Average Rate Option
An average rate option (ARO), commonly known as an Asian option, is an exotic derivative whose payoff is determined by the average price of the underlying asset over a specified observation period rather than the spot price at expiration, making it less expensive than vanilla options (because averaging reduces volatility) and particularly suited for hedging exposures based on average prices, such as monthly commodity purchases or periodic foreign exchange conversions.
Key takeaways
- ARO payoff at expiration for a call: max(A_T - K, 0), where A_T is the arithmetic or geometric average of the underlying price over the averaging period and K is the strike price.
- Average rate options cost less than vanilla options with the same strike and maturity because averaging reduces the effective volatility—the variance of the average is lower than the variance of the spot price by a factor of approximately 1/n for n equally-weighted observations.
- Arithmetic average AROs have no closed-form solution; geometric average AROs can be priced in closed form (Kemna-Vorst model) and are used as control variates in Monte Carlo pricing of arithmetic AROs.
- Corporate treasurers at multinational companies widely use AROs to hedge foreign exchange exposure arising from periodic repatriation of earnings, because their realized exchange rate reflects the average rate over the fiscal period, not the rate on any single day.
- Commodity producers and consumers use AROs to hedge exposure to the average monthly settlement price—oil producers hedging against the average NYMEX WTI price over the calendar year, for example.
Explanation
The average rate option's defining feature is the substitution of a period average price for the spot price in the payoff calculation. This seemingly simple modification has profound implications for valuation, risk management, and the set of use cases where AROs are the optimal hedging instrument. The reduction in effective volatility is the most important valuation consequence: whereas vanilla option prices scale with the volatility of the terminal price σ√T, ARO prices scale with the volatility of the average price, which is approximately σ√(T/3) for a continuously averaged option—a reduction of approximately 42% in the volatility input, and consequently a substantial reduction in option premium.
The pricing challenge arises from the arithmetic averaging convention. The sum of lognormally distributed random variables is not itself lognormal, which means no closed-form Black-Scholes-type formula exists for arithmetic average AROs. The industry relies on the Kemna-Vorst approximation (which matches the arithmetic average with a geometric average using adjusted moments), the Turnbull-Wakeman approximation (matching the first two moments of the distribution of the arithmetic average), or Monte Carlo simulation. The geometric average version, priced exactly by a modified Black-Scholes formula with adjusted volatility σ_geo = σ × √((T+Δt)/(3T)) and adjusted forward rate, is frequently used as a control variate to reduce Monte Carlo variance.
For corporate FX hedgers, the ARO's alignment with economic exposure is its primary advantage. Consider a European exporter billing in USD and converting earnings monthly at the prevailing spot rate. Their economic cost is not the EUR/USD rate on any single day but the average rate over the year. A vanilla put option struck at 1.10 EUR/USD would only pay off if the spot rate is below 1.10 at expiration—but if the rate averaged 1.05 throughout the year and recovered to 1.12 by expiration, the economic loss is unhedged while the vanilla put expires worthless. An ARO struck at 1.10 based on the average monthly rate over the year would pay the full 5 cents of economic loss.
The averaging specification—discrete vs. continuous, arithmetic vs. geometric, the observation frequency, and whether the averaging period has already started (in-progress, creating a 'fixed strike' and 'floating strike' distinction)—significantly affects ARO pricing and must be carefully specified in the contract. In-progress AROs where some observations have already been fixed use the known observations to compute the remaining uncertainty and price accordingly; the accrued average acts like a partial fixing, reducing the residual optionality.
Formula
ARO Payoff (call): max(A_T - K, 0) ARO Payoff (put): max(K - A_T, 0) Geometric Average Volatility: σ_geo = σ × √((2n+1)/(6(n+1))) for discrete observations Kemna-Vorst Approximation: Treat as vanilla option with σ_adj = σ_geo and F_adj adjusted forward
Example
A US multinational expects to receive CNY 120 million from its China operations over the next 12 months, converting approximately CNY 10 million per month at prevailing spot rates. The current USD/CNY spot rate is 7.10. The treasurer purchases an arithmetic average rate call option on USD (put on CNY) with: notional CNY 120 million, 12 monthly averaging observations, strike of 7.20 USD/CNY, maturity 12 months. If the CNY depreciates and the 12-month average rate is 7.35, the ARO payoff = CNY 120M × (7.35 - 7.20) / 7.35 = approximately USD 2.45 million, compensating for the depreciation above the strike. The ARO premium is approximately 1.8% of notional (USD 3.1M equivalent) versus 2.9% for a vanilla option with the same strike and maturity—a 38% saving reflecting the variance reduction from averaging. The treasurer's effective floor on the conversion rate is 7.20, with unlimited participation above 7.20 net of the premium paid.
Related terms
Asian Option Automatic Exercise Call Option Exchange Floor Hedging Monte Carlo Simulation Option Premium Put Option Spot Price Spot Rate