Weather Derivative

A weather derivative is a financial contract whose value depends on a measured weather variable, such as the average temperature in a city over a month, total rainfall or snowfall in a season. Businesses use weather derivatives to protect revenue or limit costs that rise and fall with the weather, such as a gas utility facing a mild winter.

Weather derivatives are a branch of derivatives rather than insurance. An insurance policy pays when the buyer proves a loss; a weather derivative pays according to an index, whether or not the holder suffered any damage. That makes settlement quick and objective, but it exposes users to basis risk, the chance that the index and the business's actual losses do not move together. Contracts trade both on exchanges and as OTC derivatives.

Imagine an ice-cream stand owner who knows a cold, rainy summer will hurt sales. She makes a deal with someone else: if the summer turns out cooler than normal, the other party pays her a fixed amount for every degree below the usual temperature; if it is warmer, she pays them. Whatever the weather does, her income becomes steadier.

Nobody needs to inspect her accounts or argue about how many cones she failed to sell. The payout depends only on the thermometer readings from an official weather station, which both sides can check. That simple, shared measuring stick is what separates a weather derivative from a normal insurance claim. It also means the deal can be agreed months in advance and settled within days of the season ending, without any loss adjuster.

The most common weather derivatives are based on degree days, which measure how far daily average temperature departs from a base of 65 degrees Fahrenheit (about 18 degrees Celsius), the level at which buildings typically need neither heating nor cooling. A heating degree day (HDD) is defined as:

• $\text{HDD}$ is the heating degree day value for one day.
• $T_{\text{avg}}$ is the day's average temperature in degrees Fahrenheit, usually the mean of the daily high and low.
• $65$ is the base temperature; a cooling degree day (CDD) is the mirror image, $\max(0,\ T_{\text{avg}} - 65)$.

Daily values are added up over a month or season to form an index. If a January day in Chicago has a high of 30 and a low of 14 degrees, the average is 22 and the day contributes $65 - 22 = 43$ HDDs. A month averaging 40 HDDs a day produces an index of $31 \times 40 = 1{,}240$. On CME contracts each index point is worth 20 dollars, so that index corresponds to a notional value of 24,800 dollars per contract, and every extra HDD above the agreed level moves the payout by 20 dollars.

Weather contracts emerged in the US energy industry in the late 1990s, after deregulation left power and gas companies exposed to swings in demand. The Chicago Mercantile Exchange (CME) received approval from the Commodity Futures Trading Commission and listed its first weather futures contracts, based on HDDs for four US cities (Atlanta, Chicago, Cincinnati and New York), in September 1999; CDD contracts followed in January 2000. CME later added more cities, options on the futures and contracts for other regions. Exchange-traded weather contracts clear through a central counterparty, while larger or more tailored deals are arranged privately between companies, banks, reinsurers and specialist funds.

• Energy companies. Utilities and gas distributors hedge the volume risk of mild winters or cool summers, which change how much fuel customers burn.
• Agriculture. Farmers and food processors can hedge rainfall or temperature that affects yields, alongside crop insurance and the commodities market.
• Retail, leisure and construction. Ski resorts, theme parks, retailers and builders use contracts tied to snowfall, rain days or frost.
• Investors and reinsurers. Weather risk has little correlation with stock and bond markets, which makes it attractive for diversification, much like other insurance-linked risk.

Unlike oil or wheat, weather cannot be stored or delivered, so the usual arbitrage link between spot and futures prices does not apply. Weather derivatives are priced instead from historical station records, weather forecasts and statistical models of temperature. Long-term warming trends complicate this, because a 30-year average may no longer describe the climate a contract will face.

Other limitations include thin trading in many contracts, the location gap between the measuring station and a business's actual sites, and the difficulty of matching an index to real losses. Companies hedging with weather contracts may also need to consider whether they qualify for hedge accounting, since the contracts are otherwise valued at mark-to-market through profit and loss.

• CME Group: Weather futures and options
• NOAA Climate Prediction Center: Degree days statistics