Weather products see nothin’ but blue skies
New York: July 14, 1999
By John R. Stephenson

For centuries weather has been a preoccupation for people. Farmers plant in expectation of a bountiful harvest, but no matter how well they tend the fields, they remain at the weather’s mercy.

But farmers aren’t alone.

Electric utilities and power traders too must fully understand the relationship between weather and demand and the role it plays in short – and long-term power markets. Failure to understand these relationships leads to bottom line disaster.

For example, misjudging the weather at a given point in time can catch a utility short. Without adequate and timely resources to meet demand, utilities may turn to the open market for power or pay prohibitive punitive damages.

Being forced to buy expensive power on the open market led to the bankruptcy of LG&E Energy after last summer’s Midwest price spikes.

Changes in the weather also affect natural gas prices, one of the utility industry’s most important raw materials.

To better understand the connections between weather and power demand, picture a relationship diagrammed as a U-shaped curve (Figure 1).

Figure 1 illustrates the relationship between load and weather, which essentially looks like a saucer and depicts the two principal effects weather has on load.

The first is a seasonal or long-term effect on load. A move in either direction away from 65 F, either getting hotter or colder, results in a load increase. As temperatures increase, the need to cool homes and businesses increases, and as temperatures decrease the need for heat rises.

The second is a short-term or daily effect weather has on load (Figure 2). In the daily market, electric utilities gear production assets toward the expected load. As temperatures deviate from the expected, the need to increase capacity and meet the resultant load becomes apparent.

Many utilities use natural gas as a fuel for generation. Storms can sometimes disrupt natural gas supply and can wreck havoc on the cash markets for natural gas. Generally, gas prices are affected more by cold weather than hot, and a seasonal pattern appears.

For example, a strong price peak usually occurs in the December/January timeframe. Also, the price patterns for natural gas parallel the injection/withdrawal cycle. In April through October suppliers inject natural gas into storage. Withdrawal then occurs from November to March.

Injection months usually trade fairly flat, with discounts occurring from November to March. However, once injection stops and the cold weather strikes, prices rise.

Withdrawals from storage result in a limited and rapidly decreasing supply of natural gas. During a 24-hour period, a minimum level of demand from “must-run” generation or baseload units gets served.

Over and above this baseload level, generation capabilities are dispatched in rank order of least to highest variable cost – the expense associated with producing an additional unit of electricity. In general, variable cost components include fuel types and their associated heat rates. From cheapest to most expensive, the order of dispatch is generally: hydro, nuclear, coal, natural gas, fuel oil and heating oil.

If utilities forecast hourly load or weather accurately, then they know with certainty which units will serve the marginal load throughout the day. By understanding and quantifying these relationships better, utilities can more accurately streamline operations, including better coordination and scheduling of maintenance.

Knowing which units are likely to serve the market at a given hour also enhances competitiveness in the real-time trading arena. With an accurate hourly load curve established and cross referenced with a marginal unit’s variable cost, power traders make better decisions.

For example, if a utility or power marketer accurately predicts their marginal unit for the 15th hour will be a nuclear plant, and if they know the marginal cost for that unit to produce power, then they know the fair price of power during that hour. This is essentially a demand side approach to economic pricing.

By knowing the accurate weather forecast for a given hour, the load or demand for that hour can be accurately determined. Coupled with the dispatch order or supply curve for that hour.

The intersection of the hourly supply curve or supply stack and the demand for that hour gives them the equilibrium price for power at a given hour. This is shown in Figure 3.

These prices or equilibrium points in this supply/demand balance vary for every hour of a given day. From the perspective of a commodity trader, understanding these relationships remains crucial.

For traders well versed in these issues, substantial profits can accrue. If they forecast the weather and the load more accurately: then they possess two pieces of critical information.

By knowing the fuel on the margin and the price for a given hour, savvy traders can buy fuel in anticipation of a run-up in demand. Similarly, these traders can also find arbitrage opportunities for the power itself. These opportunities can make or break trading organizations in the volatile electricity market.

But so much for the implications of weather on the short-term trader. What then are the long-term risk mitigation implications of superlative weather knowledge?

Weather patterns tend to follow a specified path or trend for a period of time. Because of this phenomenon, there’s a strong likelihood that not only will the weather deviate from the expected temperature for a given day, but will continue to deviate from expectations for some time.

Therefore, utilities, power traders/marketers and even end-users may want to protect themselves from prolonged unfavorable deviations that tax their resources and dry up their cash flows.

Because of this, an ever-growing number of energy commodity marketers and risk management firms now offer a wide array of weather-related financial tools.

Over the past few years, there’s been a better understanding of the quantification techniques of the weather/price relationship and a resultant increase in the research and development of methodologies to better mitigate weather risk.

Companies such as Aquila, Enron and Duke Energy are active in this market. The services offered are designed for electric utilities and the broader commercial marketplace. For example, utilities concerned about a mild winter might purchase a financial revenue relied product.

In general, these instruments offer a financial payoff upon exercise to the purchaser. These instruments are generally structured to offer this payoff based upon a cumulative heating or cooling day (HDD or CDD) figure.

Heating or cooling degree-days are defined as the sum of the products of the days in which a deviation from 65 F occurred times the magnitude of the deviation. For example, if the temperature for a given day was 72F, a utility would have seven cooling degree-days (72-65*1=7).

The heating/cooling degree-days allow for a generally accepted measurement convention for determining relevant economic units for pricing temperature variations. To price these instruments and offer them for sale in the marketplace, trading houses and risk management firms must have substantial and significant understanding of weather patterns and modern portfolio theory.

The recent emergence of these instruments allows the uncertainty created by weather to be separated from other risks and dealt with directly. For the first time in human history, utilities and other businesses can effectively deal with one of their most daunting risk factors.

Utilities purchasing weather-related instruments now plan for the future without concern for adverse changes in the weather. This allows them to concentrate on improving their operational efficiencies and allows management to deal with the remaining risks in a more streamlined manner.

And for marketers of weather related financial instruments, the early entrants into the market benefit by accruing the greatest returns. By offering these services to their clients bases, trading firms offer a full and complete range of services, and weather, once an unknown and unquantifiable risk, turns from hurricane wind to pleasant spring breeze.

John R. Stephenson is vice president in charge of the utility and related services practice at Sterling Consulting Group, a Houston-based management consulting firm.