Home Publications Presentations Climate of Victoria 10 Day Forecasts Seasonal Outlook Other Links

Stern, H. (2001) The application of weather derivatives to mitigate the financial risk of climate variability, extreme precipitation events and extreme temperature events. Symposium on climate variations, the oceans and societal impacts, 81st Annual Meeting, Amer. Meteor. Soc., Albuquerque, New Mexico, 14-19 Jan.,2001.

Evidence of the challenge faced by the meteorological community to apply risk management products from the financial markets is growing (Dischell, 2000). Papers presented to recent symposia, and papers published in prestigious journals, stand as testimony to the increasing importance of weather derivatives.

This paper presents an approach to the pricing of weather derivatives that employs a combination of empirical forecast verification, synoptic classification, and global climate indice data.

The earliest published work on the subject is that by the current author (Stern, 1992), who employed option pricing theory to establish a measure of the economic consequences of changes in the global mean temperature.

Stern's (1992) work was prompted by a 1991 call by the Australian Electricity Supply Industry Research Board to conduct greenhouse research related to electricity supply. It is the energy and power industry that has, so far, taken best advantage of the opportunities presented by weather derivatives. Clemmons et al (1999), in a paper included in Geman's fundamental (1999) work, report the first weather derivative contract as a "temperature-related power swap ... transacted in August 1996", about the same time as the subject was first being discussed in energy and power industry journals (Simpson, 1996/97).

The Australian Electricity Industry's entry into the weather derivatives market was reported by Macleay (1999) as resulting in "(South Australian) taxpayers ... (losing) $1.6 billion through electricity trading by their state-owned power companies". Locke (2000) concludes that "after a summer of extreme price volatility and poorly judged government intervention, the Australian electricity market finds itself stuck in no-man's land between regulated and free markets".

The current paper contends that some of the difficulties associated with weather derivatives may be attributed to theoretical pricing models inadequately describing the probability distribution of expected outcomes. In much the same manner as Stern (1999a) employed empirical stock market data to improve upon the option pricing derived using the Black and Scholes (1973) theoretical model, so one might also improve upon current weather derivative prices by using forecast verification data.

For example, the probability distribution of temperature outcomes, given initial anticyclonic southsoutheasterly flow, and a maximum temperature forecast for Melbourne of 35C four days in advance, is little different from the probability distribution associated with a corresponding forecast of 25C. By contrast, the probability distribution, given a maximum temperature forecast of 35C only a day in advance, is very different from that associated with the corresponding 25C forecast.

The approach may also be applied to managing risk associated with extreme cases of other weather parameters (such as precipitation) and also for other weather-sensitive industries. Stern (1999b) reports a case of extreme (and unjustified) price volatility in agricultural commodities arising from a failure to take into account forecast verification data.

In summary, the paper presents examples of the pricing of weather derivatives about the occurrence of a variety of events such as:

Home Publications Presentations Climate of Victoria 10 Day Forecasts Seasonal Outlook Other Links