1. Efficient Short-Run Pricing A Spatial “Smart Ma

PreliminaryDraftASpatial\SmartMarketforElectricPowerandTransmissionbyJereyK.MacKie-MasonUniversityofMichigan,UniversityofCaliforniaEnergyInstituteandNBERJuly1994Currentversion:March15,1995Abstract.Creatingsuccessfulcompetitiveelectricpowermarketsrequirestransmissionpricingreform.Duetothephysicsofpowertransmissionnetworks,efficienttransmissionpricesmustincorporatetwotypesofloopflowexternalities,andmustbedeterminedjointlywithefficientpower(nodal)prices.Priorproposedmarketschemesallinvolveanapproximationtotheefficientprices.Iproposeareal-time“smartmarket”thatprovidesincentivesfortruthfulrevelationoftheinformationrequiredforfindingthecorrectefficientprices.Address.Prof.JeffreyK.MacKie-Mason,Dept.ofEconomics,UniversityofMichigan,AnnArbor,MI48109-1220.E-mail:[e-mail]“SmartMarket”forElectricPowerandTransmissionJeffreyK.MacKie-Mason1.EfficientShort-RunPricingTheentireinstitutionalandmarketstructureoftheelectricpowerindustryisinturmoilinmanycountries.TheU.K.hasderegulatedthepowermarket,andcreatedacentraltransmissionpool(NewberryandGreen(1995)).NewZealandhasbeenimplementingshort-runmarginalcostpricingforpower(Ring(1994)).RegulatorsatbothstateandfederallevelsintheU.S.havebeenstudyingradicaltransformations.Forexample,theCaliforniaPUCrecentlyproposedretailcompetitionamongpowerproviders(BlumsteinandBushnell(1994)).Transmissionpricingreformisessentialtothecreationofcompetitivepowermarkets.Apowersystemconsistsofsupply(generator)anddemand(load)nodes,connectedbyatransmissiongrid.Bydeliveringpoweroverthegrid,multiplesupplierscancompetetoserveagivendemand.However,buildingatransmissiongridinvolveshugefixedcostsrelativetovariableoperatingcosts,andthusfortheforseeablefuturecrucialcomponentsofagridwillbenaturalmonopolies.Theregulationandpricingoftransmissionaccesswilldeterminethesuccessofpowermarketcompetition.Transmissiongridsareextraordinarilycomplexengineeringsystems.Thedeliveryofpowerfromonenodetoanothermayinvolveflowsalongseveral“parallel”paths;theactualflowsaredeterminedbyphysics,noteconomicsorcontractualarrangements.Asaresult,theflowsinducedbyonepowertransactionmayaffectthecostsincurredbyothersonthegridwhoarenotparticipantsinthetransaction.Twotypesofexternaleffectsarecommon:flowsinducedbyatransactionmay(1)raisemarginaltransmissionlossesforothergridusers;and(2)raisemarginalgenerationcostsiftransmissionlineconstraintsbecomebinding.Powergridphysicsrequirethatefficientpricesforbothpowerandtransmissionbefoundjointlyasthesolutiontoanoptimizationproblem.TheextenttowhicharegulatoryreformobtainsIhavebenefittedfromconversationswithEdKahn,BartMcGuireandMatthewWhite.IundertooktheresearchforthispaperwhilevisitingtheCaliforniaEnergyInstitute,UniversityofCalifornia,Berkeley;IamgratefulforthefinancialsupportprovidedbyUCEI.1thebenefitsofanefficientpowermarketdependsontheclosenessoftheresultingtransmissionandpowerpricestotheefficientsolution.Numerousproposalsfordevelopingpowermarketsthatapproximatetheefficientsolutionhavebeenmade;twoofthebetterknownareSchweppe,Caramanis,Tabors,andBohn(1988)andHogan(1992).InthispaperIreviewthebasiceconomicsofefficientpricingforpowerandtransmission,withafocusonshort-termwholesaletransactions.Iemphasizethesteepinformationalrequirementsforefficientpricing,especiallyinthewholesalemarketsthatstandbetweenthemanynon-integrated“local”grids.Ithenproposeafeasiblemarketmechanismthattheoreticallyobtainsthefullyefficientsetofpowerandtransmissionprices.Themechanismisbasedonareal-time“smartmarket”auction.Theauctionisdesignedtoelicittruthfulrevelationoftheinformationnecessarytodeterminetheefficientsetofprices.Oneusefulroleforthesmartmarketmodelistoprovideameaningfulbenchmarkagainstwhichtocomparevariousapproximateapproachestoefficientpricing.Theapproachisusefulbecausethesmartmarketcanfeasiblybeimplemented:comparisonscouldbemadethroughsimulationstudies.Sincethesmartmarketfeasiblyachievesthefirst-bestoutcome,comparisonswouldindicatehowlargearetheefficiencygainspotentiallyforegonebyvariousapproximationmethods,andthusprovideguidanceonwhereitisworthwhiletryingtoimprovetheseapproximations.Forexample,thequalityofHogan(1992)’s“contractnetwork”approximationmethoddependsontheavailabilityofsubstantial(oftenprivate)informationoncostsandpurchaservaluationsofpower.Inhisdiscussionoftheavailabilityofthatinformation,Hoganassumesthatthemarketstructurewilloperateintheformofatightlyintegratedpool,withcontrolofthegridextendingtotheleveloftheindividualgenerator.Infact,inmanycasesmarketreformislikelytooccurfirstatthewholesalelevel,withtherelevanttransmissiongridinvolvingconnectionsbetweenvariousswitchyardsthatprovidetielinestonon-integratedutilitygrids.TheinformationrequirementsforHogan’sapproximatepricingmethodaremuchmoreseverewhennodesconnectnottospecificloadsorgenerators,buttoothergrids.ThesmartmarketmechanismIproposebelowissufficientlygeneraltoinducerevelationoftherequiredinformationwithoutmuchincrementaldifficulty.11Indeed,allornearlyalloftherequiredinformationisalreadyknowntotheparticipants,sorevelationdoesnotrequireadditionaleffortontheirpart.Theproblemist