Exposure to the vagaries of the weather permeates some of the most fundamental business risks, and, in extremis, this is a risk that has been traditionally tackled by the re/insurance industry. Now, weather is being treated as a commodity, tradable within a market environment. The logical extension of this commoditisation is that it should be an exchange-traded instrument. Where are we now in the evolution of weather as a commodity?
Born in late 1997, the weather market is, like any 1/2 year old, voraciously exploring its environment and making exciting new discoveries on an almost daily basis. And like a toddler, it can sometimes appear a little unsteady on its feet. However, with the benefit of perspective and lessons learnt in other markets, we can see that the weather market's evolution is solid and that its potential is truly outstanding.
Authoritative data on market size have been recently published in a Weather Risk Management Association (WRMA)-sponsored survey. They show that our market has already closed 5,000 contracts, with a notional risk value of $7.5bn. To date, over 97% of risk covered has been in North America, although this proportion is decreasing as Europe, Asia and other areas catch on.
By way of comparison, the traded markets for electricity and gas in Europe alone are estimated at $80bn-$100bn in 2000, a figure that could increase by a factor of ten in the next five years as markets deregulate. Furthermore, the ‘cover' – the ratio of traded volume to the size of the net underlying (physical) transaction volume – is a good indication of market maturity and liquidity. Well-established markets typically see a multiple of between two and ten times the original risk being traded between dealers. This figure is still hard to yet for weather, but participants see room for substantial increase, especially outside North America.
The experience in gas and power markets has been of ever shorter lifecycles – markets are reaching maturity/full liquidity ever more rapidly – and of an S-shaped growth in volume over time. This is characterised by slow initial growth, then a period of very rapid expansion which tails off as the market approaches its natural size. However, this time to mature is still nowhere near three years and the weather market may not yet have reached the rapid growth stage.
Weather is different
Although it is tempting to see weather (and other markets) as extensions of a secular trend in energy markets, there are reasons why we should treat weather differently and on its own merits. For a start, because it has not evolved from a pre-existing cash market in an underlying commodity, it is immediately understandable that growth should be slower. Furthermore, ‘weather' is hardly a single, uniform commodity in itself.
Average temperature, maximum temperature, peak wind speed, degree of cloud cover, sunshine hours, daily rainfall, meltwater equivalent of snowfall – the list is diverse and seemingly endless. Furthermore, many real world exposures are not to the weather variable itself, but to ‘critical days' – the number where a certain condition applies (or is exceeded) within an observed period – or to the prediction, rather than the reality, of an event. Three consecutive days of rainfall may have a completely different effect from one torrential day, and businesses may need to act in anticipation of a freeze rather than react to the reality of sub-zero temperatures. So weather is complex, and the attempt to commoditise it will tackle individual elements such as average temperature over a season, or peak daily rainfall, one by one.
Weather is complex, but it is also much more pervasive than many other risks. The US Department of Commerce has famously estimated that at least $1 trillion (a million million) of the US economy is weather sensitive. This size of exposure, and the diversity of branches of the economy that are affected by it, suggests that the global foreign exchange market may be a more relevant benchmark than the energy markets. The FX market trades $1.5 trillion a day, and while the weather market currently serves ‘term' risk (for example, an entire winter season), there is no reason why a ‘spot' market (covering short-term weather risk) would not evolve. Since the market participants' perception of pricing would be constantly changing with updated forecasts, such a spot market would see much greater trading activity than the term market.
The principal forums for trading weather risk are loosely divided into primary and secondary markets. The primary market is the risk-absorbing function, where end user (‘net') risk is covered by market-making firms, largely from the energy, reinsurance and banking industries. A deal may have been developed entirely in-house by a dealer's marketing/origination and trading/structuring functions in sole contact with the client, or an intermediating channel such as a broker may have been used. The secondary market is the dealer-to-dealer market, and it is here that liquid trading allows ease of risk transfer and price discovery. Frequently, trading is done via an intermediating broker, or through a market-making firm's online trading platform.
The primary market, where deals are generally tailored to meet the customer's specific risk profile, can involve much larger deal sizes, and most closely resembles a reinsurer's normal business. The secondary market allows adjustment of trading portfolio profiles, as well as giving feedback on how other desks price specific risks. This market is concerned with reasonably generic structures (whereas a primary market product might be based on a multivariate structure, the secondary market would generally treat each weather dimension separately) and bases its pricing on observation centres with readily available data.
It could be argued that the moment when a market has reached maturity is when there are liquidly traded markets in its products on exchanges. Here, the basic units of market risk are stripped down to their lowest common denominator, and offered in a series of building blocks, for example, broken down into individual monthly contracts. Since the generic product has applicability to a wider range of traders than any tailored product, the trading interest should be commensurately larger and the market therefore more liquid.
Additionally, exchanges generally offer the benefits of clearing and centralised counterparty risk – a substantial benefit in a market composed of a wide variety of players, with varying degrees of creditworthiness. Pure weather products are unlikely to result in the delivery of the underlying commodity, so another commodity exchange benefit – standardised delivery terms – is of less significance.
Status of market trading
Authoritative market figures have put the critical mass for an exchange-traded market at around 12 to 14 market makers. WRMA's survey took data from a total of 29 weather market participants, which is held to be a good approximation for the full over-the-counter (OTC) market. However, not all of these are what one would think of as market makers – indeed, only four firms are listed as market makers for the Chicago Mercantile Exchange (CME)'s weather contracts.
The CME lists Heating Degree Day (HDD) and Cooling Degree Day (CDD) futures and options on futures on ten cities in the US, quoted for 12 consecutive calendar months. The contract size, at $100 times the Degree Day Index, is not large compared with OTC deal sizes, and trading is possible on a 24-hour basis via the Globex system. Despite this flexibility, the total number of lots traded in the whole of 2000 was only 67 HDD contracts and 20 CDD contracts. It has been suggested that the reason for the lack of activity was the inflexibility of the product, or conversely the diversity of locations, or simply that the product was launched ahead of its time. Whatever the reason, the numbers for year-to-date 2001 are even worse – respectively, zero and zero.
Meanwhile, European exchanges seem to be in danger of jumping the gun all over again – both the London International Financial Futures and Options Exchange (LIFFE) and the Deutsche Börse section of Eurex have made well-publicised product pre-launches. While evading the question of exactly when their temperature-based contracts will go live, they have both offered free access to historical data that represent the ‘underlying' for their chosen products. In LIFFE's case, that is the monthly mean of the daily average temperature in London, Paris and Berlin. Meanwhile, Eurex has adopted a wider and more traditional approach by publishing HDD and CDD data for 30 cites around Europe, stretching from Dublin to Moscow and Helsinki to Milan. While LIFFE has announced the salient features of its intended product, Eurex will define contract specifications only ‘when it becomes apparent that the market has achieved the required level of maturity'.
However, the feeling from US weather market participants has started to change recently, and there have been a number of comments from major players to the effect that the CME should not abandon its listings as the conditions may soon be right for an exchange-based market.
Whilst the primary market will always be necessary for providing tailored weather risk management products to end-users, a healthy, expanding secondary market is a necessary condition for the effectiveness of this form of risk transfer. By extension, active exchange-traded futures and options markets can deliver transparency and liquidity in the most commoditised elements of that risk. The time of the exchanges may soon be upon us.