Skip to content
QuantMax
QuantMax
  • Overview
  • Curriculum
    • FLUMental maths and numerical fluency
    • TVMTime value, rates and linear products
    • OPTOptions: fundamentals and arbitrage
      • 1Payoffs and bounds

        • Payoffs, moneyness and the bounds
      • 2Put–call parity

        • Put–call parity
      • 3Strategies

        • Option strategies: what each one is actually a bet on
      • 4Static arbitrage constraints

        • Static arbitrage: the constraints every quote must respect
      • 5Structural and corporate

        • Borrow, dividends and what happens at expiry
    • PRCOption pricing models
    • GRKThe Greeks and hedging
    • VOLVolatility
    • EXOExotics and structured products
    • SCStochastic calculus

Practise

  • Question bank
  • Mental arithmetic
  • Market simulator
  • Arbitrage trees
  • Horse racing
  • Bid book
  • Screening tests
  • Mock papers

Reference

  • Formula reference
  • Search

Your record

  • Review queue
  • Progress
  • Leaderboard
  • Profile
  • Invite friends
AccountSend feedback
  1. Curriculum
  2. /Derivatives and options
  3. /Options: fundamentals and arbitrage
  4. /Static arbitrage constraints

Static arbitrage: the constraints every quote must respect

OPT · Chapter 4·13 min read·Asked at Optiver, SIG, IMC, Akuna

Assumes Option strategies: what each one is actually a bet on.

After this lesson you should be able to

  • List the model-free constraints relating option prices across strike and expiry.
  • Spot a violated set of quotes and say what trade exploits it.
  • Explain why these are stronger statements than a model-based mispricing.

A handful of relationships between option prices must hold regardless of any model, because violating them lets someone build a portfolio with a guaranteed non-negative payoff and a negative cost. A market maker checks them continuously, and an interviewer will hand you a quote sheet with one broken in it.

ConstraintStatementThe trade if violated
Monotone in strikeC(K)C(K)C(K) falls as KKK risesBuy the cheap low strike, sell the dear high one
Call spread bound0≤C(K1)−C(K2)≤(K2−K1)e−rT0 \le C(K_1) - C(K_2) \le (K_2 - K_1)e^{-rT}0≤C(K1​)−C(K2​)≤(K2​−K1​)e−rTBuy or sell the spread against cash
Butterfly non-negativityC(K1)−2C(K2)+C(K3)≥0C(K_1) - 2C(K_2) + C(K_3) \ge 0C(K1​)−2C(K2​)+C(K3​)≥0Buy the butterfly for a credit
CalendarC(T2)≥C(T1)C(T_2) \ge C(T_1)C(T2​)≥C(T1​) for T2>T1T_2 > T_1T2​>T1​, same strikeBuy the long-dated, sell the short-dated
Put–call parityC−P=S−Ke−rTC - P = S - Ke^{-rT}C−P=S−Ke−rTConversion or reversal
Lower boundC≥max⁡(S−Ke−rT, 0)C \ge \max(S - Ke^{-rT},\, 0)C≥max(S−Ke−rT,0)Buy the call, short the stock, lend the strike
Table 4.1 · The constraints. Only the calendar row needs a caveat: it holds cleanly for European options without dividends, and a large dividend between the two expiries can invert it legitimately.

Derivation 4.2

Why prices must be convex in strike

The butterfly constraint is convexity, and convexity follows from the payoff.

  1. payoff of 12 at K1+12 at K3 ≥ payoff at K2\text{payoff of } \tfrac12\text{ at } K_1 + \tfrac12 \text{ at } K_3 \ \ge\ \text{payoff at } K_2payoff of 21​ at K1​+21​ at K3​ ≥ payoff at K2​

    With K2K_2K2​ the midpoint, the average payoff of the wings dominates pointwise.

  2. ⇒12C(K1)+12C(K3)≥C(K2)\Rightarrow \tfrac12 C(K_1) + \tfrac12 C(K_3) \ge C(K_2)⇒21​C(K1​)+21​C(K3​)≥C(K2​)

    A portfolio that never pays less must never cost less.

  3. ⇒C(K1)−2C(K2)+C(K3)≥0\Rightarrow C(K_1) - 2C(K_2) + C(K_3) \ge 0⇒C(K1​)−2C(K2​)+C(K3​)≥0
C is convex in KC \text{ is convex in } KC is convex in K
90100110010Underlying at expiryPayoff
Figure 4.3 · Why a butterfly can never pay out less than nothing. The three legs are drawn faintly and their sum in full. It is zero outside the wings, worth $10\$10$10 at the middle strike, and never below zero anywhere — which is the entire reason being paid to put it on would be an arbitrage rather than a view.

Why these are different from a mispricing. Saying an option is cheap on your model means you and the market disagree about volatility, and you might be wrong. Saying a butterfly trades at a credit means someone will pay you to take a portfolio that can never pay out less than nothing — no model, no view, no volatility assumption. That is why these constraints are checked mechanically and continuously: a violation is free money or, far more often, a sign that one of your inputs is stale.

Example 4.4

Calls on the same expiry are quoted at $12.00\$12.00$12.00 for the $90\$90$90 strike, $7.00\$7.00$7.00 for the $100\$100$100 and $3.50\$3.50$3.50 for the $110\$110$110. Is anything wrong?

Show the worked solutionHide the worked solution

Worked solution

  1. Formula
    C(K1)−2C(K2)+C(K3)≥0C(K_1) - 2C(K_2) + C(K_3) \ge 0C(K1​)−2C(K2​)+C(K3​)≥0
  2. Substitute
    =12.00−2(7.00)+3.50= 12.00 - 2(7.00) + 3.50=12.00−2(7.00)+3.50
  3. Solve
    =12.00−14.00+3.50=1.50≥0✓= 12.00 - 14.00 + 3.50 = 1.50 \ge 0 \quad \checkmark=12.00−14.00+3.50=1.50≥0✓
  4. Call spreads: 12.00−7.00=5.00≤10✓\text{Call spreads: } 12.00 - 7.00 = 5.00 \le 10 \quad \checkmarkCall spreads: 12.00−7.00=5.00≤10✓
  5. 7.00−3.50=3.50≤10✓7.00 - 3.50 = 3.50 \le 10 \quad \checkmark7.00−3.50=3.50≤10✓
  6. Answer
    consistent — no static arbitrage\text{consistent — no static arbitrage}consistent — no static arbitrage

Sanity check. Change the middle quote to $8.50\$8.50$8.50 and the butterfly becomes 12.00−17.00+3.50=−$1.5012.00 - 17.00 + 3.50 = -\$1.5012.00−17.00+3.50=−$1.50: you would be paid $1.50\$1.50$1.50 to hold a structure whose payoff is never negative. That is the violation to look for.

The rest of this lesson is in Premium

You have read the opening. 11 more sections follow, including 4 worked examples and 3 quick checks.

Start the free 7-day trialSign in

Nothing is charged for 7 days, and you can cancel before then. Or read Forwards, futures and the cost of carry in full, free.

← Option strategies: what each one is actually a bet onBorrow, dividends and what happens at expiry →
On this page
  • The constraints
  • Why prices must be convex in strike
  • Why a butterfly can never pay out less than nothing
  • Worked example

QuantMax · 141 lessons · 1342 questions · c5c0caa

  • Premium
  • Arbitrage trees
  • Horse racing
  • Invite friends
  • Account
  • About QuantMax

Firm names identify publicly reported question patterns and nothing more. QuantMax is not affiliated with, endorsed by, or recruiting for any firm named in the curriculum. Everything you do in lessons and the question bank is kept to your account.