Dynamic Vs Static Hedging

Buy an umbrella before you leave the house and you're covered for the whole day, rain or shine, without ever touching it again — that's a static hedge, set once and forgotten. Wearing a jacket you keep zipping and unzipping as the weather changes is a dynamic hedge: it works too, but it demands constant attention, and every zip is a small hassle. Both get you home dry. A trading desk faces exactly this choice for every option it sells: build a fixed portfolio of other instruments that pays off the same way no matter what happens (static), or hold a position in the underlying stock and keep adjusting it as the stock moves (dynamic).

The two strategies, precisely

A static hedge is a portfolio you assemble once, at trade inception, whose payoff matches the option's payoff — or comes close enough — at every future date and every possible path the stock could take, with no further trading required. A dynamic hedge is a position, usually in the underlying stock, that you continuously resize according to the option's delta, as covered in delta hedging frequency and costs. Delta hedging is inherently dynamic because delta itself changes as the stock price changes; a static hedge tries to sidestep that by finding a combination of other traded instruments whose combined delta already tracks the option's delta automatically, with no rebalancing needed.

Worked example 1 — a forward is (almost) free to hedge statically

The simplest static hedge in finance is a forward contract. Say a client wants to lock in buying a stock, currently $100, for delivery in one year at a forward price of $105 (with a 5% risk-free rate, that's the fair forward price: $100 \times 1.05 = 105$). The dealer's hedge: buy one share of stock today for $100, financed by borrowing $100 at 5%. In one year the dealer owes $100 \times 1.05 = 105$, i.e. $105, on the loan and holds one share, worth whatever the stock is then. Deliver that share to the client for $105, use the $105 to repay the loan exactly. Check two scenarios: if the stock ends at $130, the dealer still just delivers the one share already held and collects $105 — no adjustment was ever needed. If the stock ends at $80, same thing: deliver the same one share, collect the same $105. The hedge — one share, bought once — matches the obligation perfectly at every terminal price, with zero rebalancing trades along the way. That's what makes forwards cheap to trade: the hedge is trivially static.

Worked example 2 — statically replicating a digital option

A digital (or "binary") call pays a fixed $1 if the stock finishes above strike $K$, and nothing otherwise. Its payoff jumps instantly from $0 to $1 right at the strike, so its delta near expiry, right at the strike, becomes enormous and unstable — a dynamic hedger has to trade huge amounts of stock in a tiny price range right as expiry approaches (this is the pin risk the digital is famous for). A static alternative: replicate the digital with a tight call spread instead. Buy $1/\varepsilon$ calls struck at $K - \varepsilon/2$ and sell $1/\varepsilon$ calls struck at $K + \varepsilon/2$, where $\varepsilon$ is a small strike gap.

Take $K = 100$ ($100), $\varepsilon = 2$ ($2), so the spread is long calls at $99 and short calls at $101, sized $1/2 = 0.5$ contracts each. With the stock at $100, one year to expiry, zero rates, and 20% volatility, Black-Scholes gives the $99 call a value of about $8.43 and the $101 call a value of about $7.51. The spread costs $0.5 \times (8.43 - 7.51) = 0.5 \times 0.92 = 0.46$, or $0.46. Compare that to the digital's own Black-Scholes value under the same inputs, also about $0.46. The two agree closely — the call spread, assembled once with no further trading, approximates the digital's payoff and its price. Tighten $\varepsilon$ toward zero and the approximation becomes exact, but the position needed near the strike grows without bound — the same pin risk resurfaces, just pushed into the static portfolio's construction rather than into rebalancing.

Compare the sharp step of a digital's payoff (jumping from $0 to $1 right at the strike) against the smoother diagonal payoff of an ordinary call shown above. The steeper the jump, the harder any hedge — static or dynamic — has to work near that point.

What this means in practice

Static hedges are prized because once they're on, they're immune to two of dynamic hedging's biggest headaches: transaction costs (no more trades) and model risk from getting volatility slightly wrong (the replicating instruments are, themselves, traded market prices — you don't need a model to tell you what a call spread is worth, the market tells you directly). But static hedges only exist cleanly for a limited set of payoffs — barriers, digitals, and a handful of other exotics have known static-replication tricks; most path-dependent payoffs have no exact static hedge at all, and desks fall back to dynamic hedging, accepting its costs, because there's no alternative.

Practice

1. A client wants to lock in selling a stock in six months at a fixed price. Sketch the static hedge (what to trade today, and what happens at expiry) the way worked example 1 did for the forward-purchase case.
2. Using the call-spread method from worked example 2, replicate a digital put (pays $1 if the stock finishes below $50) with strikes $49 and $51. Which two ordinary options do you need, and are you long or short each one?