A No-Arbitrage Bound for European Call Prices at Different Strikes
Summary
The document proves an upper bound on the price difference between two European calls on the same asset with the same maturity, where the first strike is no greater than the second. The key idea is to view each call together with a risk-free bond whose principal equals its strike. At maturity, the combined portfolio pays the greater of the asset price and the strike.
The higher-strike portfolio has a payoff at least as large in every state, so no-arbitrage pricing requires its initial value to be at least as high. Rearranging that price comparison gives the stated bound: the call price difference cannot exceed the discounted difference between strikes. A second explanation gives the intuition as a call spread whose maximum payoff is the strike gap. The argument assumes the stated no-arbitrage setup and does not discuss complications such as transaction costs or market frictions.
Key ideas
- Compare each call with a risk-free bond paying its strike at maturity.
- The call-plus-bond portfolio pays the greater of the underlying price and the strike.
- The portfolio with the higher strike has a payoff at least as large in every outcome.
- No-arbitrage pricing bounds the call price difference by the discounted strike gap.
Tags
Full text
# European call options and strikes
# European call options and strikes
We consider 2 European call options with the same underlying asset, the same maturity date $T$ and with 2 different strikes $K_1$ and $K_2$ such that $K_1\leq K_2$. We denote $C^1_{0}$ and $C^{2}_{0}$ their respective prices. Show that under no arbitrage assumption:
$$ C^1_{0} - C^{2}_{0}\leq e^{-rT}(K_2-K_1) $$
with $r$ is the risk-free rate.
## Answer by dm63 (score 4)
https://quant.stackexchange.com/a/34650
The stated equation claims that: Value of call spread <= present value of its maximum payoff. This is almost self evident. For a formal proof, suppose untrue. Then sell the call spread, invest proceeds at r, giving a certain profit at T.
## Answer by Daneel Olivaw (score 1)
https://quant.stackexchange.com/a/34644
Write your equation as follows:
$$ C_0^1 +e^{-rT}K_1 \leq C_0^2 +e^{-rT}K_2 \quad (1)$$
Both sides consist on the price at $0$ of a portfolio $i$, $i \in \{1,2\}$, containing one option of strike $K_i$ and one risk-free zero-coupon bond of maturity $T$ and principal $K_i$. The payoff $P_i$ at $T$ of each portfolio is $\max(S_T,K_i)$ where $S_T$ is the price of the underlying stock. It comes:
$$ \begin{align} P_2 - P_1 & = \max(S_T,K_2) - \max(S_T,K_1) \\[12pt] & = (K_2-S_T)1_{\{K_2 > S_T > K_1\}} + (K_2-K_1)1_{\{K_1 \geq S_T \}} \end{align}$$
Under all scenarios payoff $P_2$ is greater than payoff $P_1$, hence by no arbitrage portfolio $2$ must have a greater price than portfolio $1$ at any time $0 \leq t \leq T$. The price of each portfolio is equal to the sum of the option price and the zero-coupon bond thus $(1)$ must be enforced, which proves your original inequality.Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.