Exotic Derivatives: Beyond Vanilla Options

We have spent a lot of time on vanilla calls and puts. But now Wilmott opens Part Two of the book, and things get interesting. Chapter 22 introduces exotic derivatives, contracts that keep quants employed and traders nervous.

“Exotic” just means “anything more complicated than a plain call or put.” Wilmott’s goal is to give you a mental toolkit for classifying any derivative you encounter.

Why Bother with Exotics?

Exotic options exist for practical reasons. Cheaper protection, payoffs matching specific business cashflows, currency hedging that depends on average exchange rates over a quarter. From a quant’s perspective, exotics are harder to price and more model-dependent. The risks are hidden. A trader might have a gut feeling about the price but no idea how to hedge properly. And hedging is really all that matters.

Wilmott makes an important point: do not try to decompose an exotic into a portfolio of vanillas. If you can do that exactly, it was never exotic in the first place.

The Six Features That Matter

Wilmott proposes classifying every exotic option by six mathematical features. These tell you almost everything about how to price and hedge the contract. Nothing to do with the option’s name or the underlying. Purely about the math.

1. Time Dependence

Some contracts have features that are only active at certain times. A Bermudan option, for example, can only be exercised on specific dates (not any time like an American, and not just at expiry like a European). This means when you write code to price it, your time grid has to line up with those special dates. Not difficult, just annoying.

2. Cashflows

Some contracts pay money during their life, not just at expiry. Bond coupons, range notes that pay you every day the stock stays in a band. If the cashflow is discrete, the option value “jumps” at that moment by the amount paid. If the cashflow is continuous, you add a source term to the Black-Scholes equation instead. One catch: the cashflow must be a deterministic function of the asset price and time. Random cashflows need different tools.

3. Path Dependence (The Big One)

Many exotic payoffs depend not just on where the stock price ends up, but on the entire path it took to get there. This is path dependence, and it comes in two flavors.

Weak path dependence: the contract cares about whether something happened, but you do not need extra variables. A barrier option dies if the stock hits a level. Two paths ending at the same price can have completely different payoffs. Yet you still solve the regular Black-Scholes equation with special boundary conditions. No extra dimensions.

Strong path dependence: you need an extra variable. An Asian option depends on the running average. You cannot know the option value from just the current price and time. The average becomes a new independent variable, pushing the equation from two to three dimensions. More expensive computationally.

4. Dimensionality

The number of independent variables. Vanilla: two (stock price, time). Barrier: still two. Asian: three. You also get three dimensions from two random underlyings (option on the max of two stocks). Higher dimensions mean longer computing time. Rule of thumb: low dimensions favor finite differences, high dimensions favor Monte Carlo.

5. Order

A compound option is an option on another option. A call on a put gives you the right to buy a put at some future date. To price it, first price the underlying option, then use that result. Layer cake.

The practical problem: compound options are very model-sensitive. The payoff depends on the market value of the underlying option, not your theoretical price. Second-order options force you to care about what the market does, not just what your model says.

6. Embedded Decisions

Early exercise in an American option is the simplest example. The passport option is more interesting: you get a trading account, buy and sell freely, keep profits at expiry, losses are written off. The decision is what and when to trade.

The pricing assumes the holder acts optimally, maximizing the option value. Finite-difference methods work best because you can check the optimal decision at each grid point.

Some Concrete Examples

Wilmott gives quick sketches of several exotic types.

Compound and chooser options are options on options. The chooser lets the holder pick whether to receive a call or a put at some decision date. In practice, these are sensitive to the pricing model because the payoff depends on market prices of other derivatives.

Range notes pay you based on how long the stock stays within a specified band. They are popular in FX and fixed income. Every day the stock is in the range, you accumulate money.

Barrier options (knock-in, knock-out) are the most common exotics. Cheap because the payoff is conditional on the stock not hitting (or hitting) a specific level. We will spend all of Chapter 23 on these.

Asian options depend on the average price over some period. They are strongly path-dependent, needing three dimensions. The average can be arithmetic or geometric, continuously or discretely sampled.

Lookback options depend on the realized maximum or minimum of the stock. The extreme example pays the difference between the highest and lowest prices, effectively letting you buy at the low and sell at the high. Every trader’s dream. And very, very expensive.

The Classification Table

Here is a summary of what each feature means for your coding and pricing:

The Takeaway

This chapter is really about how to think. When someone hands you a term sheet for some strange derivative, do not panic. Ask six questions: Is there time dependence? Cashflows? Path dependence (weak or strong)? How many dimensions? What order? Any embedded decisions?

The answers tell you what pricing method to use, whether you can reuse old code, how long it will take to build, and how fast it will run. That is the power of classification. It turns an impossible-looking problem into a checklist.

And we are still inside the Black-Scholes world. Later chapters will relax those assumptions. But first, we need to learn the exotics themselves.

Previous post: The Trading Game: Delta Hedging in Practice

Next post: Barrier Options: Knock-ins, Knock-outs, and How They Work