Temperature, Oil Prices, and Weather Derivatives

Lynne Kiesling

Ooooh, baby, it’s cold outside! The cold snap affecting a large proportion of the U.S. population is placing upward pressure on oil prices, particularly because the cold snap is affecting U.S. regions that rely more on home heating oil than natural gas or electric heating:

The onset of cold weather in the U.S. Northeast, the world’s biggest heating oil market, and a shift in investment flows has lifted oil from a 20-month low of $49.90 on January 18 and brought it into the danger zone for consumer nations’ economies.

This tight relationship between temperatures and oil prices raises a question (and hence, an opportunity): why can’t oil companies just forecast and produce to expectations? Of course, the problem is that oil companies, and oil traders, cannot forecast future weather perfectly. What if you produce enough heating oil for a given number of customers and an average temperature of 30 degrees, but the winter is colder than average, and the average ends up being 25 degrees? Furthermore, what if the average stays around 30 degrees, which you expected, but there’s more volatility, more cold snaps and heat waves during the winter? Either of those cases (forming expectations about the first moment of the temperature distribution, or the second moment of the temperature distribution) is one in which producers can’t form perfect expectations; temperature is a stochastic variable with some structure to it, some predictable core, and then some disturbance term as well. That disturbance term is what makes oil prices move with temperature in the winter, because the predictable core is what producers can use when planning their heating oil production. But if that disturbance term is large, producer profits can really fluctuate.

Therein lies the opportunity: financial instruments to hedge this temperature risk. The Chicago Mercantile Exchange offers a range of weather products: monthly heating degree day [that’s for winter stuff – ed.], seasonal strip of heating degree days, monthly and seasonal cooling degree days for summer, monthly and seasonal frost days, and monthly and seasonal snowfall. The CME’s contracts cover 18 U.S. cities, 9 European cities, 6 Canadian cities, and 2 Japanese cities.

Furthermore, the CME offers both futures and options (puts and calls) building off of those future contracts. Think, for example, of the heating degree monthly contract. If you think next month’s temperature is going to vary from the CME’s temperature index in a particular direction by a particular amount, then you could buy an option to cover that difference. This variety of regional and temporal contracts creates quite a few opportunities for oil producers to hedge the non-systematic portion of changes in weather.

From the CME’s weather background page:

In general, weather derivatives cover low-risk, high probability events, while weather insurance typically covers high-risk, low-probability events, as defined in highly tailored or customized policies. For example, a company might use a weather derivative to hedge against a winter that forecasters thought would be 5° Fahrenheit warmer than the historical average (a low risk, high probability event) since the company knows its revenues would be affected by that kind of weather. But the same company would most likely purchase an insurance policy for protection against damages caused by a flood or hurricane (high-risk, low-probability events.)

One thought on “Temperature, Oil Prices, and Weather Derivatives

  1. In the case of energy traders/hedgers, I don’t see why weather derivatives would be preferable to energy futures or options; what if it never gets cold but heating oil prices rise anyway for another reason? You’d then have been better off hedging in the energy markets themselves.

Comments are closed.