Trading Weather Risk
Introduction
As a business, the trading of weather risk has taken off. In the twelve months 2003-2004, per the Price Waterhouse Coopers survey, the total limit of weather transactions executed amounted to $4.7 billion. In the period 2005-2006 this number jumped nearly tenfold to $45.2 billion. This explosive growth has come overwhelmingly from weather risk trading, which now clearly has reached a level of critical mass as a class of commercial activity.
At one level, trading weather risk is an attractive business opportunity because weather essentially is uncorrelated with secular or systemic risk in general financial markets and provides an opportunity for diversification for traders.
- Weather is a physical phenomenon. It is driven by atmospheric factors and its statistical characteristics are independent of human factors. Trends in weather fundamentally have no connection with trends in the stock or bond markets, for example.
- Although weather itself has no underlying value (a difference with traditional financial instruments) it can be traded in ways that are common to trading generally. Weather has volatility, the tails of weather risk may move differently from at-the-money positions, correlations between weather in different areas converge and diverge, etc.
- Pricing methodologies are consistent with general trading market practice for pricing options, although stochastic simulation is used rather than Black-Scholes.
- In addition, portfolios of weather risk can be managed by most of the approaches used in managing portfolios of security and commodity risk: including ratio tests, Greek parameters and Value at Risk measures.
Among WRMA’s membership are several providers of data bases, pricing programs and portfolio management programs. Many of these providers also offer programs for the trading of commodities and financial instruments, demonstrating the degree to which trading in weather has similarities to trading in other fields. On this basis, trading in weather can be viewed as an independent but complimentary trading activity and as such an attractive add-on to a portfolio of traded risk.
At another level trading in weather risk is an attractive proposition because of the relationship between weather and specifically weather-sensitive commodities and financial instruments.
- Weather has obvious interrelations with a variety of agricultural and energy commodities as well as with newer objects of trading activity such as emissions and carbon credits.
- In addition, the financial performance of many corporations is sensitive to weather.
For example, a key component of certain railroads’ business is transporting agricultural produce to markets. Inadequate rainfall or ill-timed precipitation can reduce yields and harvests, which means reduced carloads of agricultural produce, which in turn means less revenue for the railroads. At the extremes, however, precipitation can force traffic from river barges to rail and truck. The rail transportation industry’s exposures to agricultural weather risk accompany the operational and maintenance risks peculiar to railroads due to snow, freeze and rain (washouts, flood).
Trading in weather can be linked to trading in specific commodities or select financial instruments. To a significant degree the one can be a proxy for the other. Combining trading in weather with trading in weather-sensitive commodities expands the possibilities of both activities.
Case Studies
The purpose of these studies is to illustrate the some basic weather trading transactions and weather-commodity trading transactions. Because trading transactions do not exist in isolation the case studies presented below describe circumstances under which those holding weather risk might adopt a particular strategy. Quantification is kept to a minimum, and where present to a relatively unsophisticated level, to facilitate explanation of what often are complex considerations. Far more advanced techniques are present in the market than those shown in the examples.
Although many of the examples are related to temperature, please keep in mind that analogous transactions can be executed on other weather variables, particularly precipitation. With respect to temperature, there is frequent reference to Degree Days (DDs). Heating Degree Days (HDDs) in the United States are 65ºF less the average daily temperature; Cooling Degree Days (CDDs) are the average daily temperature less 65ºF. Over a period of weeks, months or years HDDs and, respectively, CDDs can be totaled to give an aggregate quantitative measure of temperature over time. DDs can be pegged to other standards than 65ºF, and other measures of temperature can be used as well.
Case Study 1 – Difference in Views
Observing the Japanese market a trader notes that the market is trading November HDDs at around the 5 year average (132.5 Celsius HDDs based on 18ºC). Studying recent years’ activity the trader notes that the 10 year average is around 128 Celsius HDDs. Bearing in mind a warming trend in Tokyo (40 Year average around 145 Celsius HDDs), the trader has reason to believe that the market sentiment is distorted by the 2002 and 2004 year temperatures and is unreasonably biased to the cool side. Exercising a basic weather transaction the trader buys a November put with a strike at 132.0 Celsius HDDs. If the trader’s view is correct the put should pay out in his favor and has a reasonable probability of clearing the premium cost.
Case Study 2 - Weather - Correlation
Boston and Chicago O’Hare are heavily traded weather nodes and therefore are subject to market pressures as well as to the vagaries of climate. Winter weather in the two cities is correlated approximately 65%.
A trader may take the view that in the coming year the relationship between Chicago and Boston may de-couple, as it did in 2004. The de-coupling may be based on climatological conditions and on market expectations – i.e. the relative price for risk in these markets may diverge. For example if the expectation is that Boston will show warm while Chicago shows cold the trader may establish a collar at Boston in which the trader goes long warmth and then purchases a call on Chicago HDDs. In the event Boston goes warm the trader will collect on the collar. If, contrary to expectation, Boston runs cold, the trader will have the income from the Chicago call to offset the payment the trader may have to make under the collar.
Case Study 3 - Trading around a Weather Position
A favorable premium level entices a risk taker to write a call on summer temperatures in Phoenix. The call is triggered by total June – September CDDs in excess of 3,400 CDDs. The risk taker’s view of the trends in Phoenix temperature differs from the levels represented in market pricing such that the risk taker in its view can cover its position favorably by means of a swap centered around 3,300 CDDs at 100 CDDs each side of the swap level. Weather warmer than the swap level will finance payments on the call position. Payments to the swap counterparty can be mitigated in part by the premium from the call sold to the call counterparty. The risk taker here has taken the view that Phoenix will be warm. If the risk taker’s view is correct this combination of trades is likely to result in a profit. Under this perspective the downside could be mitigated further by the sale of a CDD put. It also may be managed by other components of the risk taker’s portfolio.
Among the critical factors are the risk taker’s view of the temperature trend, the centering of temperature and the expected direction of weather in the coming season.
Case Study 4 – Weather and Commodity
As gas trader expects that the coming March in a given region is more likely to be cold than warm and goes long gas for the month. The trader hedges this position with an HDD put on a traded weather station in the region. If it is cold, the trader makes significant money on the gas position at the relatively modest cost of the HDD put, netting out to a profit. If March turns out to be warmer than the trader expected, the trader will loose money on the gas position, which will be compensated in whole or in part by the payout from the put.
A weather trader could employ the opposite strategy, going long March cold weather at the weather station while purchasing a put on March gas. If it is cold the weather position will pay out and the trader’s cost is the cost of the gas put. If the weather is warm, the payment from the gas put will cover the weather position in whole or in part, depending on the degree to which the trader structured the hedge.
These four case studies illustrate a small fraction of the possible trading strategies in the weather and weather-commodity trading space. They can be exercised over the counter, on exchanges and through traded indices. They also suggest that successful trading requires a point of view which is based on weather market dynamics, meteorological input or movement in the commodity markets. Specialist intermediaries can assist buyers and sellers balance these factors and their portfolios of exposure. Among WRMA’s membership are several organizations which provide analytic meteorological services.
Notes on the Market
The trading market is made up primarily of monoline weather trading operations, the trading desks of financial institutions and utilities, professional commodity traders and hedge funds. Some insurance companies also are active in the trading market where the market gives them the ability to manage their weather risk portfolios or where there are arbitrage opportunities. Some participants in the trading market also take part in the risk management market. Transactions can be effected over the counter, on exchanges or through publicly traded indices. Intermediaries may play important roles in many transactions.
The Price Waterhouse Coopers survey of the weather risk market, which can be found elsewhere on this site, gives a wide range of basic data about the global weather market.
In the trading market the average trade size runs around $2.0 million to $2.5 million. Most trades are below the $10 million level in notional value. As noted in the PWC survey, a significant portion of the trading activity takes place through the Chicago Merchantile Exchange using relatively standardized exchange trading instruments. Instruments uniquely tailored to particular size, location and duration characteristics are transacted over the counter. In that most weather trading structures are akin to options trading takes place almost exclusively in ISDA swap format in the form of derivatives. Of the weather hazards traded, temperature is the most liquid, followed by precipitation and wind. Less frequently traded hazards are sunshine, growing days and humidity. As is common in many trading areas the trades closer to the money are the most liquid. Liquidity decreases as structures migrate towards the tail of the relevant distribution. Extreme events such as hurricane risk or typhoon wind risk are highly illiquid, although there is a developing crossover into catastrophe insurance markets which may ease liquidity somewhat. Insurers who are able to manage tail weather risk in conjunction with their portfolios of natural catastrophe risk, thereby achieving diversity of tail risk, may have an advantage in this particular risk space. With regard to structural characteristics, aggregate structures tend to be more liquid than event risk structures.
The Price Waterhouse Coopers survey shows that the weather market, although growing impressively, is still significantly smaller, with less depth and breadth, than most commodity markets. With a view to Case Study 4 above, it is more likely, under present conditions, that weather-commodity trading example would have a more profound impact strategically on the weather trader’s book than on the gas trader’s book, although the potential for tactical benefit to the gas trader in dealing in the weather market is strong.
Weather risk trading is statistically driven and puts a high value on pricing, risk management and portfolio management as well as on expense management. Value at Risk required and returns are a function of the types of risk incorporated into the portfolio. Tail risks, particularly extreme tail risks such as windstorm, consume significant amounts of VaR, although they can offer higher returns. Substantial books of business can be supported on a foundation of relatively modest VaR levels and may yield attractive returns provided portfolio management, pricing, credit assessment and marketing are carefully coordinated. Trading in weather risk requires many of the same disciplines and is subject to many of the same regulatory disciplines as trading in other risk space. As with all financial transactions the possibility of loss is present as well as the possibility of gain.
From the point of view of the market, insurers, banks and monolines are in an ideal position to act as aggregators of weather risk. They have the ability to manage weather risk in conjunction with their other pools of risk (e.g. natural catastrophe, credit), to warehouse risk or to restructure risk, repackage it and release it into the financial markets, starting with the weather risk trading market. In this respect the weather risk trading market has the potential to follow the path struck by other innovative markets, such as credit and mortgage risk.