For decades the insurance industry has been marketing diverse products protecting against the material damage caused by climatic events (the effects of wind, of water etc.). This type of insurance is well developed today and represents the earliest form of “weather product”. But how could the financial loss suffered by an ice-cream producer be insured when the summer season was not “sufficiently sunny” and the demand for ice-cream was sorely lacking?

Until now, this type of risk has not been insured, simply because it was “uninsurable”. It represents the “venture risk” in a certain way, given that all the “uninsurable” risks are normally retained by the client (for example, the consequences of launching a new product or opening a new retail office). Therefore, completely new types of weather products (weather insurance and weather derivatives) had to be invented.

The principal characteristics of weather products are:

  • They apply to the “weather hazard” which does not create any material damage;

  • They can reduce the negative financial impact of a “weather event”.

    Insurance or finance weather product?

    Within the term “weather product”, there are in fact two possible forms : weather insurance or weather derivative.

    Weather insurance has the form of any insurance contract. It can be an endorsement or a new section in an existing contract.

    The weather derivative is a financial tool, often an option, just like an option on stocks, except that it is in over-the-counter market. The legal structure is the International Swap and Derivative Association (ISDA) standard.

    The advantage of the weather insurance is adaptability of the insurance form. Furthermore, the product can be much more complex (for example, with multiple triggers). On the other hand, a financial option is not subject to insurance taxes. For both forms, the aim is the same: decreasing the volatility due to the weather hazard.

    Industries who have a real weather risk are numerous. Some analysts estimate that between 20% and 30% of GDP can be affected by one weather hazard. Among the most important potential users are:

  • the energy industry

  • tourism and leisure industries

  • agriculture and agribusiness

    Example of a weather product

    In practice, such a product can have numerous applications. Let us take an example of an amusement park. Looking at the daily attendance statistics, we can easily realise they can fluctuate significantly. The professionals can explain why on one given day the attendance level was less than expected. Even if the reasons for the phenomenon are not always sufficiently formalised, they can observe the lack of visitors when the weather is rainy.

    A more detailed analysis of the case would make us notice that the impact of weather conditions on the number of visitors is more significant when the morning is rainy, rather than the afternoon. (When it is raining in the afternoon, most of the visitors are usually already there). But let us leave aside this detail since it brings an additional complexity to our case.

    The amusement park will then purchase a “climate product” working in the following way:

    For each day during the “covered period”,

  • If the daily precipitation level is below 3 mm, no compensation will be granted ;

  • If the daily precipitation level is equal or above 3 mm, the park will get $80,000 of flat indemnity.

    A $800,000 deductible for the whole “covered period” is provided.

    The 3 mm trigger for the precipitation level is calculated according to the park's data and statistics (the 3 mm figure seems to fit the situation perfectly). The $80,000 of daily compensation corresponds to the average deficiency of 800 visitors times the average ticket price being $100.

    Since the indemnity calculation only depends on the precipitation level, the historical data from a recognised meteorological institution could be used for the “as-if” simulation (see stats):

    One can see that this product meets the client's needs. For rainy seasons (1997 and 1999 in our example), the cover is working and the client receives an indemnity for the supposed lack of visitors. So, this amusement park succeeds in mitigating its rain exposure, its main weather hazard.

    How does it work?

    As explained, the main characteristic of these products is not to cover material damage, but rather the increase or decrease in earnings. The difficulty is to create a product for which the indemnification calculation is beyond dispute. If the product indemnifies the “lack of turnover”, it is too vague. Lack of turnover can be due to many reasons that have nothing to do with climate! That is the reason why one always uses (at minimum) a weather index.

    In most simple cases, the product is defined in three steps:

  • the measure ;

  • the weather index ;

  • the financial characteristics of the product.

    The measure can be any kind of weather measure done by an independent institution. It can be temperature, wind speed, sunlight, snow level etc. For an amusement park in the French city of Montpellier, the measure could be “the daily rainfall measured by Météo France in Montpellier”.

    The weather index is a mathematical function of the measure previously defined. For example, it can be the sum during the risk period of all recorded measures. Or it can be the number of days within the risk period for which the measure is higher than 25 mm. Or it can be the step function which is always equal to zero, except if the measure is higher than 150 mm at least one day during the risk period. In the amusement park example, the index is the number of days for which the measure is higher than 3 mm

    Heat and cold

    Among all the possible weather indexes, there are two important ones related to the temperature: heating degree days (HDD) and cooling degree days (CDD). The HDD index is equal to the sum, day by day, of the number of heating degrees needed to reach a reference temperature level commonly fixed at 18° C or 65°F. Symmetrically, the CDD index is equal to the sum, day after day, of the number of cooling degrees needed to reach a reference temperature level.

    Those two indexes are commonly used for two or three years because they can reflect the energy consumption accurately. The HDD is often used during the winter season and the CDD index is mainly used during summer season.

    Finally, the financial characteristics of the product enable calculation of the indemnity to be paid as a function of the weather index.

    In the amusement park example, the indemnity is equal to i x $80,000 with a deductible of $800,000. The indemnity is also equal to (i-10) x $80,000 for i>=10. Often, the financial vocabulary is used, so, in this case, 10 is called the strike and $80,000 is the tick value.

    More complex possibilities

    Beyond this quite simple example of weather products, it is possible to set up very complex structures.

    For example, so far only one measure has been used: the rainfall or the temperature measured daily in only one meteorological station. But it is possible to use the information from different measuring stations, for example, to use 15 different stations for the measure of temperature and five other ones for the snowfall. In this case, the index is a function of all the collected data.

    It is also possible to use a compensatory trigger. The client is indemnified only if he can prove that he has really suffered a loss or a lack of revenue, compared to previous years.

    The only limits are the client's concerns and imagination.

    Structuring the deal

    When the client wants to buy a weather product, the definition of it may need a long and difficult study. As a matter of fact, it is necessary to optimise the cover and reduce the basis risk (the possibility for the client to suffer a real loss for which the cover is not triggered).

    The information required from the client is mainly the historical turnover or financial balance in as much detail as possible (ideally on a daily or weekly basis). Then, the arranger calculates numerous weather indexes using historical measures and retains only the most correlated one.

    Finally, the financial conditions of the contract are defined according to the client's willingness to transfer the weather risk.

    Pricing the deal

    Once the conditions of the arrangement are established, pricing the deal takes into account mainly two elements : on one hand, the expected value of the indemnity and, on the other hand, the risk transferred: that is the volatility, often measured through the standard deviation.

    Conclusion : weather products for everybody.

    Within a few years when this new market matures, some predict that it will be more important than the current catastrophic insurance market (windstorm, flood, earthquake...). Personally, I am going to Brazil this summer and I am longing for the time when a “cancellation for bad weather” insurance will be offered to me with my vacation stay. This could be reality very soon.

  • Marc Hannebert is ART manager of SOREMA, Paris. E-mail: mhannebert@sorema.fr