Chapter 7

Put-call parity estimates for American

options, bounds on option
prices,variables determining option
prices

Put-call parity

Theorem 7.1
American put-call parity estimates For a stock paying no dividends, the price of Amer-
ican call and put options, both with the same strike price K and expiry time T , satisfy

C A − P A ≤ S (0) − Ke−rT , (7.1)

C A − P A ≥ S (0) − K. (7.2)

Bounds on Option Prices

• We will assume that all the options have the same strike K and expiry time T .

• We start by noting the following obvious inequalities


0 ≤ C E ≤ C A,
(7.3)
0 ≤ P E ≤ P A.

• The option prices must be non-negative because they have non-negative payoff.

• American options should be more expensive because they give at least the same rights
as their European counterparts.

1
CHAPTER 7. PUT-CALL PARITY ESTIMATES FOR AMERICAN OPTIONS, BOUNDS ON OPTIO

Proposition 7.1
On a stock paying no dividends one has that

+
S (0) − Ke−rT =

= max 0, S(0) − Ke−rT ≤ C E < S (0) ,

+
Ke−rT − S (0) = max 0, Ke−rT − S (0) ≤ P E < Ke−rT .

Theorem 7.2
On a stock paying no dividends one has that

C E = C A. (7.4)

Remark 7.1
As C A ≥ C E and C E ≥ S(0) − Ke−rT , it follows that

C A > S(0) − K,

if r > 0. Because the price of the American option is greater than its payoff, the
option will sooner be sold than exercised at time 0. Similar inequalities hold for t < T
and one can repeat the arguments to conclude that the American option will never
be exercised prior to the expiry time. This also shows that the American option is
equivalent to the European option.

Proposition 7.2
On a stock paying no dividends one has that

+
S (0) − Ke−rT = max 0, S (0) − Ke−rT ≤ C A < S (0) , (7.5)

(K − S (0))+ = max 0, Ke−rT − S (0) ≤ P A < K. (7.6)

CHAPTER 7. PUT-CALL PARITY ESTIMATES FOR AMERICAN OPTIONS, BOUNDS ON OPTIO

Variables Determining Option Prices

Here we will study how the option prices depend on variables such the strike K, the
˙ and the expiry time T . We shall analyse option prices
current price of the underlying S (0),
as functions of one of the variables, keeping the remaining variables constant.
European options: dependence on the strike price

Proposition 7.3
If K1 < K2 , then

1. Monotonicity: C E (K1 ) > C E (K2 ) and P E (K1 ) < P E (K2 ) .

2. Lipschitz continuity:

C E (K1 ) − C E (K2 ) < e−rT (K2 − K1 ) ,

P E (K2 ) − P E (K1 ) < e−rT (K2 − K1 ) .

3. Convexity: For α ∈ (0, 1) we have

C E (αK1 + (1 − α) K2 ) ≤ αC E (K1 ) + (1 − α) C E (K2 ) ,

P E (αK1 + (1 − α) K2 ) ≤ αP E (K1 ) + (1 − α) P E (K2 ) .

European options: dependence on the underlying asset price

The current price S (0) of the underlying asset is given by the market and cannot be
changed. But we can consider an option on a portfolio of x shares, worth S = xS (0) . The
payoff of a European call with strike K on such portfolio, to be exercised at time T, will be
(xS (T ) − K)+ . We shall study the dependence of option prices on S. We will denote the
call and put prices by C E (S) and P E (S).

Proposition 7.4
If S1 < S2 , then

1. Monotonicity: C E (S1 ) < C E (S2 ) and P E (S1 ) > P E (S2 ) .

2. Lipschitz continuity:

C E (S2 ) − C E (S1 ) < S2 − S1 ,

P E (S1 ) − P E (S2 ) < S2 − S1 .

3. Convexity: For α ∈ (0, 1) we have

C E (αS1 + (1 − α) S2 ) ≤ αC E (S1 ) + (1 − α) C E (S2 ) ,

P E (αS1 + (1 − α) S2 ) ≤ αP E (S1 ) + (1 − α) P E (S2 ) .
CHAPTER 7. PUT-CALL PARITY ESTIMATES FOR AMERICAN OPTIONS, BOUNDS ON OPTIO

American options: dependence on the strike price

Proposition 7.5
If K1 < K2 , then

1. Monotonicity: C A (K1 ) > C A (K2 ) and P A (K1 ) < P A (K2 ) .

2. Lipschitz continuity:

C A (K1 ) − C A (K2 ) < (K2 − K1 ) , (7.7)

A A
P (K2 ) − P (K1 ) < (K2 − K1 ) .

3. Convexity: For α ∈ (0, 1) we have

C A (αK1 + (1 − α) K2 ) ≤ αC A (K1 ) + (1 − α) C A (K2 ) ,

P A (αK1 + (1 − α) K2 ) ≤ αP A (K1 ) + (1 − α) P A (K2 ) .

American options: dependence on the underlying asset price

As in the European case, we consider options on a portfolio of x shares worthS = xS (0).

Proposition 7.6
If S1 < S2 then

1. Monotonicity: C A (S1 ) < C A (S2 ) and P A (S1 ) > P A (S2 ) .

2. Lipschitz continuity:

C A (S2 ) − C A (S1 ) < S2 − S1 ,

P A (S1 ) − P A (S2 ) < S2 − S1 .

3. Convexity: For α ∈ (0, 1) we have

C E (αS1 + (1 − α) S2 ) ≤ αC E (S1 ) + (1 − α) C E (S2 ) ,

P E (αS1 + (1 − α) S2 ) ≤ αP E (S1 ) + (1 − α) P E (S2 ) .

American options: dependence on expiry time

Proposition 7.7

If T1 < T2 , then C A (T1 ) ≤ C A (T2 ) and P A (T1 ) ≤ P A (T2 ) .

Proof: (only for calls, the proof for puts being analogous)
Suppose that C A (T1 ) > C A (T2 ), then
CHAPTER 7. PUT-CALL PARITY ESTIMATES FOR AMERICAN OPTIONS, BOUNDS ON OPTIO

• Sell the option with shorter time to expiry and buy the one with longer time to expiry,
investing the balance risk free.

– If the option sold is exercised at time t ≤ T1 , we can exercise the other option to
cover our liability.
– The risk-less profit will be C A (T1 ) − C A (T2 ) ert > 0.

Remark 7.2
The previous arguments do not work for European options because early exercise is
not possible.

Time value of options

Definition 7.1
We say that at time 0 ≤ t ≤ T a call option with strike K is
(≫)
• (deep) in the money if S (t) > K,

• at the money if S (t) = K,

(≪)
• (deep) out of the money if S (t) < K.

The same terminology applies to put options but with the inequalities reversed.

Definition 7.2
At time 0 ≤ t ≤ T , the intrinsic value of a call (put) option with strike K is
equal to (S (t) − K)+ ((K − S (t))+ ).

Remark 7.3
The intrinsic value of out of the money or at the money options is zero. Options in
the money have positive intrinsic value. The price of an American option prior to
expiry must be greater than its intrinsic value. The price of a European option prior
to expiry may be greater or smaller than its intrinsic value.

Definition 7.3
The time value of an option is the difference between the price of the option and
its intrinsic value, that is,

C E (t) − (S (t) − K)+ , European call

E
P (t) − (K − S (t)) , +
European put
C A (t) − (S (t) − K)+ , American call
+
P A (t) − (K − S (t)) . American put.

Here, the argument t in the option prices denotes the current time and NOT the
expiry time.
CHAPTER 7. PUT-CALL PARITY ESTIMATES FOR AMERICAN OPTIONS, BOUNDS ON OPTIO

The time value of a European call as a function of S (t) is always nonnegative. For
in the money calls, the time value is bigger than K − Ke−r(T −t) , due to the inequality
C E (t) ≥ S (t) − Ke−r(T −t)
The same applies to an American call because their prices coincide.

The time value of a European put may be negative. This happens if the put option
is deep in the money, because we can only exercise the option at time T and there is a
considerable risk that in the meanwhile the stock price rises.

The time value of an American put is always nonnegative.

CHAPTER 7. PUT-CALL PARITY ESTIMATES FOR AMERICAN OPTIONS, BOUNDS ON OPTIO

Proposition 7.8
For any European or American call or put with strike price K, the time value attains
its maximum at S = K.

Proof. (only for European calls) If S ≤ K the intrinsic value is zero. Since C E (S) is an
increasing function of S, this means that the time value of the call is increasing for S ≤ K.
If K ≤ S1 < S2 , we have that C E (S2 ) − C E (S1 ) ≤ S2 − S1 and, hence,

C E (S2 ) − S2 ≤ C E (S1 ) − S1
⇕
C E (S2 ) − (S2 − K)+ = C E (S2 ) − S2 + K
≤ C E (S1 ) − S1 + K = C E (S1 ) − (S1 − K)+ ,

which yields that the time value of the call is a decreasing function of S if S ≥ K. Therefore,
the maximum is at S = K.