Weather Derivatives necessity, methods and application

1
Weather Derivativesnecessity, methods and
application

• Reinhard Hagenbrock
• Seminar of the
• working group on Climate Dynamics
• Bonn, 16. Mai 2003

2
Outline

• History
• What is a weather derivative?
• Idealised example
• Market players, -places and requirements
• Use of meteorology
• Summary and outlook

3
History of weather derivatives

• risk management of weather risks has always
  been part of insurance business
• storms, crop failure, floods, ...
• accident caused by weather extremes is insured
• starting point for weather derivatives
  dependency of profit on weather
• approx. 20 of business activities in western
  economies (partly) dependent on weather

4
History of weather derivatives

• price risk
• higher acquisition prices (e.g. for crop)
• higher energy consumption
• extra costs (e.g. for irrigation)
• volume risk
• in the production (e.g. agriculture)
• in the sales (e.g. ice cream)

5
History of weather derivatives

• Price risks may generally be managed with options
  / long term contracts
• weather risk generally is a volume risk, price
  risk should by managed independently

6
History of weather derivatives

• Starting of weather derivatives dependency of
  energy sales on temperature

7
History of weather derivatives

• First weather derivative Sep 1997 between two
  energy suppliers
• aim to balance electricity sales caused by
  temperature fluctuations in winter 1997/98
• concept seemed simple, benefit obvious
• new, exotic derivatives dealt at Chicago
  Mercantile Exchange since Sep. 1999

8
What is a weather derivative?

• ... derivative financial instrument in which
  meteorological data - e.g. temperature - is used
  as a basis product
• Degree-Day
• Heating Degree Day HDD(t) max(65-T(t),0)
• Cooling Degree Day CDD(t) min(T(t)-65,0)
• usually summed up over a month/season
• sometimes DD with other reference temperatures,
  average temperature

9
What is a weather derivative?

• Other indices
• precipitation
• Indices are dealt like goods

10
What is a weather derivative?

• 70-80 of the weather derivative deals are
  options
• Put pay at end of contract if index is small
  P T ? min((max(X-V),0),C)
• T tick size or notional, e.g. 100 /HDD
• V value of index at end of contract
• X strike of the option
• C cap-strike upper limit of pay
• Call counterpart to Put

11
What is a weather derivative?

• Swaps Interchange between Put and Call, no
  premium
• more complex contracts Collars, spreads to
  chose appropriate chance/risk balance
• other contracts
• hybrid contracts
• non-linear pay function
• critical-day contracts

12
What is a weather derivative?

• Differences between weather insurance and weather
  derivative
• proof of damage
• no strict link between index value and damage
• trade with contracts in a secondary market
• standardised contracts
• differences in accounting and fiscal aspects

13
What is a weather derivative?

• Multitude of derivatives
• Location of measurement (USA 10, Xelsius 30)
• Type of asset (HDD, CDD, precipitation, )
• Strike
• Time period
• Tick size

14
Idealised example

• Risk analysis
• Electricity Enterprises finds out electricity
  sales drop by 400 MWh/day if temperature rises by
  1C
• monthly loss (31? 400 ? 18) 223.200 ?
• ave. 1969-1998 HDD(Frankfurt) 686.4
• in 18/30 years HDD(Frankfurt) lt 500
• Contract
• Tick size (400 ? 18) 7200 ?
• Strike 500 HDD
• Cap 100 HDD ? 720.000 ?
• premium 120.000 ?

15
Idealised example

• if winter is cold (HDD gt 500) ? no payment
• if winter moderately warm option in the money
• break even 483.3 HDD
• if winter is extremely warm cap limits payment

16
Market

• Hedger energy, agriculture, food and drink
  industry, building, tourism, ... ? management
  of exogenous risks
• Risk taker (re-)insurance companies,
  (investment) banks, energy suppliers, ... ?
  diversified portfolio, balance of risks

17
Market

• Market places
• Chicago Mercantile Exchange
• London International Financial Futures and
  Options Exchange (LIFFE)
• Eurex (Frankfurt) ? xelsius.com

18
Market

• CME expects that the products are not dealt by
  end customers but by risk traders ? secondary
  market

trader
(online-) broker stock exchange
(re-) insurance companies
(investment-) banks
19
Market

• price model
• Black/Scholes model, accepted for option prices,
  is not applicable
• no other widely accepted price model ? premiums
  not transparent, may vary by a factor 10!?
  possible hedgers are discouraged from entering
  the market
• possible widely accepted price model must
  reflect reality, otherwise market prices and
  economic cost of weather differ

20
Market

• market needs to be complete
• Any payoff vector ... may be realised.
• number of traded derivatives matches at least the
  number of uncertainties (meteorological
  parameter, time period, place of measurement,
  ...)

21
Use of meteorology

• Listed under problem fields!
• methods require an estimate on the variability of
  the weather variable
• generally taken from historic data
• pricing may depend on length of historic times
  series
• 30 years seem to be generally accepted
• station data from national weather services is
  strictly preferred

22
Use of meteorology

• Problems like heat islands or relocation of
  stations are known, data needs to be corrected
• stationarity of the stochastic of the weather
  variable not generally given
• higher confidence is given to more recent
  measurements

23
Use of meteorology

• Meteorological data needs to be of high quality,
  cheap and quickly/easily available

24
Use of meteorology

• Problem connection between DD value and business
  performance is often only weak
• profit dependent on economic factors
• external economic cycles, general
  social/economic changes, ...
• internal higher efficiency, new markets, ...

25
Use of meteorology

• HDD is a bad predictor for the predictand
  business performance
• use of additional meteorological information
  reduces the amount of unexplained variance

Unexplained variance 42
Unexplained variance 61
26
Use of meteorology

• Disadvantage of using additional meteorological
  data (e.g. model output, objective analysis)
  number of control variables increases ? number of
  different markets increases ? liquidity
  decreases
• Generally no interest in more complex
  meteorological data than station values.
• Clash of Cultures

27
Use of meteorology

• Specific market traders (e.g. re-insurance
  companies) may have special interest in more
  complex meteorological methods
• would reduce risk, increase profit (especially,
  if market prices are based on less appropriate
  methods)
• Methods include seasonal prediction and Monte
  Carlo modelling
• !!!TOP SECRET!!!
• Reduces possibility for a generally accepted
  pricing method

28
Summary and Outlook

• Weather derivatives Measured weather is traded
  like goods
• large market for business activities with a
  dependency on weather
• Most common HDD and CDD as integrals over period
  (month, season)
• trade market established in Chicago in 1997,
  difficult start in London, stagnation in Frankfurt

29
Summary and Outlook

• Success of trading weather derivatives relies on
  the simplicity of the products
• needed for liquidity of market and accepted
  pricing method
• simple statistical use of plain weather
  measurements hardly appropriate to reflect
  dependency on weather

30
Summary and Outlook

• Trade with weather derivatives in the USA
  connected with liberalisation of energy market
  and thus increased competition
• need to manage risk of energy suppliers/traders
• need to react to energy consumers needs
• Energy market in Germany is only partly
  liberalised, competition is low
• little need to compete for the consumers
• relatively large regions make it possible to
  manage risk within the enterprise