vega of an at-the-money 2yr forward Caplet with strike X = 4.84%

The Vega is calculated and graphed as a function of the Forward Rate F

Caplet Vega

3.00%

1 2 Vega is largest
Vega = e - r ( T - t ) F e -(1 2)d1 (T - t ) 2.50% at the strike
2π
2.00%

ln(F / X) + (1 / 2)σ 2 (T - t ) Vega smaller

d1 = 1.50% Vega smaller as call goes into
σ (T - t ) out of the money the money
1.00%
- r(T- t )
e is the discount factor for time T
0.50%

(T - t) = term to Maturity of the Option 0.00%

0.00% 1.00% 2.00% 3.00% 4.00% 5.00% 6.00% 7.00% 8.00% 9.00% 10.00%

out of the money Forward Rate in the money

 24

Gamma
The Gamma of an option measures sensitivity of Delta to change in the underlying Index

Here I show the Gamma of an at-the-money 2yr forward Caplet with strike X = 4.84%

The Gamma is calculated and graphed as a function of the Forward Rate F

Caplet Gamma

30.00
1 2 1 at the money
Gamma = e - r ( T - t ) e - (1 2)d1 Delta changing
2π Fσ (T - t ) 25.00
most quickly
20.00

ln(F / X) + (1 / 2)σ 2 (T - t )
d1 = 15.00
σ (T - t ) Delta changes
10.00 less as call goes
- r(T- t ) into the money
e is the discount factor for time T
5.00

(T - t) = term to Maturity of the Option 0.00

0.00% 1.00% 2.00% 3.00% 4.00% 5.00% 6.00% 7.00% 8.00% 9.00% 10.00%

out of the money Forward Rate in the money

 25

Call Spread

A Call Spread is a combination of a bought Call and a sold Call at different strikes.

The Long Call is at the Low Strike, the Short Call at the High strike

Payoff = Max (0, IT – X1) – Max (0, IT– X2)

0 if IT < X1 < X2
= (IT – X1) if X1 < IT < X2
(X2 – X1) if X1 < X2 < IT

X
X2 – X1
where … X1 = Low Strike
IT = Index fixing at Maturity (time T) X2 = High Strike

X1 = Low Strike
X2 = High Strike X1 X2 Index
 26

Put Spread

A Put spread is a combination of a bought Put and a sold Put at different strikes.

The Long Put is at the High Strike, the Short Put at the Low strike

Payoff = Max (0, X2 – IT) – Max (0, X1 – IT)

(X2 – X1) x DayCount if L < X1 < X2

= (X2 – L) x DayCount if X1 < L < X2
0 if X1 < X2 < L

X L = Libor
X2 – X1
X1 = Low Strike
where … X2 = High Strike
IT = Index fixing at Maturity (time T)
X1 = Low Strike
X2 = High Strike X1 X2 Index
 27

FX Options
An FX Option maturing at time T gives the option holder the right to exchange say M units of CCY 1 for
N units of CCY2.

This is best explained by looking at an example.

Exercise of EUR Put / USD Call

Consider an option that allows the option holder at Maturity to …
EUR P€
sell P€ EUR in return for receiving 1.30 x P€ USD. Option Option
Buyer Seller
We say that this option is a EUR Put / USD Call struck at X = 1.30 USD 1.30 x P€

Conversely a EUR Call / USD Put struck at 1.30 allows the holder
Exercise of EUR Call / USD Put
the right to …
USD 1.30 x P€
buy P€ EUR in return for paying 1.30 x P€ USD Option Option
Buyer Seller
X
EUR P€
 28

FX Options
Note that in an FX Option there are by definition 2 currencies involved and potentially exchanged.
This can create confusion when comparing to regular calls and puts which are settled in a single currency.

Consequently, in considering FX Options, it is very useful to divide the 2 currencies involved into a
Commodity Currency and a Terms currency.

It is supposed then that the Spot Rate between the 2 currencies equates 1 unit of the Commodity Currency
to S units of the Terms Currency.

Spot 1 unit of Comm CCY = S units of Terms CCY S = Spot Rate

The option holder then has the right to buy (or sell) 1 unit of the Commodity Currency in return for paying X
units of the Terms CCY.

In our example … EUR is the Commodity Currency

USD is the Terms Currency

Exercise of EUR Put / USD Call Exercise of EUR Call / USD Put

EUR 1 EUR 1
Option Option Option Option
Buyer Seller Buyer Seller
USD X USD X

Option Buyer has the right to Option Buyer has the right to
sell EUR 1 against USD X buy EUR 1 against USD X
 29

FX Option Price … Commodity Call / Terms Put

We assume as usual that the distribution of the Spot FX rate at time T in the future is lognormal.

A Commodity CCY Call / Terms CCY Put maturing at time T will have the payoff (in Terms CCY) …

Payoff = Max (0, ST – X) value in Terms CCY Comm Call / Terms Put

Note .. at time T, 1 Comm CCY = ST Terms CCY Comm 1

Option Option
The Black-Scholes equation states that the price in Terms CCY of the Buyer Seller
Comm Call / Terms Put is given by Terms X

Comm Call Pr ice = e - r ( T - t ) { FT N (d1) - X N (d2 ) } price in Terms CCY

Standard Normal
Normal Distribution
Distribution
where …
ln(FT / X) + (1 / 2)σ 2 (T - t )
FT = Fwd FX Rate to time T d1 =
σ (T - t )
X = Strike
σ = Spot Volatility d2 = d1 - σ (T - t )
N (x)
e - r ( T - t ) = discount factor for time T N(x) = Prob (Z<x)
Z is a N(0,1) variable
(std Normal distribution)
Note that FT is given by x

dc ( T ) dc(T) is the Comm discount factor for time T

x
1 2
FT = S ×
dt (T ) dt(T) is the Terms discount factor for time T N (x) = ∫ 2π e -(1 2)z dz
-∞
 30

FX Option Price … Commodity Put / Terms Call

We assume again that the distribution of the Spot FX rate at time T in the future is lognormal.

A Commodity CCY Put / Terms CCY Call maturing at time T will have the payoff (in Terms CCY) …

Payoff = Max (0, X – ST) value in Terms CCY Comm Put / Terms Call

Note .. at time T, 1 Comm CCY = ST Terms CCY Comm 1

Option Option
The Black-Scholes equation states that the price in Terms CCY of the Buyer Seller
Comm Put / Terms Call is given by Terms X

Comm Put Pr ice = e - r ( T - t ) { X N (-d2 ) - FT N (-d1) } price in Terms CCY

Standard Normal
Normal Distribution
Distribution
where …
ln(FT / X) + (1 / 2)σ 2 (T - t )
FT = Fwd FX Rate to time T d1 =
σ (T - t )
X = Strike
σ = Spot Volatility d2 = d1 - σ (T - t )
N (x)
e - r ( T - t ) = discount factor for time T N(x) = Prob (Z<x)
Z is a N(0,1) variable
(std Normal distribution)
Note that FT is given by x

dc ( T ) dc(T) is the Comm discount factor for time T

x
1 2
FT = S ×
dt (T ) dt(T) is the Terms discount factor for time T N (x) = ∫ 2π e -(1 2)z dz
-∞
 31

Caps
A Cap is a basic Call on a floating Index, for example Libor or Euribor.

They frequently occur in structures where coupons dependent on a floating Index are either …

• capped by a maximum (a coupon cap)

• floored by a minimum (a coupon floor)

The Cap payoff, which is typically paid at the end of the coupon period to which it applies, is defined as …

Cap Payoff = Max (0, L - X) x DayCount = (L - X) x DayCount if L > X

0 otherwise

where …
Caplet X
Payoff
L = Libor fixing
X = Strike
DayCount = day count of the underlying period.

XX Libor
In practice, we usually call this option a Caplet, reserving the term Cap
for a series of consecutive caplets.

For example, a 5yr Cap might be defined as the series of 10 semi-annual caplets.
This is a series of calls on the Index, each with a common strike.
 32

Caplets
The Payoff in a Caplet usually occurs at the end of the underlying period. This helps ensure that the caplet
acts in harmony with floating rate coupons to “cap” the coupon.

For example, suppose we have a floating rate Note bearing a coupon of

3m USD Libor + 0.50% paid in arrears

If we impose the condition that the maximum coupon permitted in a period is 6.00%, then we have
implicitly added a caplet struck at 5.50% to the coupon.

Then the coupon can be decomposed as ..

Min (3m USD Libor + 0.50%, 6.00%) = (3m USD Libor + 0.50%) + Min (0, 5.50% – 3m USD Libor)
= (3m USD Libor + 0.50%) - Max (0, 3m USD Libor – 5.50%)

Long a floating coupon Short a Caplet struck at 5.50%

Then the Coupon payment at the end of a 91 day period is ..

Coupon
if Libor <= 5.50% [(3m Libor + 0.50%) x 91/360] 6.00%

if Libor > 5.50% [(3m Libor + 0.50%) x 91/360]

– [(3m Libor – 5.50%) x 91/360]
Caplet payoff
5.50% X Libor
= 6.00% x 91/360

Thus the addition of a short caplet position has “capped” the coupon at 6.00%
 33

Cap Pricing
We now look to use the Black-Scholes methodology to price a Caplet.

We assume a lognormal distribution for Libor, and adopt the following time-line

- Today is denoted by time t

t T T+∆ - the caplet matures (is fixed) at time T
- the payoff of the caplet is at time T+ ∆
Option Period Underlying

A caplet is a call on Libor, maturing at time T with strike X, and paid at time T+ ∆.

Caplet Payoff
Caplet Payoff = Max (0, L - X) x ∆ = (L - X) x ∆ if L > X
0 otherwise
X
Thus the price of the caplet is given by
XX Libor
∆
Caplet Pr ice = × dT × { FT N (d1) - X N (d2 ) }
(1 + FT ∆ )

where … ln(FT / X) + (1/ 2)σ2 (T - t )

FT = Libor Forward Rate (T, T+∆) d1 =
σ (T - t )
X = Strike
σ = Volatility of Libor d2 = d1 - σ (T - t )
dT = discount factor to time T (dT dT + ∆ - 1)
N(x) = Prob (Z < x), Z~ N(0,1) FT =
∆
 34

Floors
A Floor is similarly a basic Put on a floating Index, for example Libor or Euribor.

Again they frequently occur in structures where the coupons dependent on a floating Index are either
capped by a maximum or floored by a minimum.

The Floor payoff, which is typically paid at the end of the coupon period to which it applies, is defined as …

Floor Payoff = Max (0, X - L) x DayCount = (X - L) x DayCount if L < X

0 otherwise

where …
X Payoff
Floorlet
L = Libor fixing
X = Strike
DayCount = day count of the underlying period.

X Libor
As for caps, we often call this option a Floorlet, reserving the term Floor
for a series of consecutive floorlets.
 35

Floorlets
The Payoff in a Floorlet usually occurs at the end of the underlying period. This helps ensure that the
floorlet acts in harmony with floating rate coupons to “floor” the coupon.

For example, suppose we have a floating rate Note bearing a coupon of

3m USD Libor + 0.50% paid in arrears

If we impose the condition that the minimum coupon permitted in a period is 2.00%, then we have implicitly
added a floorlet struck at 1.50% to the coupon.

Then the coupon can be decomposed as ..

Max (3m USD Libor + 0.50%, 2.00%) = (3m USD Libor + 0.50%) + Max (0, 1.50% – 3m USD Libor)

Long a floating coupon Long a Floorlet struck at 1.50%

Then the Coupon payment at the end of a 91 day period is .. Coupon

X
if Libor >= 1.50% [(3m Libor + 0.50%) x 91/360]
2.00%
if Libor < 1.50% [(3m Libor + 0.50%) x 91/360]
+ [(1.50% – 3m Libor) x 91/360]
Floorlet payoff
1.50% Libor
= 2.00% x 91/360

Thus the addition of a long floorlet position has “floored” the coupon at 2.00%
 36

Floor Pricing
We now use the Black-Scholes methodology to price a Floorlet.

We assume a lognormal distribution for Libor, and adopt the following time-line

- Today is denoted by time t

t T T+∆ - the floorlet matures (is fixed) at time T
- the payoff of the floorlet is at time T+ ∆
Option Period Underlying

A floorlet is a put on Libor, maturing at time T with strike X, and paid at time T+ ∆.

Floorlet Payoff
Caplet Payoff = Max (0, X - L) x ∆ = (X - L) x ∆ if L < X
0 otherwise
X
Thus the price of the floorlet is given by
XX Libor
∆
Floorlet Pr ice = × dT × { X N (-d2 ) - FT N (-d1) }
(1 + FT ∆ )

where … ln(FT / X) + (1/ 2)σ2 (T - t )

FT = Libor Forward Rate (T, T+∆) d1 =
σ (T - t )
X = Strike
σ = Volatility of Libor d2 = d1 - σ (T - t )
dT = discount factor to time T (dT dT + ∆ - 1)
N(x) = Prob (Z < x), Z~ N(0,1) FT =
∆
 37

Cap Vols and Caplet Vols

As mentioned, a Cap is essentially a portfolio of Caplets.
The Caplets have identical strikes and are temporally sequential, covering the life of the Cap.

t1 t2 t3 t4 tn tn+1 - caplets are c1, c2,.., cn

- caplet cj is fixed at time tj and paid at time tj+1
c1 c2 c3 cn - each caplet cj has an individual volatility σj

Total Period of the Cap

To allow for “skew”, caplet vols are functions of not just maturity but strike as well.

Each caplet cj has an individual volatility σj(X), a function of strike X.

The caplet is priced using its own vol. But we cannot observe these vols directly.

Instead, what we can observe on trader screens are Cap Vols, which are average (or flat) vols.
They are typically quoted on screens for maturities 1yr, 2yr, 3yr, … etc

If each caplet in the Cap were to be priced using the flat Cap Vol σ(X), then the Total Cap price should
match the Total Cap price obtained by pricing each caplet cj at its individual volatility σj(X)

Thus by observing the flat Cap Vols σ(X) for a series of maturities, we can use a bootstrap
method to obtain caplet vols σj(X) for each underlying period and strike X.
 38

Cap & Floor Put-Call Parity

The Put-Call Parity is easy to visualize graphically ...

Payoff Payoff Payoff

+ =
X Libor

X Libor X Libor

long Caplet short Floorlet pay X, receive Libor

Payoff Payoff Payoff

short Caplet + long Floorlet = receive X, pay Libor

X Libor X Libor X Libor