Storm guard
As the world's energy markets have deregulated it has become clear
that major energy projects demand increasingly sophisticated and
innovative risk management. The market environment under which
energy assets are used and energy companies raise investments for
new projects are becoming more challenging and therefore
innovative. This is particularly true of the power industry. The
commoditisation of electricity has created new merchant capacity
with new challenges for both sponsors and bankers.
The changing environment
In the US, most regulatory regimes allowed energy companies to pass on the changing energy feedstocks and operating costs to their customer in the form of rates. This meant that customers were essentially bearing some of the costs of generation assets and were providing a component of the risk capital needed to build generation capacity. The banks were therefore happy with high leverage and high debt ratings, assured that the source of cashflow for debt servicing could be adjusted by simply charging higher rates to customers.
In a deregulated environment however, the situation is not as predictable. Merchant power plants (MPPs) which can be defined as power generation facilities for which the source of cashflow is not covered by long-term offtake agreement are beginning to increase their share of capacity. In the medium term it is expected that merchant generation will be the standard in those countries introducing competition in the form of a wholesale or pool electricity market.
Where traditional IPPs with long term take or pay contracts can lock the spread between production costs and electricity prices, few merchant power plants will be able to allocate all offtake, price or financial risk to third parties. This poses problems for lenders. Without long term Power Purchase Agreements (PPAs) lenders are asked to take on market risk when financing these plants. The lack of isolation from market forces makes long term viability of the project, in other words the source of cash flow for debt servicing and return on investment obligations, central. This cash flow volatility resulting from interrelated markets variables then places the focus of the credit analysis on the project's income and cash flow, emphasising the sale of future output from the project and the security of such sales. The rationale then becomes obvious as to why forward sales of output must have adequate hedging to assure lenders and how such assurances can increase debt capacity and leverage.
The use of derivatives to mitigate price risk has been a key mechanism in the evolution of risk management practices in many new industries. In the last two years or so, a new area of risk is being tackled by derivatives: the mitigation of volume risk related to the effects of weather. Weather derivatives fills in a gap left by traditional risk management solutions, by addressing the uncatastrophic and seasonal weather related volume fluctuations. Demand for electricity from a generator is very well correlated to the weather, creating a volume risk by making a significant impact on the cash flow.
It therefore comes as no surprise that weather derivatives made their debut in the United States in 1997, when the US power community realised that it could insulate itself from negative shifts in volume risk. World wide there have been some 3,500 weather deals, representing a notional risk value of around $5 billion. The first publicly known European weather contract was signed in August 1998. The latest deal was a swap between TXU and a counter-party, whose name remains undisclosed. It is believed that more than 100 deals have been done in Europe so far, by companies wishing not only to hedge their weather risks, but also to familiarise themselves with the products, so that they can take advantage of future opportunities.
Weather derivatives
Weather derivatives are conceptually very simple. The weather is indexed ? measured in a quantifiable way like the inches of rain that fall in region X, during period Z ? and a formulated payout is based upon this chosen index. For instance $5000 per inch of rain up to a maximum of $200,000. The structures are built from the same financial building blocks that most derivatives are based upon and can be sold and bought as puts and calls or traded as swaps and collars. There are also many other variations of exotic structures that can be tailored to suit particular needs.
Contracts
The contracts are based on the ISDA (International Swaps and Derivatives Association) master agreement and defined by their:
? Location ? city (or basket of cities) in which transaction takes place and measurement are taken
? Type ? Cooling Degree Days or Heating Degree Days, a precipitation, a wind-chill factor
? Start date & finish date ? duration
? Strike ? can be anything the parties agree
? Premium ? upfront cost of the hedge (for puts, calls etc.)
? Notional ? this is the $ amount per DD (it could also be rain, wind...)
? Cap ? maximum pay out of the deal
There are many different weather contract indices, but since the utility industry has been the most active within this market the most popular index is temperature or degree-days. A degree-day is the difference between 65 degrees Fahrenheit and the average temperature of the day. This value is theoretically the temperature at which users turn on their heating or air-conditioning. If the average temperature of the day is above 65F, say 75F, this difference is called cooling degree-days (CDD). If the temperature is below 65F say 55F, the difference is called heating degree-days (HDD). For example, 30 days at an average of 55F would cumulate to 300 HDDs during that month.
The contracts also specify a designated weather station at which measurement should be taken. In the US for example this could be #WBAN: 14739, in the UK it could be London Heathrow weather station #IWMO 03772. There will also be a second back up station.
The contracts are normally written for a season, say the November to March winter season, but have also been written for a number of seasons over a number of years. A strike or the agreed upon value of the index, is the point at which compensation flows from one party to another. If a contract is for the winter season, the average HDDs could be 500 HDDs per month or 500 X 5 months = 2500 HDDs. And the strike, could be one standard deviation away from this datum.
Contracts concluded vary in size between $1,000 per degree-day notional with a limit of $2 million to deals with notional values of $50,000 with a cap of $20 million. The average deal size is around the $5,000 notional with a $2 million cap.
Within eight days of the contract running out the parties are paid. Three days is generally made available after the last date of the contract for weather information to be made available to both parties and a further 5 days for payment to change hands. In the UK the weather data is checked and reviewed by the UK Meteorological Office and the quality reassessed which can take up to 95 days. At this time a review of the settlement is made and any adjustments needed are made.
The various products
As mentioned the products can be in the form of put or call options, swaps and collars. The weather derivatives providers charge a premium or a fee for the put and call options contracts that they provide. This can however be avoided by using a structure such as a swap or costless collar in which no initial payment changes hands. The weather market's favourite structures are collars and swaps as they offset the premium costs of buying an option.
Put and Call Options
A Degree day put option places a minimum on the number of DDs for example, ensuring that the buyer is paid for every DD below the specified floor (see diagram 1). If at the expiration of a put option contract, the strike price ? a preagreed number of degree-days ? is higher than the actual number of HDD or CDDs, the seller of the option pays the buyer a certain amount. This amount is equal to the strike level less the actual number of degree-days for the period, multiplied by a pre-arranged dollar/pound amount. If the strike is lower than the actual DDs then the seller pays nothing and retains the premium.
If at the expiration of a call option contract the strike price is lower than the actual number of HDD or CDDs, the seller pays the buyer a certain amount. This amount is equal to the actual number of degree days less the strike level of degree-days multiplied by a pre-arranged dollar amount. If the strike is higher than actual DDs the seller pays nothing and retains the premium.
Swaps and Collars
A DD swap is a structure in which a DD value is agreed on, one side of which the buyer gets paid by the seller and on the other side of which the seller gets paid by the buyer (see diagram 2). The seller is compensated pro rata per DD whenever DDs settle below an agreed strike level (in the following diagram it is set at 1700 CDDs). The buyer of the swap sees a mirror effect.
The same in principle applies to a collar in which the buyer pays the seller per degree day and the seller pays the buyer per DD, the difference with a swap being that the compensation does not occur form one party to another at the same DD value or strike. In diagram 3, the compensation begins to flow at 1800 HDDs or 1700 HDDs.
Expanding market
The more mature US derivatives market is estimated to house around 70 players including heating oil marketers, power marketers, utilities, re-insurers, insurance companies, banks, brokers and other financial institutions. The three main types of market makers and weather derivative providers are energy majors such as Enron, insurers and re-insurers such as Swiss Re and banks such as Societe Generale.
An area of expansion is exchange-traded contracts, and Internet portals offering weather contracts, such as the Chicago Mercantile Exchange, I-Wex and Swiss Re on-line.
Weather derivatives in project finance
1999 and early 2000 saw a significant number of financing transactions with either fully merchant or semi-merchant characteristics and leverage from 50-65% to 85%. In the US the main merchant financing funding sources have been the banks for the short term and the capital markets in the long term.
Weather risk management can be put in use in a number of distinct areas, some more directly applicable to project finance than others:
1 ? cash flow predictability ? the rationale would run as follows:
? The loss of PPAs and take or pay agreements has created merchant risk,which is passed on to the source of funding.
? These agreements created predictable commodity volumes and therefore cash flows improving the credit profile and debt capacity of the project leading to increased leverage.
? One of the most important variables altering the dynamics of supply and demand in many commodity markets, especially electricity and gas, is the seasonal fluctuation of weather.
? By dampening or hedging the effects of weather on volume, predictability in project income can be improved, which should assure lenders allowing better ratings and higher debt capacity.
This is explored further below.
2 ? weather derivatives perhaps transformed into insurance products can be used to insure progress of physical work on the construction of the second or third phase or during the initial short term financing package of Greenfield projects. By compensating sponsors in case of delays due to bad weather the sponsor guarantees no additional burden is placed on contingent sources for existing debt payments.
3 ? Initial capital costs can be partially financed by the sale of a weather derivative. A power company could finance the purchase of a peaking plant for example by selling a maximum temperature call option struck at a temperature that is likely to produce high profits for a peaking plant, say at 100 degrees. A maximum temperature call option is a digital structure in which the seller pays a set amount say, $1m, when a maximum strike temperature is exceeded. The option seller can then use the option premium collected to help finance the project. The above suggestion however is ambitious and may not suit some sponsors and stand alone projects.
Weather derivatives were developed to mitigate volume risk of existing assets, by ensuring that either a plant is dispatched or the resulting cash flow, which would have been generated, is present. PPAs which are used as assurance to lenders, essentially hedge forward output. In the oil industry exploration and production companies are more familiar with selling forward production volumes in order to assure lenders and attract more debt funding. The volumetric production payment (VPP) is in essence a forward sales contract tied to a loan, with legal assurances and safeguards for the lenders.
Owners of merchant power plants face a very similar problem, there is no reason why the important assurance of stable cash flow from the project can not be met by a weather hedge as a part of a wider embedded hedging structure. There are however some differences between PPA's and forward/derivatives that have to be noted:
For a commodity the hedging instrument used depends on the economics of the production model. For a low cost production facility a forward contract is often the most efficient hedging method. In a situation where production is not stopped by low prices, forward sales usually incur lower transaction costs than options.
On the other hand, for a high cost production facility, the production function can be described as an option. This analogy of a plant as an option works where the underlying price is the market price of the output, the strike price is the marginal cost of generation and the option premium is the fixed cost of owning the plant. A peaking power plant for example, can be considered as a physical call option on temperature, here as temperature increases the demand for electricity increase to a point where it is profitable to dispatch the plant. A higher cost production model therefore has economic characteristics similar to options and is most efficiently hedged in the options market. This hedging can therefore take place as price or volume hedging or a combination of both.
The use of derivatives in energy project finance has been limited to oil and gas markets so far. But almost all the highly leveraged projects that have gone through in the power markets have had some degree of risk hedging. The primary risk hedged so far from the lenders perspective has been the price curve. Will actual prices be as projected? Very detailed supply and demand models try and resolve the validity of price projections, but to do so with sufficient margin of error to avoid the risk of default in bad times is difficult. This margin of error can be reduced by hedging of volumes. Clearly price hedging can reduce the impact of price risk on a project. The overall impact of price hedging though depends on the certainty of the volume of the product. If volumes are very uncertain then the effect of price hedging is smaller. Consider the diagram below.
A hedging strategy that encompasses both price and volume leads to a much narrower range of possibilities in cash flow. The reduced range of cash flows has a key consequence from a project finance point of view, which becomes evident on default probabilities. To get projects debt rated, agencies evaluate the cashflow forecasts. On the basis of the probability of the default implicit in the free cashflow, an anticipated rating of debt can be determined.
The project may therefore support more debt funding in the first few years, possibly increasing gains from leverage. Higher leverage may lead to higher equity returns because less equity is needed. Many developers seek to minimise hedging given a fixed leverage ratio, because it leaves the most room for upside in equity returns.
Despite this, the problem for some developers wanting to hedge in the derivative markets may be the tenor of the instruments. In the weather market many OTC marketers have a very flexible approach to weather contracts which can be one way of overcoming the problem. The longer deals done have been for two or three years with the longest deal on record spanning 10 years. n
The authors would like to acknowledge contributions from the Weather Risk Advisory based in Cambridge, England.
? EJC Energy, a London-based consultancy, has published a report on weather derivatives sponsored by Enron Europe and the Chicago Mercantile Exchange. Please contact Ali Tahghighi or Philippe Carpentier on +44 (0) 207 287 8605 or e-mail: alit@ejc.co.uk
? Weather Risk Advisory, a Cambridge-based advisory specialises in weather derivatives and is developing a range of proprietary software for this market. Please contact Peter Brewer on +44 (0) 1954 206250 or email: info@WeatherRiskAdvisory.com
The changing environment
In the US, most regulatory regimes allowed energy companies to pass on the changing energy feedstocks and operating costs to their customer in the form of rates. This meant that customers were essentially bearing some of the costs of generation assets and were providing a component of the risk capital needed to build generation capacity. The banks were therefore happy with high leverage and high debt ratings, assured that the source of cashflow for debt servicing could be adjusted by simply charging higher rates to customers.
In a deregulated environment however, the situation is not as predictable. Merchant power plants (MPPs) which can be defined as power generation facilities for which the source of cashflow is not covered by long-term offtake agreement are beginning to increase their share of capacity. In the medium term it is expected that merchant generation will be the standard in those countries introducing competition in the form of a wholesale or pool electricity market.
Where traditional IPPs with long term take or pay contracts can lock the spread between production costs and electricity prices, few merchant power plants will be able to allocate all offtake, price or financial risk to third parties. This poses problems for lenders. Without long term Power Purchase Agreements (PPAs) lenders are asked to take on market risk when financing these plants. The lack of isolation from market forces makes long term viability of the project, in other words the source of cash flow for debt servicing and return on investment obligations, central. This cash flow volatility resulting from interrelated markets variables then places the focus of the credit analysis on the project's income and cash flow, emphasising the sale of future output from the project and the security of such sales. The rationale then becomes obvious as to why forward sales of output must have adequate hedging to assure lenders and how such assurances can increase debt capacity and leverage.
The use of derivatives to mitigate price risk has been a key mechanism in the evolution of risk management practices in many new industries. In the last two years or so, a new area of risk is being tackled by derivatives: the mitigation of volume risk related to the effects of weather. Weather derivatives fills in a gap left by traditional risk management solutions, by addressing the uncatastrophic and seasonal weather related volume fluctuations. Demand for electricity from a generator is very well correlated to the weather, creating a volume risk by making a significant impact on the cash flow.
It therefore comes as no surprise that weather derivatives made their debut in the United States in 1997, when the US power community realised that it could insulate itself from negative shifts in volume risk. World wide there have been some 3,500 weather deals, representing a notional risk value of around $5 billion. The first publicly known European weather contract was signed in August 1998. The latest deal was a swap between TXU and a counter-party, whose name remains undisclosed. It is believed that more than 100 deals have been done in Europe so far, by companies wishing not only to hedge their weather risks, but also to familiarise themselves with the products, so that they can take advantage of future opportunities.
Weather derivatives
Weather derivatives are conceptually very simple. The weather is indexed ? measured in a quantifiable way like the inches of rain that fall in region X, during period Z ? and a formulated payout is based upon this chosen index. For instance $5000 per inch of rain up to a maximum of $200,000. The structures are built from the same financial building blocks that most derivatives are based upon and can be sold and bought as puts and calls or traded as swaps and collars. There are also many other variations of exotic structures that can be tailored to suit particular needs.
Contracts
The contracts are based on the ISDA (International Swaps and Derivatives Association) master agreement and defined by their:
? Location ? city (or basket of cities) in which transaction takes place and measurement are taken
? Type ? Cooling Degree Days or Heating Degree Days, a precipitation, a wind-chill factor
? Start date & finish date ? duration
? Strike ? can be anything the parties agree
? Premium ? upfront cost of the hedge (for puts, calls etc.)
? Notional ? this is the $ amount per DD (it could also be rain, wind...)
? Cap ? maximum pay out of the deal
There are many different weather contract indices, but since the utility industry has been the most active within this market the most popular index is temperature or degree-days. A degree-day is the difference between 65 degrees Fahrenheit and the average temperature of the day. This value is theoretically the temperature at which users turn on their heating or air-conditioning. If the average temperature of the day is above 65F, say 75F, this difference is called cooling degree-days (CDD). If the temperature is below 65F say 55F, the difference is called heating degree-days (HDD). For example, 30 days at an average of 55F would cumulate to 300 HDDs during that month.
The contracts also specify a designated weather station at which measurement should be taken. In the US for example this could be #WBAN: 14739, in the UK it could be London Heathrow weather station #IWMO 03772. There will also be a second back up station.
The contracts are normally written for a season, say the November to March winter season, but have also been written for a number of seasons over a number of years. A strike or the agreed upon value of the index, is the point at which compensation flows from one party to another. If a contract is for the winter season, the average HDDs could be 500 HDDs per month or 500 X 5 months = 2500 HDDs. And the strike, could be one standard deviation away from this datum.
Contracts concluded vary in size between $1,000 per degree-day notional with a limit of $2 million to deals with notional values of $50,000 with a cap of $20 million. The average deal size is around the $5,000 notional with a $2 million cap.
Within eight days of the contract running out the parties are paid. Three days is generally made available after the last date of the contract for weather information to be made available to both parties and a further 5 days for payment to change hands. In the UK the weather data is checked and reviewed by the UK Meteorological Office and the quality reassessed which can take up to 95 days. At this time a review of the settlement is made and any adjustments needed are made.
The various products
As mentioned the products can be in the form of put or call options, swaps and collars. The weather derivatives providers charge a premium or a fee for the put and call options contracts that they provide. This can however be avoided by using a structure such as a swap or costless collar in which no initial payment changes hands. The weather market's favourite structures are collars and swaps as they offset the premium costs of buying an option.
Put and Call Options
A Degree day put option places a minimum on the number of DDs for example, ensuring that the buyer is paid for every DD below the specified floor (see diagram 1). If at the expiration of a put option contract, the strike price ? a preagreed number of degree-days ? is higher than the actual number of HDD or CDDs, the seller of the option pays the buyer a certain amount. This amount is equal to the strike level less the actual number of degree-days for the period, multiplied by a pre-arranged dollar/pound amount. If the strike is lower than the actual DDs then the seller pays nothing and retains the premium.
If at the expiration of a call option contract the strike price is lower than the actual number of HDD or CDDs, the seller pays the buyer a certain amount. This amount is equal to the actual number of degree days less the strike level of degree-days multiplied by a pre-arranged dollar amount. If the strike is higher than actual DDs the seller pays nothing and retains the premium.
Swaps and Collars
A DD swap is a structure in which a DD value is agreed on, one side of which the buyer gets paid by the seller and on the other side of which the seller gets paid by the buyer (see diagram 2). The seller is compensated pro rata per DD whenever DDs settle below an agreed strike level (in the following diagram it is set at 1700 CDDs). The buyer of the swap sees a mirror effect.
The same in principle applies to a collar in which the buyer pays the seller per degree day and the seller pays the buyer per DD, the difference with a swap being that the compensation does not occur form one party to another at the same DD value or strike. In diagram 3, the compensation begins to flow at 1800 HDDs or 1700 HDDs.
Expanding market
The more mature US derivatives market is estimated to house around 70 players including heating oil marketers, power marketers, utilities, re-insurers, insurance companies, banks, brokers and other financial institutions. The three main types of market makers and weather derivative providers are energy majors such as Enron, insurers and re-insurers such as Swiss Re and banks such as Societe Generale.
An area of expansion is exchange-traded contracts, and Internet portals offering weather contracts, such as the Chicago Mercantile Exchange, I-Wex and Swiss Re on-line.
Weather derivatives in project finance
1999 and early 2000 saw a significant number of financing transactions with either fully merchant or semi-merchant characteristics and leverage from 50-65% to 85%. In the US the main merchant financing funding sources have been the banks for the short term and the capital markets in the long term.
Weather risk management can be put in use in a number of distinct areas, some more directly applicable to project finance than others:
1 ? cash flow predictability ? the rationale would run as follows:
? The loss of PPAs and take or pay agreements has created merchant risk,which is passed on to the source of funding.
? These agreements created predictable commodity volumes and therefore cash flows improving the credit profile and debt capacity of the project leading to increased leverage.
? One of the most important variables altering the dynamics of supply and demand in many commodity markets, especially electricity and gas, is the seasonal fluctuation of weather.
? By dampening or hedging the effects of weather on volume, predictability in project income can be improved, which should assure lenders allowing better ratings and higher debt capacity.
This is explored further below.
2 ? weather derivatives perhaps transformed into insurance products can be used to insure progress of physical work on the construction of the second or third phase or during the initial short term financing package of Greenfield projects. By compensating sponsors in case of delays due to bad weather the sponsor guarantees no additional burden is placed on contingent sources for existing debt payments.
3 ? Initial capital costs can be partially financed by the sale of a weather derivative. A power company could finance the purchase of a peaking plant for example by selling a maximum temperature call option struck at a temperature that is likely to produce high profits for a peaking plant, say at 100 degrees. A maximum temperature call option is a digital structure in which the seller pays a set amount say, $1m, when a maximum strike temperature is exceeded. The option seller can then use the option premium collected to help finance the project. The above suggestion however is ambitious and may not suit some sponsors and stand alone projects.
Weather derivatives were developed to mitigate volume risk of existing assets, by ensuring that either a plant is dispatched or the resulting cash flow, which would have been generated, is present. PPAs which are used as assurance to lenders, essentially hedge forward output. In the oil industry exploration and production companies are more familiar with selling forward production volumes in order to assure lenders and attract more debt funding. The volumetric production payment (VPP) is in essence a forward sales contract tied to a loan, with legal assurances and safeguards for the lenders.
Owners of merchant power plants face a very similar problem, there is no reason why the important assurance of stable cash flow from the project can not be met by a weather hedge as a part of a wider embedded hedging structure. There are however some differences between PPA's and forward/derivatives that have to be noted:
For a commodity the hedging instrument used depends on the economics of the production model. For a low cost production facility a forward contract is often the most efficient hedging method. In a situation where production is not stopped by low prices, forward sales usually incur lower transaction costs than options.
On the other hand, for a high cost production facility, the production function can be described as an option. This analogy of a plant as an option works where the underlying price is the market price of the output, the strike price is the marginal cost of generation and the option premium is the fixed cost of owning the plant. A peaking power plant for example, can be considered as a physical call option on temperature, here as temperature increases the demand for electricity increase to a point where it is profitable to dispatch the plant. A higher cost production model therefore has economic characteristics similar to options and is most efficiently hedged in the options market. This hedging can therefore take place as price or volume hedging or a combination of both.
The use of derivatives in energy project finance has been limited to oil and gas markets so far. But almost all the highly leveraged projects that have gone through in the power markets have had some degree of risk hedging. The primary risk hedged so far from the lenders perspective has been the price curve. Will actual prices be as projected? Very detailed supply and demand models try and resolve the validity of price projections, but to do so with sufficient margin of error to avoid the risk of default in bad times is difficult. This margin of error can be reduced by hedging of volumes. Clearly price hedging can reduce the impact of price risk on a project. The overall impact of price hedging though depends on the certainty of the volume of the product. If volumes are very uncertain then the effect of price hedging is smaller. Consider the diagram below.
A hedging strategy that encompasses both price and volume leads to a much narrower range of possibilities in cash flow. The reduced range of cash flows has a key consequence from a project finance point of view, which becomes evident on default probabilities. To get projects debt rated, agencies evaluate the cashflow forecasts. On the basis of the probability of the default implicit in the free cashflow, an anticipated rating of debt can be determined.
The project may therefore support more debt funding in the first few years, possibly increasing gains from leverage. Higher leverage may lead to higher equity returns because less equity is needed. Many developers seek to minimise hedging given a fixed leverage ratio, because it leaves the most room for upside in equity returns.
Despite this, the problem for some developers wanting to hedge in the derivative markets may be the tenor of the instruments. In the weather market many OTC marketers have a very flexible approach to weather contracts which can be one way of overcoming the problem. The longer deals done have been for two or three years with the longest deal on record spanning 10 years. n
The authors would like to acknowledge contributions from the Weather Risk Advisory based in Cambridge, England.
? EJC Energy, a London-based consultancy, has published a report on weather derivatives sponsored by Enron Europe and the Chicago Mercantile Exchange. Please contact Ali Tahghighi or Philippe Carpentier on +44 (0) 207 287 8605 or e-mail: alit@ejc.co.uk
? Weather Risk Advisory, a Cambridge-based advisory specialises in weather derivatives and is developing a range of proprietary software for this market. Please contact Peter Brewer on +44 (0) 1954 206250 or email: info@WeatherRiskAdvisory.com
Request a Demo
Interested in IJGlobal? Request a demo to discuss a trial with a member of our team. Talk to the team to explore the value of our asset and transaction databases, our market-leading news, league tables and much more.