Diapositive 1

1
Weather Derivatives
By Majdala Khater and Céline Levy
2
Table of Contents

• Weather derivatives overview
• Weather derivatives markets
• Weather derivatives structure
• Pricing weather derivatives

3
Weather Derivatives Usefulness

• Most industries in the world are directly or
  indirectly affected by the unexpected weather
  events
• Weather risk affects a great number of businesses
  all around the world
• In fact, a survey has proved that more than 200
  major US companies suffered from huge losses due
  to unexpected weather change events
• Weather derivatives have been introduced to the
  financial market as a protection or insurance
  against potential losses due to weather changes
• Weather derivatives concern not only the energy
  market but also the tourism, the transportation
  and the agricultural industry.

4
Weather Derivatives what are they?

• Financial risk management tools used by companies
  and enterprises to hedge against the risk of
  weather changes
• they cover the high probability for unexpected
  events such as a dryer or freezer weather, while
  an insurance covers the high probability events
  such as hurricanes...
• Purpose
• Stabilize revenues that may be affected by severe
  change in weather
• Control Pricing and Volume Risk
• Enhance financial portfolios
• Competitive advantage versus the simple
  insurance
• No moral hazard

5
Weather Derivatives limitations

• Increasing interest in weather risk management
  but
• volume of trade in weather derivatives have been
  growing slowly
• Lack of liquidity
• The weather derivatives market needs to be more
  developed and potential research opportunities
  are needed to be analysed
• Capital pricing model assumptions are not
  realistic for weather-linked financial
  instruments

6
Weather derivatives markets

• One of the most exposed sector power industry
• 730,087 contracts traded worldwide from April
  2006 to March 2007
• Total value of contracts on temperature in OTC
  markets and CME 18.9 billion
• Rain 142 million
• Wind 36 million

7
Volume of contracts traded worldwide
8
Exchange traded market

• At the beginning swaps between companies like
  Enron Koch Energy
• Two main markets
• Chicago Mercantile Exchange (CME)
• London International Financial Futures and
  Options Exchange (LIFFE)

9
CME

• 10 cities in the US
• Reference temperature 65F (18.33C)
• Heating Degree Days (HDD) October to April
• Cooling Degree Days (CDD) May to September
• Arithmetic mean between highest and lowest
  temperature
• Contract size 100, a tick being 1F

10
Euronext - LIFFE

• London, Paris, Berlin
• London contracts of 3,000
• Paris and Berlin 3,000
• Tick of 0.01C ? 30 or 30 a tick

11
Nextweather

• Météo-France Euronexxt
• National and regional (5 regions)
• Temperatures reported daily and quarterly

12
Weather derivative structure

• Most of weather derivatives traded are call or
  put options, swaps, collars
• Weather futures contract
• Provides or requires payments according to the
  level of a weather index
• No initial premium is required
• The payment of a HDD-Future (Heating Degree Days)
  contract is
• F-NHDD
• F future price (amount of money)
• N contract size (affecting a financial value)
• For a CME temperature contract N100/degree
  

13
Weather derivative structure call and put options

• The underlying assets of a call or a put option
  are either HDD or CDD (Cooling Degree Days)
• A dollar amount is associated with every degree
• In order to limit the maximum payout by the
  counterparties, the contracts are usually capped
• For a call option Payoff P(/DD)Max(ST-X,0)
• For a put option Payoff P(/DD)Max(X-ST,0)
• Example
• Consider a CDD call option with a strike price of
  1000CDDs paying 4000 per degree day.
• Payoff 4000 Max (CDDt-1000, 0)
• CDDt is the cumulative cooling degree days over
  the life of the contract

14
Weather derivative structure swaps

• Combination of a call and put option having the
  same strike price and the same underlying
  location
• The simplest kinds uncapped swaps
• The payoff is calculated
• Payoff MinP(/DD)Max(ST-X,0),
  h-MinP(/DD)Max(X-ST,0), h
• Example
• An icemaker and a cinemas operator want to cover
  respectively against a fresh summer and warm
  summer. They put in place a swap in which the
  exercise price is 24 C. If the average
  temperature over the period agreed exceeds 24 C,
  the producer of ice pay to the operator of
  cinemas the provision in the contract

15
Weather derivative structure collars

• Purchase an OTM put (call) with a particular
  strike price (K2) and sell in the same time an
  OTM call (put) with different strike price (K1).
• Payoff MinP(/DD)Max(ST-K1,0),
  h-MinP(/DD)Max(K2-ST,0), h
• Example 
• A company who wish to protect itself against a
  hard winter will buy a HDD Cap having the
  following characteristics Strike 500HDD with
  2000 per HDD and a payoff capped to 2 000 000.
• Payoff CapMin (2000Max (0,CumHDD-500),2
  000 000)
• PL Payoff Cap-premium for buying the cap

16
Weather derivative Pricing

• The no-arbitrage option pricing model is not
  practical pricing tool for weather derivatives
• the underlynig weather indexes are not a tradable
  instruments
• The underlying weather indexes are not
  stationary
• Hard to implement pricing techniques
• Historical data are characterized by high degree
  of autocorrelation
• reduces the number of independent observations

17
Weather derivative Pricing Simple option pricing

• This model can be constructed using a probability
  distribution that will fit the historical data
  collected of monthly CDDs or HDDs
• The expected payoff of a CDD option can be
  calculated using the formula
• Payoff (CDD) M
  P(CDD)Q(CDD)d(CDD)
• The expected value of a weather derivative
  depends on
• the strike price
• the probability distribution that describes the
  CDDs
• the number of dollar per CDD

18
Weather derivative Pricing

• Many complications encounter the pricing of
  weather derivatives
• A simple distribution should not explain the
  historical data.
• Variability in weather trends
• Data given by the atmospheric community cant be
  used directly

19
Weather derivatives PricingBlack and Scholes

• The BS model shows that, under certain
  assumptions, the price of options can be
  determined from the price of the underlying asset
  
• The asset can be traded continuously with no
  transaction costs
• The asset price follows geometric Brownian motion
• The asset is uninfluenced by the trading of the
  asset that is undertaken to hedge the option
• No riskless arbitrage
• Some of these assumptions do not probably fit the
  weather derivative because
• The underlying instruments in weather derivatives
  is not tradable
• Its also impossible to create a risk-neutral
  portfolio

20
Weather derivatives Pricingthe actuarial method

• The average final payment EM is calculated using
  the expected value of the function fp of the
  contract payoff under the measure of the
  probability chosen
• EM ?R fp(x)g(x)dx
• The actuarial method is much more robust than the
  Burn analysis

21
Weather derivatives PricingBurn Analysis

• Called simulation on historical data, the burn
  analysis answers the question what would have
  been the average option payoff in the past n
  years
• There are 6 steps to be considered in a Burn
  analysis process
• Collecting the historical data
• Converting them to degree day either HDD or CDD
• Some corrections should be made to the data
  converted.
• Then, for every year in the past and for each
  weather pattern, we determine the payout of the
  option
• Find the average of these payout amounts
• Discount back to the settlement date.
• Burn analysis formula consists on estimating an
  average final payment based on historical data EM
  and risk hedge standard deviation sM.
• Pd(EM?sM)

Where ? positive constant
d discount factor
22
Weather derivatives PricingPruning Analysis

• The Pruning technique consists of integrating
  into daily simulation models various weather
  forecasts
• The integration of climate forecast is slightly
  more difficult, because it requires the
  calculation of conditional probability
  distributions
• Solution use of historical scenarios for each
  forecast

23
Conclusion

• The weather derivatives efficiently meet the
  specific needs of coverage
• Climate uncertainty reamins high despite improved
  weather forecasts
• Existing pricing models described are not
  necessarily used by market makers
• Need for a standard pricing model

24
Sources

• La lettre mensuelle dEuronext Paris - Février
  2002 - N 41 - Bourse Information
• Introduction to Weather Derivatives
• de Geoffrey Considine, Ph.D, Weather Derivatives
  Group, Aquila Energy
•  Weather Derivatives Instruments and Pricing
  Issues
• Mark Garman', Carlos Blanco and Robert Erickson
• Quel avenir pour les dérivés climatiques en
  Europe ?
• Phillipe Smadja et André Yves Ponts
•  Evaluation des dérivés climatiques , Michael
  Moreno
•  Options, futures and other derivatives 6ème
  edition, JOHN C. HULL
• La Gestion du Risque Climatique 
• Didier Marteau, Jean CARLE, Stéphane Fourneaux,
  Ralph Holz, Michael Moreno

25
Sources

• Websites
• www.edubourse.com
• www.artemis.bm/html/press_releases.
• www.matif.fr/monep/pub/NextweatherbrochureFR
• http//www.lesinfos.com/news17993.html
• http//www.novethic.fr
• www.bfinance.fr
• La Grande Relève gt  Articles gt  N 1070 -
  novembre 2006 gt Énergie et climat
• http//www.atmos.washington.edu
• http//www.derivativesstrategy.com

26
Thank you for your attention!!