Volumetric Hedging in Electricity ProcurementYumi OumDepartment of Industrial Engineeringand Operations Research,University of California,Berkeley, CA, 94720-1777Email: [email protected] Email: [email protected] Email: [email protected] Shmuel OrenDepartment of Industrial Engineering and Operations Research, University of California Berkeley, CA 94720-1777Shijie DengSchool of Industrial and System Engineering Georgia Tech Atlanta, GA, 30332-0205Abstract— Load serving entities (LSE) providing electricityservice at regulated prices in restructured electricity markets,face price and quantity risk. We address the hedging problemof such a risk averse LSE. Exploiting the correlation betweenconsumption quantities and spot prices, we developed an optimal,zero-cost hedging function characterized by payoff as functionof spot price. We then show how such a hedging strategy can beimplemented through a portfolio of call and put options.I. INTRODUCTIONThe introduction of competitive wholesale markets in theelectricity industry has put high price risk on market partici-pants, particularly on load serving entities (LSEs). The uniquenon-storable nature of electricity as a commodity eliminatesthe buffering effect associated with holding inventory, andmakes the possibility of sudden large price changes morelikely. Significant market risks that are faced by LSEs are notrelated to price alone. Volumetric risk (or quantity risk), causedby uncertainty in the electricity load is also an importantexposure for LSEs since they are obligated to serve the varyingdemand of their customers at fixed regulated prices. Electricityvolume directly affects the company’s net earnings and moreimportantly the spot price itself. Hence, hedging strategiesthat only concern price risks for a fixed amount of volumecannot fully hedge market risks faced by LSEs. The priceand volumetric risks are especially severe to LSEs becausesupply and demand conditions usually shift adversely togetheras demonstrated by the California electricity crisis in 2000 and2001, which led three large LSEs in California to bankruptcyor near bankruptcy.As a way of mitigating price risks in electricity markets,derivatives such as futures, forwards and options have beenused. An electricity forward contracts obligates a party to buyand the other party to sell a specified quantity on a given datein the future at a predetermined fixed price. At the maturitydate if the market price is higher than the contracted forwardprice, then the buyer would make a profit, conversely, if themarket price is lower than the forward price then the buyer willsuffer a loss. The profits and losses are paid when the deliveryis completed. Put or call options are also used for differenttypes of risk hedging: A call (put) option on electricity supplyobligates the seller to reimburse the buyer for spot prices above(below) the strike price. LSEs would also use call options toavoid the risk of higher prices while still being able to forgothe contract and enjoy the benefit of lower spot prices.While it is relatively simple to hedge price risks for aspecific quantity, such hedging becomes difficult when thedemand quantity is uncertain, i.e., volumetric risks are in-volved. When volumetric risks are involved a company shouldbe hedged against fluctuations in total cost, i.e., quantity timesprice but unfortunately, there are no simple market instrumentsthat would enable such hedging. Furthermore, the commonapproach of dealing with demand fluctuations for commoditiesby means of inventories is not possible in electricity marketswhere the underlying commodity is not storable.The non-storability of electricity combined with the steeplyrising supply function and long lead time for capacity ex-pansion results in strong positive correlation between demandand price. When demand is high, for instance due to a heatwave, the spot prices will be high as well and vice versa.For example, the correlation coefficient between hourly priceand load for two years from April 1998 in California1was0.539. [2] also calculated the correlation coefficients betweennormalized average weekday price and load for 13 markets:for example, 0.70 for Spain, 0.58 for Britain, and 0.53 forScandinavia. There are some markets where this price andload relationship is weak but in most markets load is the mostimportant factor affecting price of electricity.The correlation between load and price amplifies the ex-posure of an LSE having to serve the varying demand atfixed regulated prices and accentuates the need for volumetricrisk hedging. An LSE purchasing a forward contract for afixed quantity at a fixed price based on the forecasted demandquantity will find that when demand exceeds its forecast and itis underhedged the spot price will be high and most likely willexceed its regulated sale price, resulting in losses. Likewise,when demand is low below its forecast, the spot price at whichthe LSE will have to settle its surplus will be low and mostlikely below its purchase price, again resulting in losses.Because of the strong causal relationship between electricityconsumption and temperature, weather derivatives have beenconsidered to be an effective means of hedging volumetricrisks in the electricity market. Weather derivatives, whose1During this period, all the regulated utilities in the California marketprocured electricity from the spot market at the Power Exchange (PX). Theywere deterred from entering into long term contracts through direct limitationson contract prices and disincentives due to ex post prudence requirements.Authorized licensed use limited to: Univ of Calif Berkeley. Downloaded on March 19,2010 at 18:59:34 EDT from IEEE Xplore. Restrictions apply.payoff is triggered by weather conditions, exploit the correla-tion between electricity demand and weather condition. Forexample, if the upcoming winter is milder than usual, theelectricity demand would be low leaving an LSE with lowrevenue. The LSE can protect against such situation by buyinga Heating Degree Days (HDD) put option, which gives apositive payoff if the winter was realized milder than the HDDstrike value denotes and zero payoff otherwise. However, thespeculative image of such instruments makes them undesirablefor a regulated utility having to justify its risk managementpractices and the cost associated with such practices to aregulator.In this