Benefits of Using Climate Forecast Information

Skillful seasonal climate forecasts have many uses. In agriculture, they may allow a farmer to match cropping decisions more closely to expected climatic events. In the energy industry, improved forecast skill might help gas companies with inventory management and with anticipating price fluctuations. Hydro-dependent utilities might benefit from seasonal forecasts of precipitation and runoff, and utilities with seasonal demand profiles might benefit from seasonal forecasts of heating or cooling degree-days. Climate forecasts can also improve decision making in the insurance industry, construction, fishery, and water management.

Climate forecasts are useful only in relation to the actions people can take, given forecast information, to improve their outcomes. One necessary ingredient of a useful forecast is its skill. Although the forecast skill can be mathematically estimated, studies of climate forecast value show that managers do not always know what the minimum level of forecast skill is required to start acting upon forecast information (e.g., see Making Climate Forecasts Matter). 

With the advent of weather derivatives in the late 1990s, it became much easier to attach a monetary value to a forecast. Weather derivatives are a type of contract whose payoff depends on occurrence or nonoccurence of specific weather events, for example, when temperature exceeds a certain level. It is not surprising that most active participants in weather derivative markets are utilities and energy management companies, whose earnings are strongly dependent on weather and climate. The markets, however, are growing rapidly, and companies in other sectors, such as construction, entertainment, and leisure, have started purchasing weather index instruments. Below are two examples showing how our climate forecasts can help you save money when managing financial risk related to weather (hedging) or make money betting on weather.

Hedging

One possibility to manage the risks associated with weather and climate fluctuations is to hedge it with a financial transaction, such as a weather option. Weather options, unlike traditional equity options, have a strike level based on the relevant measure of weather, such as heating degree days (HDD), cooling degree days (CDD), or cumulative average temperature (CAT). A typical example is a HDD put option that pays out a certain amount of money, when the current winter is much warmer than expected. Such a transaction may look like this:

Reference weather station:  Chicago O’Hare International Airport (WBAN #94846)
Underlying index:  Heating Degree Days
Term:  Nov. 1 – Mar. 31
Structure:  Put option
   Strike = 4850 HDD
   Tick size = $5,000
   Limit = $1 million
   Premium = $150,000

Let’s say you want to protect your company from a warmer than normal winter and buy this option. Since this is a put option, you would pay upfront a nonrefundable premium of $150,000. If the winter, indeed, is too warm, you would be paid back $5,000 for each HDD below 4850 until the limit is reached, as shown in Fig. 1. But if the winter turns out to be anomalously cold, with HDD > 4850, you would not receive any funds or a refund of the premium. Each year you face a decision whether to buy this option or it’s a waste of money. Climate Logic can help you to make this decision. For example, if you contacted us prior to the winter of 2008, we would advice you not to buy this option (or, at least, change the strike level to lower the premium), because our forecast for Chicago was 5130 HDD (or 330 HDD above the average for the past 10 years). The actual HDD (its settlement value) for the winter (Nov-Mar) of 2008 was 5360. As a result, you would save $150,000 on the premium. 

Fig. 1 (Source)

Betting on Weather

A skillful seasonal forecast can help a professional trader to make money when trading weather derivatives. Climate Logic issues seasonal HDD, CDD and CAT forecasts for all the cities listed on the Chicago Mercantile Exchange (CME). Weather derivatives can also be traded on WeatherBill.com, the first online weather risk management service. Table 1 shows how much money could be made by trading on WeatherBill with the help of our winter 2008 forecasts for the United States and Canada. In this example, the bets were made on seasonal (Nov 1 – Mar 31) HDDs for those cities that satisfied the following criteria:

1. Forecast HDD anomaly should be of the same sign relatively to both the 1971-2000 and 1998-2007 base periods. 

2. Magnitude of forecast HDD anomaly from the past 10-yr average should be greater than 0.2σ (σ - standard deviation).

A total of nine U.S. and four Canadian cities satisfied these criteris. Although this is an illustrative example only, the forecast numbers are from our real-time forecast. The contract prices are applicable to the 2009 winter season, if the contracts were purchased on April 29, 2008. We assume that the prices were about the same for the 2008 winter season. A contract price may vary over time and usually gets higher as the forecast period approaches. The lead time of our forecasts, however, is long enough to keep contract prices relatively low.

As Table1 shows, 10 out of 13 bets were profitable, with the cumulative return of 85%. The return for some cities would be even higher, if we placed bets relative to the 1971-2000 average. For example, the return for Des Moines and Kansas City would increase to 228% and 293%, respectively. The return would also be higher, if we restricted our bets to only those cities where magnitudes of forecast HDD anomalies were greater than 0.4σ. However, due to occasional disagreements between local and regional temperature anomalies, it is important to reasonably diversify your bets. For example, our forecast for the winter of 2008 called for warmer than normal conditions in the southeastern U.S. with a relatively high confidence. Indeed, the winter temperature anomaly for the entire southeastern U.S. climatological region was 0.7σ. For Atlanta, however, the temperature anomaly was only 0.2σ. As a result, the payout for the contract on that city was less than the cost of the contract. 

For comparison, we repeated the betting for the United States using the CPC forecast as a guidance. The bets were placed on those cities were the probability of Above and Below normal categories in the forecast was greater than 40%. Table 2 shows that it would result in a 79% loss. Restricting the bets to the cities with the probability of forecast categories greater than 50% (Chicago, Dallas, Houston, Kansas City) or changing the level to the 1971-2000 average could not make the outcome profitable. 

Contact us with your questions about the betting and hedging strategies. The first month of consultations is free with the subscription.

Table 1. Results of betting on HDDs for the winter (Nov-Mar) of 2008 using the forecast from Climate Logic

• Forecast Skill:
• HDD
• USA
• Canada
• Europe
• CDD
• USA
• CAT
• Canada
• Europe
• WD Forecast