Skip to content

Summary

Black–Scholes–Merton (BSM) prices a European option from five inputs — spot, strike, time, rates, and volatility — assuming the underlying follows geometric Brownian motion with constant volatility. Its deepest insight is that the option price is independent of the stock's expected return, enabling risk-neutral valuation1. In practice traders invert the model: they observe prices and solve for the one unobservable input, implied volatility2.

Inputs and Intuition

Input Symbol Observable? Role
Spot price S₀ Yes Level of the payoff distribution
Strike K Yes Payoff boundary; N(d₂) ≈ risk-neutral exercise probability
Time to expiry T Yes Grows the distribution as σ√T
Risk-free rate r Yes (approximately) Carrying cost; drives put–call parity term
Volatility σ No Width of the distribution; the trader's input
Dividend yield q Estimated Extension: replace S₀ with S₀e^(−qT)

Formulas (European, no dividends): c = S₀N(d₁) − Ke^(−rT)N(d₂), with d₁ = [ln(S₀/K) + (r + σ²/2)T] / (σ√T) and d₂ = d₁ − σ√T1.

The volatility input dominates: it is the only input not directly observable, and small changes in σ move option values materially. This is why the model is used in reverse — quote a price, back out an IV, and trade the vol, not the dollar price2.

What Each Assumption Buys and Costs

Assumption What it buys What it costs
Geometric Brownian motion, lognormal prices Closed-form formula; no negative prices Fat tails and gaps in real markets are underpriced
Constant volatility σ tractability; single vol per option Reality: IV varies by strike and maturity (skew/smile)
Continuous trading & hedging Riskless dynamic portfolio → risk-neutral pricing Real hedging is discrete; hedge slippage is a real cost
No transactions costs or taxes Clean arbitrage arguments Market makers price spreads and fees in
Constant risk-free rate, no dividends in the base case Simple parity relationships Extended via q or PV-of-dividends adjustments1
Frictionless short selling Two-sided arbitrage enforcement Borrow costs and restrictions bind in stress

The constant-σ assumption is the one traders live with daily: a single flat IV almost never reproduces an option chain2.

Where the Model Breaks

  • Volatility skew/smile: implied volatility varies systematically by strike — OTM equity puts carry higher IV than OTM calls, reflecting demand for downside protection and the market's rejection of lognormality2. A flat-IV BSM quote will misprice wings relative to the body; skew must be handled explicitly (see Volatility Concepts).
  • Jumps and gaps: earnings, Fed decisions, and crashes violate the continuous-path assumption. BSM systematically underprices short-dated OTM puts around events; the "rush and crush" in IV around events is the market re-pricing this jump risk3.
  • American-style early exercise: no closed-form BSM price exists; binomial or other numerical methods are required (see Binomial Model).
  • Term structure of rates and vol: a single flat r and σ fails on long-dated chains; solve inputs month-by-month to reproduce the market3.

Despite these breaks, BSM remains the lingua franca: greeks, IV, and parity conventions are all defined relative to it.

Links

Footnotes

  1. Hull, Options, Futures, and Other Derivatives — BSM topic: assumptions, differential equation, formulas, dividend-yield extensions. ↩↩↩

  2. Natenberg, Option Volatility & Pricing — volatility topic: IV as the unobservable input; lognormal assumption; skews. ↩↩↩↩

  3. Passarelli, Trading Option Greeks — volatility chapters: event-driven IV behavior and chain-consistency checks. ↩↩