Chapter - 13

Options on Stock Indexes, Foreign Currencies,

Futures Contracts, and Volatility Indexes

1 The Merton Model

Before we go on to analyze options on stock indexes, foreign currencies, and futures

contracts, let us first derive an equivalent of the Black-Scholes formula for a stock
that pays a continuous dividend yield. This model was derived by Merton and
has implications for option pricing models for other financial assets such as stock
indexes and foreign currencies.

Take the case of a stock that evolves from a current price of St to a value of
ST by time T. If this stock were to pay a continuous dividend yield at the rate
of δ, the dividend can be construed as a leakage of value from it. Thus, if it
were to pay such a dividend it would evolve to a value of ST e−δ(T −t) by time T.
This price movement is identical to what a non-dividend paying stock which is
currently price at St e−δ(T −t) would experience. Thus the Black-Scholes formula
can be applied to the stock paying a continuous yield of δ, if we replace its price
St with St e−δ(T −t).

We know that for a non-dividend paying stock

CE,t = StN(d1 ) − Xe−r(T −t)N(d2 ) (13.1)

where
σ2
!
St
 
ln + r+ (T − t)
X 2
d1 = q (13.2)
σ (T − t)
and q
d2 = d1 − σ (T − t) (13.3)

658
If we substitute St with St e−δ(T −t) we get

CE,t = Ste−δ(T −t)N(d1 ) − Xe−r(T −t)N(d2 ) (13.4)

where
σ2
!
St
 
ln + r−δ+ (T − t)
X 2
d1 = q (13.5)
σ (T − t)
and q
d2 = d1 − σ (T − t) (13.6)
The corresponding formula for European puts on a stock that pays a continuous
dividend yield is

PE,t = Xe−r(T −t) N(−d2 ) − St e−δ(T −t)N(−d1 ) (13.7)

where d1 and d2 are as defined for the call options.

1.1 The Underlying Rationale

We defined the Ito process for the stock price of a non-dividend paying stock as

dS = µSdt + σSdZ (13.8)

The equivalent process for a stock that pays a continuous dividend yield is

dS = (µ − δ)Sdt + σSdZ (13.9)

Consider a derivative security F that is a function of S and t.

1 ∂ 2F 2 2
" #
∂F ∂F ∂F
dF = (µ − δ)S + + × 2
σ S dt + σSdZ (13.10)
∂S ∂t 2 ∂S ∂S
∂F
Let π be the value of a portfolio that is long in units of the stock and short
∂S
in one unit of the derivative.
∂F
π= S−F (13.11)
∂S
The change in π over a small interval of time 4t is given by
∂F
4π = [4S + δS4t] − 4F
∂S
659
 1 ∂ 2F 2 2 ∂F
" #
∂F
= − − σ S + δS 4t (13.12)
∂t 2 ∂S 2 ∂S
Since this portfolio by construction is instantaneously riskless

4π = rπ4t

where r is the riskless rate of interest. Hence

1 ∂ 2F 2 2 ∂F
" #
∂F
− − σ S + δS 4t
∂t 2 ∂S 2 ∂S
!
∂F
= r −F + S 4t
∂S
∂F ∂F 1 ∂ 2F 2 2
⇒ + (r − δ)S + σ S − rF = 0 (13.13)
∂t ∂S 2 ∂S 2
The Merton model is the solution to this equation by applying the relevant bound-
ary conditions. For a European call option, the conditions are:

• C(S, T ) = Max[0, ST − X]

• C(0, t) = 0

• C(S, t) ∼ Se−δ(T −t) as S → ∞

The first two conditions are identical to those for a non-dividend paying stock.
The third condition is however different, and reflects the fact that there is leakage
of value from the stock price.

1.1.1 Example

Consider a stock that is currently priced at $ 100. Call and put options with an
exercise price of $ 100 and six months to maturity are available. The riskless rate
of interest is 10% per annum and the volatility of stock returns is 30% per annum.
The stock pays a continuous dividend yield of 5% per annum.

100 0.30 × 0.30

   
ln + 0.10 − 0.05 + 0.50
d1 = 100 √ 2
0.30 0.50
660
 = 0.2240

d2 = 0.2240 − .2121 = 0.0119

N(d1 ) = 0.5886 and N(d2 ) = 0.5047

CE,t = 100e−.05×0.5 × 0.5886 − 100e−.10×0.5 × 0.5047

= $ 9.3982

PE,t = 100e−.10×0.5 × 0.4953 − 100e−.05×0.5 × 0.4114

= $ 6.9902

1.1.2 Applying the Binomial Model for a Given Value of Sigma

The Binomial model can be used to value options on a stock paying a continuous
dividend yield. The parameters are typically defined as follows.
√
u = eσ 4t (13.14)
√
d = e−σ 4t (13.15)

The risk neutral probabilities can be derived as follows. Consider a portfolio

consisting of α units of the stock and $ B of riskless debt. Assume that the stock
pays a dividend at a rate of δ% per annum and that the dividends are re-invested
in the stock. Let us choose α and B such that:

αuSteδ4t + Ber4t = Cu (13.16)

and αdSt eδ4t + Ber4t = Cd (13.17)

Cu − Cd −δ4t
So α = e (13.18)
St (u − d)
and B = [Cu − αuSt eδ4t ]e−r4t
u(Cu − Cd ) −r4t
= [Cu − ]e (13.19)
u−d
αSt + B = Ct . Therefore
Cu − Cd −δ4t u(Cu − Cd ) −r4t
e + [Cu − ]e = Ct
(u − d) u−d

661
 Cu − Cd (r−δ)4t u(Cu − Cd )
⇒ e + [Cu − ] = Ct er4t
(u − d) u−d
e(r−δ)4t − d u − e(r−δ)4t
" # " #
⇒ Cu + Cd = Ct er4t
u−d u−d
e(r−δ)4t − d u − e(r−δ)4t
( " # " #)
⇒ Ct = Cu + Cd e−r4t (13.20)
u−d u−d
e(r−δ)4t − d
" #
This is of the form Ct = [pCu + (1 − p)Cd ]e−r4t where p = .
u−d

1.2 Arbitrage Restrictions

2 Lower Bound for European Call Options

h i
It can be demonstrated that CE,t ≥ Max 0, St e−δ(T −t) − Xe−r(T −t) . If the ex-
pression St e−δ(T −t) − Xe−r(T −t) were to be less than zero, then all that we can
assert is that CE,t ≥ 0, because we know that an option cannot have a negative
premium. However, if the expression were to be positive, then the option premium
must be greater than or equal to it to preclude arbitrage.

2.1 Proof

Assume that CE,t < St e−δ(T −t) − Xe−r(T −t) > 0 or that St e−δ(T −t) − Xe−r(T −t) −
Ct > 0. Consider the following table.

662
 Table 13.1
Lower Bound for a European Call Option
Action Initial Terminal
Cash Flow Cash Flow
If ST > X If ST < X
Short sell e−δ(T −t) units of St e−δ(T −t) -ST -ST
the Stock
Buy the Call -CE,t (ST − X) 0
−r(T −t)
Lend the P.V. of X -Xe X X
Total St e−δ(T −t) − Xe−r(T −t) 0 X − ST > 0
−CE,t

The initial strategy entails the short sale of e−δ(T −t) units of the stock. If the
lender had not parted with the stock he could have reinvested the dividends in the
stock itself, and consequently would have had one unit of the stock at the time of
expiration of the option. Consequently the short seller must return one share of
the stock to cover his short position, which explains the outflow of ST at time T.

In this table the cash flow at inception is positive by assumption. The cash
flows at expiration are non-negative. Thus this table reflects an arbitrage oppor-
tunity. To preclude it, it therefore must be the case that

St e−δ(T −t) − Xe−r(T −t) − CE,t ≤ 0

⇒ CE,t ≥ St e−δ(T −t) − Xe−r(T −t) (13.21)

3 Lower Bound for European Put Options

In the case of European puts, the no-arbitrage condition is:

PE,t ≥ Max[0, Xe−r(T −t) − Ste−δ(T −t)]

This condition can be interpreted as follows. If Xe−r(T −t) − St e−δ(T −t) < 0, then
all that we can assert is that PE,t > 0, because an option cannot have a negative
premium. However, if Xe−r(T −t) − St e−δ(T −t) > 0 then we can show that PE,t ≥
Xe−r(T −t) − St e−δ(T −t). To prove it, consider the following strategy.
663
 Table 13.2
Lower Bound for a European Put Option
Action Initial Terminal
Cash Flow Cash Flow
If ST > X If ST < X
Buy the Put -PE,t 0 (X − ST )
Buy the e−δ(T −t) units of -St e−δ(T −t) ST ST
the Stock
Borrow the P.V. of X Xe−r(T −t) -X -X
Total Xe−r(T −t) − PE,t − St e−δ(T −t) ST − X > 0 0

To rule out arbitrage we require that

Xe−r(T −t) − PE,t − Ste−δ(T −t) ≤ 0

⇒ PE,t ≥ Xe−r(T −t) − Ste−δ(T −t) (13.22)

3.1 Put Call Parity for European Options

The put-call parity condition for European options on a stock paying a continuous
dividend yield is:
CE,t − PE,t = St e−δ(T −t) − Xe−r(T −t)

The proof is given in Table 13.3.

664
 Table 13.3
Put Call Parity for European Options
Action Initial Terminal
Cash Flow Cash Flow
If ST > X If ST < X
Sell the Call CE,t -(ST − X) 0
Buy the Put -PE,t 0 (X − ST )
−δ(T −t) −δ(T −t)
Buy e units -St e ST ST
of the Stock
Borrow the P.V. of X Xe−r(T −t) -X -X
Total CE,t + Xe−r(T −t) − PE,t − Ste−δ(T −t) 0 0

The initial cash flow has to be less than or equal to zero to preclude arbitrage.
However, if it were to be less, then we could reverse the above strategy and make
arbitrage profits. Consequently to rule out any form of arbitrage, the initial cash
flow must be zero. This implies that

CE,t − PE,t = St e−δ(T −t) − Xe−r(T −t) (13.23)

3.2 The Greeks for European Options on a Stock Paying

a Continuous Dividend Yield

3.3 Delta

For a European call: Delta = e−δ(T −t)N(d1 )

For a European put: Delta = -e−δ(T −t) N(−d1 )

3.4 Gamma
e−δ(T −t)n(d1 )
Gamma for both calls and puts is given by: √
St σ T − t

665
3.5 Vega
√
Vega for both calls and puts is given by: e−δ(T −t)St n(d1 ) T − t

3.6 Theta

For a European call:

e−δ(T −t)St σn(d1)
Theta = δe−δ(T −t)St N(d1 ) − rXe−r(T −t) N(d2 ) − √
2 T −t
For a European put:
e−δ(T −t)St σn(d1 )
Theta = −δe−δ(T −t)St N(−d1 ) + rXe−r(T −t) N(−d2 ) − √
2 T −t

3.7 Rho

For a European call: Rho = e−r(T −t)X(T − t)N(d2 )

For a European put: Rho = -e−r(T −t)X(T − t)N(−d2)

4 Index Options

Options on prominent stock indices are traded in venues across the globe. In the
U.S. the CBOE offers options contracts on a number of indices. These include:

• The Dow Jones Industrial Average

• The S&P 100 Index

• The S&P 500 Index

• The NASDAQ 100 Index

Both European as well as American style options are offered on the S&P
100 index, whereas on the other indices only European options are available.
All options expire on the Saturday following the third Friday of the expiration
666
month. The contracts are cash settled and the multiplier is $ 100. Thus if a call
option is exercised, the holder will receive $ 100 × (IT − X), whereas if a put is
exercised he will receive $ 100 × (X − IT ). IT is the index level at the time of
exercise/expiration.1

4.1 Example

European call options on the S&P 100 index are available with an exercise price
of 700, and six months to expiration. The riskless rate is 6% per annum and the
volatility is 20% per annum. The dividend yield is 4% per annum and the current
level of the index is 700.

The option price can be calculated using the Merton model.

0.2 × 0.2
 
0.06 − 0.04 + 0.5
d1 = √ 2 = 0.1414
0.2 × 0.5
√
d2 = 0.1414 − 0.2 × 0.5 = 0

N(d1 ) = 0.5562 and N(d2 ) = 0.5000

CE,t = 700 × e−0.04×0.5 × 0.5562 − 700 × e−0.06×0.5 × 0.5000

= 381.6306 − 339.6559 = 41.9747

The premium per contract is 100 × 41.9747 = $ 4, 197.47.

5 Foreign Currency Options

Foreign currency options are traded on a number of exchanges. The Philadel-

phia Stock Exchange (PHLX) is the world’s leading platform for exchange traded
currency options. The exchange currently lists options contracts on 10 foreign
currencies. These currencies and the contract sizes for the respective options are
given in the table below. The options are European style. At any point in time,
1
For options on the Dow Jones index, the index level is taken to be one hundredth of the
value of the index.

667
contracts are available for the four quarterly months, March, June. September
and December, and for the two nearest calendar months. Contracts expire on
the Saturday following the third Friday of the expiration month. Exercise prices
are expressed in terms of U.S. cents per unit of foreign currency. For instance
a call option on British pounds with an exercise price of $ 195 would give the
option buyer the right to buy pounds at $ 1.95 per pound. Premiums for op-
tions are quoted in U.S. cents per unit of the underlying currency. For instance
a premium of 2.78 for a given option on British pounds is $ 0.0278 per pound.
Since each option contract is for 10,000 GBP, the total premium per contract is
0.0278 × 10, 000 = $ 278.00.

Table 13.4
Currencies and Contract Sizes
Currency Contract Size
Australian Dollar 10,000 AUD
British Pound 10,000 GBP
Canadian Dollar 10,000 CAD
Euro 10,000 EUR
Japanese Yen 1,000,000 JPY
Mexican Pesos 100,000 MXN
New Zealand Dollar 10,000 NZD
South African Rand 100,000 ZAR
Swedish Krona 100,000 SEK
Swiss Franc 10,000 CHF

All the above contracts are cash settled.

5.1 Arbitrage Restrictions

A foreign currency is like a stock paying a continuous dividend yield. The leakage
of value in this case, takes place at a rate of rf , which is the riskless rate of interest
in the foreign currency. So the lower bounds for European call and put options

668
may be expressed as:

CE,t ≥ Max[0, Ste−rf (T −t) − Xe−r(T −t) ] (13.24)

and PE,t ≥ Max[0, Xe−r(T −t) − St e−rf (T −t)] (13.25)

The put-call parity relationship for European options is given by:

CE,t − PE,t = Ste−rf (T −t) − Xe−r(T −t) (13.26)

6 The Garman Kohlhagen Model

This model is an extension of the Black-Scholes model. According to it

CE = St e−rf (T −t) N(d1 ) − Xe−r(T −t)N(d2 ) (13.27)

where
σ2
!
St
 
ln + r − rf + (T − t)
X 2
d1 = √ (13.28)
σ T −t
√
and d2 = d1 − σ T − t.

6.1 Example

Consider European calls on British pounds.

Let X = 200, and the spot rate be 185. The riskless rate of interest in the U.S
is 4.5%, while the rate in the U.K is 6%. The time to expiration is six months,
and the volatility of the spot rate is 30%.
(0.30)2
!
185
 
ln + 0.045 − 0.06 + 0.5
200 2
d1 = √
0.30 0.5
= −.2970
d2 = −.5091
N(d1 ) = 0.3832 and N(d2 ) = 0.3053
CE = 185e−0.06×0.5 × 0.3832 − 200e−0.045×0.5 × 0.3053 = 9.0953
669
6.2 The Binomial Model with Given Parameters

Let us take the case of call options on British pounds. Let St be the spot exchange
rate quoted as GBP/USD. Consider a portfolio consisting of α USD of riskless
debt in the foreign currency, and B USD of riskless debt in the domestic currency.

Let r = 1 + domestic riskless rate, and rf = 1 + foreign riskless rate.

In a given period, the spot rate can either go up to uSt, or go down to dSt .

If the spot rate were to increase, the portfolio will be worth

α
rf uSt + Br (13.29)
St
while, if it were to decline, it will be worth
α
rf dSt + Br (13.30)
St
Let us choose α and B such that

αurf + Br = Cu (13.31)

and
αdrf + Br = Cd (13.32)

Therefore
Cu − Cd
α= (13.33)
rf (u − d)
and " #
1 uCd − dCu
B= (13.34)
r u−d
Thus
pCu + (1 − p)Cd
Ct = α + B = (13.35)
r
where
r
−d
rf
p= (13.36)
u−d
and
r
u−
rf
(1 − p) = (13.37)
u−d
670
6.3 Example

We will now compute the price of a call option on British pounds using a two
period Binomial model. Let u = 1.2, d = .8, St = 185, r = 1.045, rf = 1.06, and
X = 200.

The spot exchange rate tree, may be depicted as follows.

266.4





222

HH
 H
H
 H
 HH
 
185 H177.6
H 
H 
H 
HH 
H 

148
HH
H
HH
H
H
HH
H118.4

Figure 13.1

Cuu,T = 66.40, Cud,T = 0, Cdd,T = 0

1.045
− 0.8
p= 1.06 = 0.4646 and (1 − p) = 0.5354
0.4
0.4646 × 66.4 + 0.5354 × 0
Cu,T −1 = = 29.5210
1.045
The intrinsic value at this node is 22.
0.4646 × 0 + 0.5354 × 0
Cd,(T −1) = =0
1.045
The intrinsic value at this node is zero.
0.4646 × 29.5210 + 0.5354 × 0
Ct = = 13.1248
1.045
671
The intrinsic value at this node is zero. Thus the call will be worth 13.1248 cents.

6.3.1 Applying the Binomial Model for Given Value of Sigma

The formula for the option price using continuously compounded rates of interest
r and rf and the volatility of the exchange rate σ is similar to what we derived for
a stock paying a continuous dividend yield. The only difference is that we replace
δ with the riskless rate in the foreign currency rf .

6.4 The Greeks

The greeks for foreign currency options are similar to those for a stock paying
a continuous dividend yield. We have to once again replace δ with the foreign
riskless rate rf . In the case of Rho however we get two expressions, one for the
partial derivative with respect to the domestic interest rate, ρr , and the other for
the partial derivative with respect to the foreign riskless rate, ρrf .

For a European call:

ρr = e−r(T −t)X(T − t)N(d2 ) and ρrf = −e−rf (T −t)St (T − t)N(d1 ) (13.38)

For a European put:

ρr = −e−r(T −t)X(T − t)N(−d2 ) and ρrf = e−rf (T −t)St (T − t)N(−d1 ) (13.39)

7 Futures Options

A futures option is an option on a futures contract. Such options are available

on a number of successful futures contracts in many of the leading exchanges.
Examples include options on index futures, interest rate futures, FOREX futures,
and commodity futures.

672
7.1 Call Options

A call futures option gives the buyer the right to assume a long position in a futures
contract. If and when the call holder decides to exercise, a long position will be
established for him in the futures contract, and the contract will be immediately
marked to market. Once the long position is entered into, the investor must
either deposit adequate funds to meet the initial margin requirement of the futures
contract, or else he must offset the futures position.

When the call holder exercises, a short position in the futures contract is
established for the writer of the call. The writer’s position will also be marked
to market, and he must also deposit adequate margin for the futures position, or
else, liquidate it.

7.1.1 Example

Consider a person who buys a call futures option on crude oil. Each contract is
for 1,000 barrels of oil. We will assume that the current futures price of oil is
$ 75.00 per barrel, and that the exercise price of the call is $ 72.50 per barrel.

If the call holder exercises, he will get a long position in the futures contract.
At the same time, due to marking to market, he will receive an inflow of (75.00 −
72.50) × 1, 000 = $ 2, 500. How can we explain this inflow of $ 2,500? The futures
option contract gives the holder the right to go long in a futures contract at a
price of $ 72.50. If he exercises, and a long position is established for him, it will
be at the prevailing futures price of $ 75.00. To ensure that he effectively gets a
long position at $ 72.50, the difference between $ 75.00 and $ 72.50 must be paid
to him. Notice, that if Ft is the current futures price, and X the exercise price,
the call holder will exercise only if Ft > X. When he exercises, he will get an
inflow equal to $ (Ft − X), which is the intrinsic value of the option.

673
7.2 Put Options

A put futures option gives the holder the right to go short in the futures contract
at the exercise price. If the holder exercises, a long position in the futures contract
will be established for the writer. Both the holder’s and the writer’s positions will
be marked to market when the option is exercised, and they will both have to post
the required margin for the futures contract, or else, liquidate their positions.

7.2.1 Example

Consider a person who owns a put futures option on crude oil. The exercise price
is $ 75.00 per barrel, and the current futures price of oil is $ 72.00 per barrel.

If the holder chooses to exercise, he will get a short position in the futures
contract. At the same time, due to marking to market, he will receive a cash
inflow of (75.00 − 72.00) × 1, 000 = $ 3, 000. The rationale for this inflow should
be obvious from the logic presented for call futures options. Notice, that if Ft is
the current futures price, and X the exercise price, the call holder will exercise
only if Ft < X. When he exercises, he will get an inflow equal to $ (X − Ft),
which is the intrinsic value of the option.

8 Arbitrage Restrictions

8.1 Lower Bounds for European Call Futures Options

It can be demonstrated that CE,t ≥ Max[0, (Ft − X)e−r(T −t)]. As before we know
that CE,t ≥ 0 and consequently that explains the first part of the expression.
However, if (Ft −X)e−r(T −t) ≥ 0 then we can demonstrate that the option premium
must be greater in order to preclude arbitrage.

674
8.1.1 Proof

Assume that CE,t < (Ft − X)e−r(T −t) > 0 or that (Ft − X)e−r(T −t) − CE,t > 0.
Consider the following table.

Table 13.5
Lower Bound for European Calls
Action Initial Terminal
Cash Flow Cash Flow
FT > X FT ≤ X
Buy a call −CE,t FT − X 0
Sell a futures contract 0 Ft − FT Ft − FT
−r(T −t)
Borrow PV of (Ft − X) (Ft − X)e −(Ft − X) −(Ft − X)
Total −CE,t + (Ft − X)e−r(T −t) 0 X − FT

In this table the cash flow at inception is positive by assumption and the sub-
sequent cash flows are non-negative. Consequently this table reflects an arbitrage
opportunity. Therefore, in order to preclude arbitrage it must be the case that

−CE,t + (Ft − X)e−r(T −t) ≤ 0

⇒ CE,t ≥ (Ft − X)e−r(T −t) (13.40)

We also know that CE,t must be greater than zero. Therefore, the lower bound is

CE.t ≥ Max[0, (Ft − X)e−r(T −t) ] (13.41)

The lower bound for a European call on the underlying asset, with the same
expiration date is
CE,t ≥ Max[0, St − Xe−r(T −t)] (13.42)
For an asset which does not make any payouts, Ft = St er(T −t). Therefore, the
lower bound for a European futures call option may be written as

CE,t ≥ Max[0, St − Xe−r(T −t)] (13.43)

Thus, the two lower bounds are equivalent. Hence, if the futures option, and the
futures contract expire at the same time, a European call on the futures contract
675
is equivalent to a European call on the spot commodity. The logic is that, at
expiration, ST = FT . Hence Max[0, ST − X] = Max[0, FT − X]. If the payoffs
from the two options are the same, the options must be equivalent.

What about American call futures options? An American option must always
be worth at least its intrinsic value. So for an American futures option,

CA ≥ Max[0, Ft − X] (13.44)

This is a tighter lower bound because (Ft − X) > (Ft − X)e−r(T −t). Hence an
American option will be priced higher than a corresponding European option, and
may be exercised early. Remember that for a stock which makes no payouts, we
showed that American call options will never be exercised early. American calls
on futures contracts are clearly different.

8.2 Lower Bound for European Put Futures Options

In the case of European puts, the no-arbitrage condition is:

PE,t ≥ Max[0, (X − Ft)e−r(T −t)] (13.45)

This condition has the usual interpretation. If (X − Ft)e−r(T −t) < 0, then all that
we can assert is that PE,t > 0. However, if the expression is positive then we can
demonstrate that PE,t ≥ (X − Ft)e−r(T −t). To prove it, consider the following
strategy.

676
8.2.1 Proof

Table 13.6
Lower Bound for European Puts
Action Initial Terminal
Cash Flow Cash Flow
FT ≥ X FT < X
Buy a put −PE,t 0 X − FT
Buy a futures contract 0 FT − Ft FT − Ft
−r(T −t)
Lend PV of (Ft − X) −(Ft − X)e (Ft − X) (Ft − X)
Total −PE,t − (Ft − X)e−r(T −t) FT − X 0

To preclude arbitrage, we require that

−PE,t − (Ft − X)e−r(T −t) ≤ 0

⇒ PE,t ≥ (X − Ft)e−r(T −t)

Thus, we can assert that

PE,t ≥ Max[0, (X − Ft)e−r(T −t)]

For an asset which does not make any payouts, Ft = St er(T −t). Therefore, the
lower bound for a European futures put option may be written as

PE,t ≥ Max[0, Xe−r(T −t) − St] (13.46)

Thus the lower bound is equivalent to what we derived for a put option on the
underlying asset. Hence, if the futures option, and the futures contract expire at
the same time, a European put on the futures contract is equivalent to a European
put on the spot commodity. The intrinsic value of an American put futures option
is Max(0, X − Ft), which provides a tighter lower bound. Hence American put
futures options may be exercised early.

8.3 Put-Call Parity

The put-call parity condition for European options on a futures contract is:

CE,t − PE,t = (Ft − X)e−r(T −t)

677
The proof is given in the table below.

Table 13.7
Put Call Parity
Action Initial Terminal
Cash Flow Cash Flow
FT ≥ X FT < X
Sell a call CE,t −(FT − X) 0
Buy a put −PE,t 0 (X − FT )
Buy futures 0 FT − Ft FT − Ft
Lend PV of (Ft − X) −(Ft − X)e−r(T −t) (Ft − X) (Ft − X)
Total CE,t − PE,t − (Ft − X)e−r(T −t) 0 0

So to preclude arbitrage, we require that

CE,t = PE,t + (Ft − X)e−r(T −t) (13.47)

9 The Black Model

This model is a variation of the Black-Scholes model and is applicable for pric-
ing European options on futures contracts. Assuming that the options and the
underlying futures expire at the same time, Black showed that

CE,t = e−r(T −t)[FtN(d1 ) − XN(d2 )] (13.48)

and PE = e−r(T −t)[XN(−d2 ) − FtN(−d1 )] (13.49)

where
σ2
!
F
 
ln + T
X 2
d1 = √ (13.50)
σ T −t
√
and d2 = d1 − σ T − t

678
9.1 Example

The price of a six months futures contract on crude oil is $ 75. Options are
available with an exercise price of $ 70 and six months to expiration. The riskless
rate is 8% per annum and the volatility is 30% per annum.
75 0.30 × 0.30
   
ln + 0.5
d1 = 70 √ 2
0.30 0.5
0.0915
= = 0.4314
0.2121
d2 = 0.4314 − .2121 = 0.2193
N(0.4314) = 0.6669; N(0.2193) = 0.5868

CE,t = e−0.08×.5[75 × 0.6669 − 70 × 0.5868] = 8.5909

PE,t = e−0.08×.5[70 × 0.4132 − 75 × 0.3331] = 3.7870

9.2 The Binomial Model

We will now see as to how the binomial model can be used to price futures options.
To start with we will focus on the one period model. Let the stock price follow a
binomial model as described below.

uS
 t










S
t
H
HH
H
H
HH
H
H
HH
H
H
HH
HdSt
Figure 13.2

From the cost of carry model, the futures price at any node is given by F =
Sr(T −t), where r is 1 + riskless rate and T-t is the number of periods left till the
maturity of the futures contract.
679
 Hence, the futures price can also be modeled using a binomial process as shown
below.

uSt










 = rSt
F
Ht
HH
H
H
HH
H
H
HH
H
H
HH
HdSt
Figure 13.3

Let us form a hedge portfolio consisting of a long position in α futures con-

tracts, and $ B in riskless debt. The initial investment = $ B, because it costs
nothing to get into the futures contract. If this portfolio is to replicate the call,
we require that
u
 
α Ft − Ft + Br = Cu (13.51)
r
!
d
and α Ft − Ft + Br = Cd (13.52)
r
Therefore
Cu − Cd
α= ! (13.53)
u d
Ft −
r r
u d
Let u∗ = and d∗ = . Therefore
r r
Cu − Cd
α= (13.54)
Ft(u∗ − d∗ )
Substituting for α we get

1 1 − d∗ u∗ − 1
" #
B= C u + Cd (13.55)
r u∗ − d∗ u∗ − d∗

d
1 − d∗ 1−
p= ∗ = r = r−d (13.56)
u −d ∗ u d u−d
−
r r
680
Thus p and (1-p) can be calculated using either u∗ and d∗ or u and d. Therefore

pCu + (1 − p)Cd
B= (13.57)
r
Since the payoffs from the portfolio are the same as that from the call, Ct = B

9.2.1 Example

A stock is currently valued at $ 100. Every period the stock price may go up by
20% or go down by 20%. The riskless rate of interest is 5% per period. What is
the value of a call futures option with two periods to expiration?

The stock price tree may be depicted as follows.

T-2 T-1 T
144


 

120

HH
 H
  H
 HH
100

HH
H96

H 
H
HH 

80
H
HH
H
H
HH
H64

Figure 13.4

The futures price tree can be depicted as follows for a contract with two periods
to expiration. As you can see, at expiration, the futures price is equal to the spot
price.

681
 T-2 T-1 T
144




126

HH
 H
 H
H

110.25

H
HH96
HH 

H
H 
HH84

H
HH
H
H
HH64

Figure 13.5

If we assume that the futures contract and the option on the futures contract
expire at the same time, then

Cuu,T = 44, Cud,T = 0, and Cdd,T = 0; p = 0.625 and 1 − p = 0.375

0.625 × 44 + 0.375 × 0
Cu,T −1 = = 26.1905
1.05
If we are dealing with American options, we have to check and see as to whether
this is less than the intrinsic value. The I.V at this node is 26, which is less. So
the option will not be exercised early.
0.625 × 0 + 0.375 × 0
Cd,T −1 = =0
1.05
The intrinsic value at this node is zero since the option is out of the money.
Therefore
0.625 × 26.1905 + 0.375 × 0
Ct = = 15.5896
1.05
Once again, we have to check for early exercise. The intrinsic value at ‘T-2’ is
10.25, so the option will not be exercised early. Thus the value of the American
call futures option = $ 15.5896.

682
9.3 Applying the Binomial Model for a Given Value of
Sigma

The binomial model can be used to value futures options using the following
parameters. √
σ 4t
u=e (13.58)
√
d = e−σ 4t
(13.59)
1−d
p= (13.60)
u−d

10 Options on Futures versus Options on the

Underlying

Futures options are often more actively traded than options on the underlying
asset. One of the main reasons is that the underlying asset in the case of a futures
option, namely the futures contract, is more liquid and consequently its prices
are more reliable. In practice futures prices can be easily ascertained from the
futures exchange, while the price of the underlying commodity may not be so
transparent. This is particularly true for underlying assets like Treasury bonds
which are largely traded over-the-counter.2

The other reason for the popularity of futures options is that the holder is
not required to take possession of the underlying asset upon exercise, which is
normally the case for options on the underlying asset. Since the exercise of a
futures option leads to the establishment of a futures position for the holder, he
has the option to offset and exit the market without taking delivery.

We have demonstrated earlier that a European futures option, whether a call

or a put, must be worth the same as a European option on the underlying asset
with the same time to maturity. However, if the European futures option were
to expire before the expiration date of the underlying futures contract, then this
would no longer be the case. In such a situation, the futures call will be worth more
2
See Hull(2006).

683
than a call on the underlying asset if the market is in contango, whereas it will be
worth less than a call on the underlying asset if the market is in backwardation.
For puts however, the case is the opposite. Puts on futures will be worth less than
puts on the underlying if the futures market is in contango, whereas they will be
worth more than puts on the underlying if the market is in backwardation. The
same is the case for American futures options, irrespective of whether the futures
contract expires at the same time as the option or after it.3

11 Portfolio Insurance

The term portfolio insurance refers to a strategy for protecting the principal value
of a portfolio against a decline in the price. The strategy entails the acquisition
of a put option, such that the floor value of the portfolio at the time of expiration
of the put is the principal value of the portfolio.

This kind of a protective put strategy may be accomplished by acquiring an

exchange traded put option, or by replicating a put option using positions in the
underlying asset and the riskless asset, as we have demonstrated in our discussion
of the Binomial model in the previous chapter. The first type of insurance may
be termed as static portfolio insurance while the second may be considered as
dynamic portfolio insurance.4

11.1 Dynamic Option Replication

We will assume in this illustration that we have more capital at the outset than
our target insurance floor. More specifically, we are going to assume we have
enough capital buy one unit of the underlying asset plus one put option on it.

Consider an asset that is currently valued at $ 100. We will assume that every
period the stock price may go up by 20% or go down by 20%. The riskless rate of
interest is 5% per period. The evolution of the underlying asset over three periods
3
See Hull(2006).
4
See Stoll and Whaley (1993).

684
may be depicted as follows.
T-3 T-2 T-1 T


172.8




144

H
 H
HH

 H
120
  HH115.2
H 
 H
HH 
 
 H 
 
100 96
HH 
H H
H
HH  H
HH

H  H
80
HH  HH76.8
H 
H 
HH 
H 
64
HH 
H
H
HH
H
HH51.2

Figure 13.6

Let us now consider a European put with an exercise price of $ 100. The value
of the put at each node can be computed using the binomial model. The risk
neutral probabilities are:
1.05 − 0.80
p= = 0.625 and 1 − p = 0.375
1.20 − 0.80
Puuu = 0; Puud = 0; Pudd = 23.20; Pddd = 48.80

Puu is obviously zero. It can be shown that Pud = 8.29, Pdd = 31.24, Pu = 2.96,
Pd = 16.09,and Pt = 7.51. The evolution of the put price can consequently be
depicted as follows.

685
 T-3 T-2 T-1 T


0



0
 HH
 H
 H
H

2.96

H
H0

 HH 
 H
H 
 
7.51

H
H 8.29

 HH
HH  H
H
H  H
H

H 16.09 H 23.20

H 
HH 
H
H 

H 31.24

H
HH
H
H
H 48.80

Figure 13.7

So the insured portfolio consisting of one unit of the underlying asset and a
put option follows the following price process.
T-3 T-2 T-1 T


172.8



144

H H
 H
HH


122.96

 HH
H 115.2

 H
HH 
 
 
107.51

H
H 104.29

 HH
H
H
HH  H
HH


H 96.09 H 100.00

H
H 
H
HH 


H 95.24

H
H
H
HH
H 100.00

Figure 13.8

As we have already seen from the study of the binomial model, a put option

686
can be replicated by a combination of positions in the underlying asset and the
riskless asset. We will denote the respective positions by α and B. For instance
Pu − Pd 2.96 − 16.09
αT −3 = = = −0.32825
ST −3 × (u − d) 100 × (1.20 − 0.80)

uPd − dPu 1.20 × 16.09 − 0.8 × 2.96

BT −3 = = = $ 40.3333
r × (u − d) 1.05 × 0.40
The composition of the replicating portfolios at the other nodes can be derived in
a similar fashion. The insured portfolio consists of one unit of the underlying asset
and a put option. Consequently, to replicate the insured portfolio, we need to start
with a portfolio consisting of 1.0 − 0.32825 = 0.67175 units of the underlying asset
and $ 40.3333 of riskless debt. The composition of such a replicating portfolio at
each node is depicted below.
T-3 T-2 T-1

α = 1.0000
B = $ 0.0000
α = 0.8273
B = $ 23.6857
α = 0.67175 α = 0.3958
B = $ 40.3333 B = $ 66.2857
α = 0.2828
B = $ 73.4667
α = 0.0000
B = $ 95.2381

Figure 13.9

As we have demonstrated in the previous chapter, this dynamic replication

process is self-financing at each node.

687
11.2 The Cost of Insurance

One way of looking at the cost of insurance is by considering the price of the
implied put option in the dynamic replication strategy. In our example we had
to take out $ 7.51 out of our initial capital of $ 107.51 to insure the portfolio at
a floor value of $ 100.

Another way of looking at the cost of insurance is by considering the return

foregone on the upside. In our case, by creating a put worth $ 7.51, the manager
100
is able to devote only ≡ 93% of the total capital to the underlying asset.
107.51
On the other hand, if he had not insured he would have been able to devote the
entire $ 107.51 to the underlying asset. The consequence is that, in the two
states where the insurance is not required, namely ST = 172.80 and ST = 115.20,
the insured portfolio is able to capture only 93% of the value that it could have
captured had the insurance not been availed off.

11.3 Portfolio Insurance and the Black-Scholes Model

From put-call parity we know that:

X
CE,t − PE,t = St −
(1 + r)T −t
X
⇒ St + PE,t = CE,t +
(1 + r)T −t
X X
= St N(d1 ) − T −t N(d2 ) +
(1 + r) (1 + r)T −t
X
= StN(d1 ) + N(−d2 ) (13.61)
(1 + r)T −t
Thus in the Black-Scholes world, an insured portfolio can be created by investing
in N(d1 ) shares and investing an amount of N(−d2 ) times the present value of
the exercise price in the riskless asset. Thus, in this scenario, N(d1 ) or the delta
of the call option plays the role of the hedge ratio for the insured portfolio.

From put-call parity

X
⇒ St + PE,t = CE,t + (13.62)
(1 + r)T −t
688
Thus a portfolio consisting of a call option with an exercise price of $ 100 plus
an investment equal to the present value of the exercise price in the riskless asset,
will behave in the same fashion as our insured portfolio consisting of positions
in a share plus a put option. Thus this combination offers an alternative way of
creating our insured portfolio.

12 Options on Volatility

The CBOE offers options on volatility computed using data for S&P500 index
options. These trade under the symbol VIX where VIX is an acronym for volatility
index. It is an estimate of expected volatility computed using the bid and ask
quotes for options on the S&P500 index. For the purpose of computation, data is
used for the nearby and second nearby options contracts with at least eight days
left to expiration. The data spans a range of exercise prices. The computation is
independent of any option pricing model.

The significance of VIX may be explained as follows. Option values are influ-
enced by a number of variables, of which volatility is one factor. VIX isolates the
expected volatility from the other factors, and allows investors to directly take a
position on the volatility.

The contract multiplier is $ 100, and the expiration date is the Wednesday
thirty days prior to the third Friday of the calendar month immediately following
the expiration month. The exercise style is European. Generally up to three
near-term contract months plus up to three additional months from the February
cycle are listed.

13 SPAN

SPAN is an acronym for Standard Portfolio Analysis of Risk. As the name sug-
gests it is a portfolio based margining system, and was developed by the Chicago
Mercantile Exchange (CME) in 1988. SPAN attempts to identify the overall risk
of a portfolio of derivatives on a given underlying asset.

689
 SPAN is extremely simple to implement in practice. The complex aspects such
as option price calculations are performed by the exchanges and clearinghouses
that use SPAN. The output of these calculations is known as a Risk Array. The
risk array and other required inputs for margin computation are then packaged in
a file called the SPAN risk parameter file. This process is referred to as the SPAN
front end. The clearing firms and other end users of SPAN use the data contained
in these risk parameter files in conjunction with their portfolios containing posi-
tions in various derivative securities on a given underlying asset, to determine the
required margin. This entails only simple arithmetic calculations, and is referred
to as the SPAN back end. Thus the end users are not required to concern them-
selves with the complex aspects of portfolio margin computation such as option
valuation. This simplicity is largely responsible for the wide acceptance of SPAN.
One important aspect of SPAN is that since the valuation and revaluation of the
underlying derivatives is undertaken by the exchanges and clearinghouses, they
are in effect able to ensure that the risk parameters being used are the same for
all end users and reflect their margin policies.

The objective of SPAN is to identify the overall risk of a portfolio of derivative

securities. Thus it simultaneously evaluates all derivatives with a common under-
lying asset, be they options, futures, or futures options. These derivatives which
are defined on a common underlying security are said in the parlance of SPAN
to belong to a single combined commodity. SPAN also takes cognizance of both
inter-commodity and inter-month risk relationships. These concepts will become
clear as we proceed.

At the core of the SPAN system is what is called the risk array. The risk
array for a derivative security represents the change in its value over a specified
period of time called the look-ahead time, under various scenarios that represent
changing market conditions. In practice the look-ahead period is usually taken to
be one trading day. The various market scenarios that are considered are called
risk scenarios. Each scenario is defined in terms of how much the price of the
underlying asset is expected to change from its current value over the look-ahead
period, as well as how much the volatility of the rate of return on the underlying
asset is expected to change in the same period. The change in value for a derivative
security for each risk scenario is termed as the risk array value for that particular

690
risk scenario. The vector of risk array values for a derivative under the full set of
‘what-if scenarios’ (risk scenarios) is termed as the risk array for that security.

Risk array values are calculated for a long position in one unit of the derivative
security. Since SPAN is more concerned with potential losses rather than potential
gains, losses are represented as positive values, whereas gains are depicted as
negative values.

SPAN evaluates the change in value for a security over sixteen risk scenarios.
Every scenario is specified in terms of two parameters called the price scan range
and the volatility scan range. The price scan range is a specified value for the
potential change in the price of the underlying asset over the look-ahead period,
while the volatility scan range is a specified value for the potential change in the
volatility of the rate of return on the underlying asset over the same period. Let
us denote the price scan range as 4P and the volatility scan range as 4σ. The
sixteen scenarios are defined in Table 13.8.

SPAN endeavours to determine the largest loss that a portfolio may be reason-
ably expected to suffer over the look-ahead period. The exchanges and clearing-
houses that use SPAN have to define what they consider to be a reasonable loss
over this period. For instance, an exchange may define the loss as a value that is
unlikely to be exceeded over 99% of the trading days. It will then set the price
scan range and the volatility scan range accordingly.

691
 Table 13.8
Definition of Risk Scenarios
Scenario No. Change in Price Change in Volatility
1 0 +4σ
2 0 −4σ
4P
3 + +4σ
3
4P
4 + −4σ
3
4P
5 − +4σ
3
4P
6 − −4σ
3
24P
7 + +4σ
3
24P
8 + −4σ
3
24P
9 − +4σ
3
24P
10 − −4σ
3
11 +4P +4σ
12 +4P −4σ
13 −4P +4σ
14 −4P −4σ
15 +24P 0
16 −24P 0

13.1 Composite Delta

Delta which is the rate of change of the value of the derivative with respect to
the price of the underlying asset is required by SPAN to form spreads. Delta for
a long futures position is always 1.0, whereas for options it will range from -1.0
to +1.0. For options, as we have seen earlier, delta is not a constant and is a
function of the price of the underlying asset. SPAN calculates one delta value for
the entire contract, called the Composite Delta. It is derived as a weighted average
of the deltas associated with each of the price scan points. The weights associated
with a price scan point is based on the probability of the price movement. Thus,

692
more likely price changes receive higher weights.5 The composite delta may be
perceived as the best estimate of what the delta for the contract will be at the
end of the look-ahead period. The value that is computed in this fashion, is a
constituent of the risk parameter file that an exchange/clearinghouse transmits to
its members.

13.2 Illustration

We will illustrate the mechanics of SPAN using futures and futures options on
crude oil. First let us take the case of futures contracts. Let the current futures
price be $ 75. We will set the price scan range, 4P , as $ 15, and the volatility scan
range as 2.5%. The volatility is of no consequence while valuing futures contracts,
but will obviously have a role to play in the valuation of futures options. Each
crude oil contract is for 1,000 barrels of oil. The gains and losses under the various
risk scenarios are depicted below.
5
See www.cme.com

693
 Table 13.9
Risk Array for a Futures Contract
Scenario No. Change in Price Change in Gain/Loss Gain/Loss
Volatility per Barrel per Contract
1 0 +2.5% 0 0
2 0 -2.5% 0 0
3 +$ 5 +2.5% -$ 5 -$ 5,000
4 +$ 5 -2.5% -$ 5 -$ 5,000
5 -$ 5 +2.5% +$ 5 +$ 5,000
6 -$ 5 -2.5% +$ 5 +$ 5,000
7 +$ 10 +2.5% -$ 10 -$ 10,000
8 +$ 10 -2.5% -$ 10 -$ 10,000
9 -$ 10 +2.5% +$ 10 +$ 10,000
10 -$ 10 -2.5% +$ 10 + $ 10,000
11 +$ 15 +2.5% -$ 15 -$ 15,000
12 +$ 15 -2.5% -$ 15 -$ 15,000
13 -$ 15 +2.5% +$ 15 +$ 15,000
14 -$ 15 -2.5% +$ 15 +$ 15,000
15 +$ 30 0 -$ 10.5 -$ 10,500
16 -$ 30 0 +$ 10.5 +$ 10,500

Let us analyze the above table. The gain/loss for a long futures position over
a one day horizon is Ft+1 − Ft. However since SPAN considers the profits to be
negative numbers and losses to be positive, we define the gain/loss per barrel as
−[Ft+1 − Ft]. The gain/loss per contract is obviously −[Ft+1 − Ft] × 1, 000. This
explains the first 14 entries in the above table. The last two scenarios depict
what are called extreme value scenarios. These are specified to capture the risk
associated with deep out of the money short option positions. The price change
corresponding to these scenarios is a multiple of the price scan range. Most
exchanges take it to be twice the price scan range. That is, the price change is
considered to be ±24P , keeping the volatility constant. However, only 35% of
the change in value is considered for these extremes by most exchanges. Since the
total change in value is ±30, 000 the gain/loss is taken as -$ 10,500 in scenario 15
694
and +$ 10,500 in scenario 16.

Now we will apply the principles of SPAN to a long position in a call futures
option. Assume that the current futures price is $ 75, and that the volatility of
the rate of return is 30%. We will consider a call option with an exercise price of
$ 75 and 61 days to maturity. The risk-less rate is assumed to be 8% per annum.
Using the Black model, the theoretical options price is:

CE,t = e−r(T −t)[FtN(d1 ) − XN(d2 )]

61
We will take the time to maturity as = 0.1671. The option premium is
365
$ 3.6183 per barrel or $ 3,618.30 per contract.

The risk array is illustrated in the following table.

Table 13.10
Risk Array for a Call Futures Option
Scenario No. Change in Price Change in Gain/Loss Gain/Loss
Volatility per Barrel per Contract
1 0 +2.5% -$ 0.2702 -$ 270.19
2 0 -2.5% $ 0.3274 $ 327.36
3 +$ 5 +2.5% -$ 3.3387 -$ 3,338.74
4 +$ 5 -2.5% -$ 2.8031 -$ 2,803.10
5 -$ 5 +2.5% +$ 1.8240 +$ 1,824.01
6 -$ 5 -2.5% +$ 2.3150 +$ 2,315.03
7 +$ 10 +2.5% -$7.2008 -$ 7,200.78
8 +$ 10 -2.5% -$ 6.8270 -$ 6,827.01
9 -$ 10 +2.5% +$ 2.9786 +$ 2,978.60
10 -$ 10 -2.5% +$ 3.2563 + $ 3,256.34
11 +$ 15 +2.5% -$ 11.5913 -$ 11,591.28
12 +$ 15 -2.5% -$ 11.3788 -$ 11,378.82
13 -$ 15 +2.5% +$ 3.4564 +$ 3,456.42
14 -$ 15 -2.5% +$ 3.5559 +$ 3,555.91
15 +$ 30 0 -$ 9.1033 -$ 9,103.30
16 -$ 30 0 +$ 1.2664 +$ 1,266.40

695
 Let us analyze the first entry in the above table. The change in value for a
long position in a call option is [Ct+1 − Ct ] where Ct is the theoretical value of
the option at the time of margin computation and Ct+1 is the theoretical value
at the end of the look-ahead period for a specified set of parameters. Remember
that in this case, the option which has 61 days to expiration at the time of margin
computation will have only 60 days left at the end of the look-ahead period.
The first row in the table is for a scenario where the futures price is assumed to
remain unchanged while the volatility is expected to increase by 2.50% from 30%
to 32.50%. The value of the call in the this scenario is $ 3.8885. Thus the change
in value is
3.8885 − 3.6183 = $ 0.2702

Since SPAN considers gains to be negative, the gain/loss is defined as −[Ct+1 −Ct].
The gain/loss per contract is

−0.2702 × 1, 000 = −$ 270.20

The other entries can be interpreted in a similar fashion. Once again, for scenarios
15 and 16, only 35% of the gain/loss has been taken into account.

Let us compute the gain/loss under each scenario for a portfolio consisting of
one futures contract and a short position in four call futures option contracts.

696
 Table 13.11
Portfolio Gain/Loss
Scenario Gain/Loss Gain/Loss Portfolio
From Futures From Options Gain/Loss
Position Position
1 0 $ 1,080.75 $ 1,080.75
2 0 -$ 1,309.45 -$ 1,309.45
3 -$ 5,000 $ 13,354.97 $ 8,354.97
4 -$ 5,000 $ 11,212.41 $ 6,212.41
5 $ 5,000 -$ 7,296.04 -$ 2,296.04
6 $ 5,000 -$ 9,260.13 -$ 4,260.13
7 -$ 10,000 $ 28,803.10 $ 18,803.10
8 -$ 10,000 $ 27,308.03 $ 17,308.03
9 $ 10,000 -$ 11,914.40 -$ 1,914.40
10 $ 10,000 -$ 13,025.35 -$ 3,025.35
11 -$ 15,000 $ 46,365.11 $ 31,365.11
12 -$ 15,000 $ 45,515.30 $ 30,515.30
13 $ 15,000 -$ 13,825.66 $ 1,174.34
14 $ 15,000 -$ 14,223.62 $ 776.38
15 -$ 10,500 $ 36,413.20 $ 25,913.20
16 $ 10,500 -$ 5,065.60 $ 5,434.40

13.3 Scanning Risk Charge

The scanning risk charge for a portfolio is the largest loss that it is likely to suffer
under the 16 scenarios being considered. In our illustration (see Table 13.11) the
largest loss is $ 31,365.11 which corresponds to scenario 11.

13.4 Inter-Commodity Spread Credits

A derivatives exchange may list contracts on two different products whose prices
are correlated. Consequently, positions in the two contracts on opposite sides

697
of the market may be risk reducing because losses from one instrument may be
offset against gains from the other. In such cases, the exchange may set an inter-
commodity spread to take cognizance of the fact that the risk of the combined
position is lower than the sum of the risks for the two positions taken in isolation.
An inter-commodity spread credit, if applicable, reduces the required margin for
the combined position. Here is an illustration.6

Effect of Inter-commodity Spread

Margins as of April 12, 2001

Margin for a short position in one

S&P500 September futures contract $ 17, 250

Margin for a long position in one

NASDAQ 100 June futures contract $ 27, 000

Margin if the two positions are

treated separately $ 44, 250

Inter-commodity Spread Credit ($ 26, 993)

Margin if the position is

treated as a portfolio $ 17, 257

13.5 Intra-Commodity Spread Risk Charge

SPAN assumes that the price movements in the underlying instrument correlate
perfectly across contract months. For instance, it will assume that the price scan
range for a three month futures contract is the same as that for a six month
futures contract. However, in practice, the correlation will be less than perfect
for price movements of different contract months. In order to take this factor into
account, SPAN allows the exchanges to levy an intra-commodity spread charge.
In the absence of such a charge, a position that is long in three month futures
6
See www.cftc.gov.

698
and short in six month futures, would have a margin requirement of zero. The
intra-commodity charge is also referred to as a calendar spread charge, and is
added to the scanning risk associated with each futures and options contract. For
each contract, SPAN identifies the associated delta, and then forms spreads using
these deltas across contract months. The calendar spread charge is assessed for
each spread that is formed in this fashion.

13.6 Short Option Minimum Charge

Short positions in options contracts which are extremely deep out of the money,
may appear to have little risk across the entire scanning range. However, If
the underlying market conditions were to change significantly, such options may
move into the money, thereby generating large losses for the writer. To cover the
risk associated with such options, SPAN levies a minimum margin for each short
option position in the portfolio. This serves as a minimum charge towards margin
requirements for each short position in an options contract. The short option
minimum charge is usually applied to the greater of the short calls or short puts
in the product, for it is unlikely that both short calls as well as short puts will
lose due to a large one way price movement in the underlying asset. However,
clearing members are given the option to calculate the short option minimum for
their customers based on the total of short calls and puts.

13.7 Spot Month Add-on Charge

Most exchanges and clearinghouses that use SPAN, levy a charge to recognize the
additional risk inherent in portfolios which have positions in futures contracts that
are in their delivery month. This charge is usually only for physically deliverable
products and applies only to open positions held in their delivery month.

13.8 Computing the Final Risk Based Margin

The procedure for determining the total margin that has to be posted as per
SPAN may be defined as follows.
699
1. The scanning risk charge for the entire portfolio of derivatives has to be first
computed.

2. Intra-commodity spread risk charges and spot month add-on charges, if

applicable, have to be added to the scanning risk charge.

3. Any inter-commodity spread credits should be subtracted from the figure

arrived at in step (2).

4. The short option minimum charge has to be computed.

5. The margin as computed in step (3) has to be compared with the short
option minimum charge that is computed in step (4). The larger of the two
amounts represents the risk margin requirement as per SPAN.

700
 Suggestions for Further Reading

1. Chance D.M. An Introduction to Derivatives & Risk Management. Thomson;South-

Western, 2004.

2. Dewynne J., Howison S., and P. Wilmott The Mathematics of Financial

Derivatives: A Student Introduction. Cambridge University Press, 1999.

3. Hull J.C. Options, Futures, and Other Derivatives. Prentice Hall of India,
2005.

4. Kolb R.W. Futures, Options, & Swaps. Blackwell, 1999.

5. Stoll, H.R. and R.E. Whaley Futures and Options: Theory and Applications.
South Western, 1993.

References

• www.cboe.com

• www.cftc.gov

• www.cme.com

701
 Concept Check

State whether the following statements are True or False.

1. Gamma is identical for European calls and puts, as per the Merton model,
on a stock that gives a continuous dividend yield.

2. Vega is identical for European calls and puts, as per the Merton model, for
stocks that give a continuous dividend yield.

3. Vega is positive for European calls and puts on a stock that gives a contin-
uous dividend yield.

4. Theta may be positive for European calls on a stock that gives a continuous
dividend yield.

5. European calls on a stock that gives a continuous dividend yield may not
always be wasting assets.

6. European puts on a stock that gives a continuous dividend yield are always
wasting assets.

7. A call futures option gives the holder the right to assume a long position in
a futures contract.

8. If a call futures option is exercised it will always lead to an inflow for the
option holder.

9. If a call futures option is exercised, both the holder and the writer must
post margins or else offset their respective positions.

10. If a put futures option is exercised it will lead to an inflow for the option
writer.

11. A put futures option gives the writer the right to assume a short position in
a futures contract.

702
12. A European call option on a futures contract will be worth the same as a
European call option on the underlying asset if the futures contract and the
options expire at the same time.

13. American calls on a futures contract may be exercised early even if the
underlying asset does not make any payouts.

14. Futures options are usually more liquid than option on the underlying assets.

15. Portfolio insurance can be created using put options but not with call op-
tions.

16. The SPAN back-end requires only simple arithmetic calculations.

17. SPAN treats futures, options, and futures options on a given underlying
asset, as a single combined entity.

18. Inter-commodity spreads serve to increase the applicable margin for the
overall position.

19. The short option minimum charge is usually applied to the sum of short
calls and short puts.

20. The spot-month add-on charge is applicable to futures contracts that are in
their delivery month.

703
 Questions & Problems

Question-I

A stock is currently priced at $ 50. Call and put options with an exercise price of
$ 45 and nine months to maturity are available. The risk-less interest rate is 8%
per annum, and the volatility of the rate of return is 25% per annum. The stock
gives a continuous dividend yield of 10% per annum.

1. Compute the value of European calls and puts using the Merton model.

2. Compute the value of delta, gamma, vega, theta, and rho, for both calls and
puts.

Question-II

Call and put options on Euros are available. The spot rate is 135 and the exercise
price of the options is 150. Every period the spot rate may go up by 20% or go
down by 20%. The risk-less rate of interest is 8% per period in the U.S., and is
6% per period in Europe.

Assuming that the options are European in nature, value both calls and puts
expiring after three periods using the Binomial model.

Question-III

Futures options are often more actively traded than options on the underlying
assets. Discuss.

Question-IV

European futures options, both calls and puts, must be worth the same as options
on the underlying asset, if both the options as well as the futures contracts expire

704
at the same time. Discuss.

Question-V

What is portfolio insurance, and how can it be achieved using put options? Discuss
the difference between static and dynamic insurance.

Question-VI

What is SPAN? What the components of the margin as calculated using SPAN?
What is the concept of a short option minimum charge in SPAN?

Question-VII

The spot price of an asset is $ 100. The volatility of the rate of return is 25%
per annum, and the risk-less rate of interest is 10% per annum. Put options are
available with an exercise price of $ 105 and 91 days to maturity. Each contract
is for 5,000 units of the underlying asset.

An investor has a short position in 10 option contracts and 4 futures contracts.

The price scan range is $ 12 and the volatility scan range is 2.5%.

Compute the scanning risk charge using SPAN. For the extreme scenarios you
need consider only 35% of the gain/loss.

Question-VIII

The volatility of the rate of return is 25% per annum and the risk-less rate is 10%
per annum. The stock pays a continuous dividend yield at the rate of 7.5% per
annum. Each period is equal to 0.25 years. The current spot price is $ 100 and
the exercise price is also $ 100.

Value European call options with three periods to expiration, using the Bino-
mial model.

705
Question-IX

The price of a 6 months futures contract on gold is $ 850 per ounce. Call and put
options with 6 months to expiration are available with an exercise price of $ 870.
The risk-less rate is 10% per annum and the volatility is 40% per annum. Each
future contract is for 100 ounces.

Compute the prices of European calls and puts using the Black model.

Question-X

A stock is currently priced at $ 80. Every period the price may go up by 25% or
go down by 20%. The risk-less rate of interest is 10% per period.

Futures contracts are available with three period to expiration, and an exercise
price of $ 80. Compute the values of European call and put options with three
period to expiration, on the futures contract.

706