layout: default title: The Black–Scholes Model ——————————
The Black–Scholes Model
Sukrit Mittal Franklin Templeton Investments
Positioning of This Lecture
Students already know:
- One-step binomial model
- Replication and no-arbitrage pricing
This lecture answers:
What happens when time becomes continuous?
Black–Scholes is not a new idea.
It is the binomial model pushed to its logical limit.
Outline
- From binomial trees to continuous time
- Market model and assumptions
- Replication strategy in continuous time
- Self-financing portfolios
- Derivation of the Black–Scholes PDE
- Boundary and terminal conditions
- Black–Scholes formula for European options
- Interpretation and limitations
- Exercises
1. From Binomial to Black–Scholes
Recall the one-step binomial model:
- Stock moves up or down
- Option is replicated exactly
- Price follows from no-arbitrage
Now increase the number of steps.
Let time steps shrink to zero.
This limit is the Black–Scholes world.
What Survives the Limit
From the binomial model we keep:
- Replication
- Self-financing strategies
- No-arbitrage pricing
What changes:
- Prices evolve continuously
- Risk is driven by Brownian motion
The logic does not change.
Only the mathematics does.
2. Market Model and Assumptions
We assume:
- One risky asset $S_t$
- One risk-free asset $B_t$
Risk-free asset:
\[\frac{dB_t}{dt} = r B_t\]Risky asset follows:
\[\frac{dS_t}{S_t} = \mu , dt + \sigma , dW_t\]These are modeling choices.
Not truths.
Interpretation of Parameters
- $\mu$: expected return (irrelevant for pricing)
- $\sigma$: volatility (central)
- $r$: risk-free rate
Notice:
Expected return disappears from option prices.
This is not an accident.
3. Derivative Pricing by Replication
Let $V(t,S)$ denote the option price.
Construct a portfolio:
- Hold $\Delta_t$ shares of stock
- Hold $\beta_t$ units of the bond
Portfolio value:
\[\Pi_t = \Delta_t S_t + \beta_t B_t\]Choose $\Delta_t, \beta_t$ to replicate $V$.
Self-Financing Condition
The portfolio is self-financing if:
\[d\Pi_t = \Delta_t , dS_t + \beta_t , dB_t\]No external cash is injected.
Replication relies on this constraint.
This mirrors the binomial model exactly.
4. Ito’s Formula (Used, Not Worshipped)
Apply Ito’s lemma to $V(t,S_t)$:
\[dV = \partial_t V , dt + \partial_S V , dS_t + \tfrac12 \sigma^2 S_t^2 \partial_{SS} V , dt\]Ito’s lemma is bookkeeping.
Replication is the idea.
5. Derivation of the Black–Scholes Equation
Match the dynamics of $V$ and $\Pi$.
Choose:
\[\Delta_t = \partial_S V\]The stochastic term disappears.
The portfolio becomes locally risk-free.
No-arbitrage implies:
\[\partial_t V + \tfrac12 \sigma^2 S^2 \partial_{SS} V + r S \partial_S V - r V = 0\]This is the Black–Scholes PDE.
What Just Happened
- Risk was eliminated by replication
- Expected return $\mu$ vanished
- Pricing became deterministic
This is no-arbitrage in action.
Not probability magic.
6. Boundary and Terminal Conditions
For a European call option:
Terminal condition:
\[V(T,S) = (S - K)^+\]Boundary conditions:
- $V(t,0)=0$
- Linear growth as $S \to \infty$
A PDE without conditions is meaningless.
7. Black–Scholes Formula
Solving the PDE yields:
\[C(t,S) = S \Phi(d_1) - K e^{-r(T-t)} \Phi(d_2)\]where:
\[d_1 = \frac{\ln(S/K) + (r + \tfrac12 \sigma^2)(T-t)}{\sigma \sqrt{T-t}}, \quad d_2 = d_1 - \sigma \sqrt{T-t}\]This is a solution—not an assumption.
Interpretation
- $\Phi(d_1)$: delta-adjusted probability
- $\Phi(d_2)$: risk-neutral exercise probability
Probabilities appear.
But pricing came first.
8. Why Risk-Neutral Valuation Works
Replication implies:
All equivalent martingale measures give the same price.
Risk-neutral pricing is a shortcut.
Replication is the foundation.
Never confuse the two.
9. Limitations of Black–Scholes
- Constant volatility
- Continuous trading
- No transaction costs
Markets violate all three.
The model survives anyway.
Structure beats realism.
10. Exercises
Exercise 1
Derive the delta-hedging strategy explicitly for a European call.
Exercise 2
Show that the Black–Scholes price satisfies the PDE.
Exercise 3
Explain why $\mu$ does not appear in the pricing formula.
Relate this to no-arbitrage.
Final Takeaways
- Black–Scholes is the continuous-time binomial model
- Replication, not preferences, drives pricing
- PDEs replace backward induction
- Assumptions are strong—but explicit
Next: Greeks and hedging errors.
Where theory meets practice.