new uses for an old tool - final...new uses for an old tool 1 about eq research eq research, llc...

NewUsesforanOldTool:UsingCostofServiceStudiestoDesignRatesinToday’sElectricUtilityServiceWorld

PamelaMorgan&KellyCrandallEQResearch,LLC

April2017

NewUsesforanOldTool 1

AboutEQResearch

EQResearch,LLCprovidescleanenergy intelligenceforclients inthenonprofit,business,andgovernment sectors. Our analysts monitor energy policy and regulations in all fifty states,stayingabreastof trends indistributed solarprogramsandpolicies,utility rate cases, energystorage,gridmodernization,andotherareas.Weofferbothmonthlysubscriptionservicesandspecialized consulting, including policy analysis and expert testimony. EQ Research’s staff islocatednationwide,withaprimaryofficeinCary,NorthCarolina.

AboutPamelaMorgan

PamelaMorgan is anEnergyPolicyExpert forEQResearch. Sheoffers consulting, facilitationand coaching to groups and individuals seeking strategic and creative thinking and acting inrelationtothechallengesoftoday’senergyindustry.Ms.Morganworkedintheelectricutilityindustry forover20years,primarilyatPortlandGeneralElectric,withresponsibilities rangingfrom all aspects of state and federal economic regulation to resource planning, governmentaffairsanddistributionoperations.Herprojectshaveincludeddesigninganddeliveringsystemsthinkingworkshops for state regulators; researchinganddraftingamajorupdateof resourceplanningandprocurementrulesfortheMontanaPublicServiceCommission;preparingamajorstudyofdecouplinganddecouplingadjustmentsand testifyingondecoupling for theNaturalResources Defense Council (NRDC); developing a comprehensive set of national energyefficiency policies for NRDC to advocate in Chile; and advising the national electric utility inTanzaniaonconstructiveregulatoryrelations.Shespecializesinhelpingindividualsanddiversegroupsofpeopleseeclearlythecontextwithinwhichthestrategicchallengesconfrontingthemarise,andcreativelydevelop iterativeactionsthatmovetowardan idealstate inwhichthosechallengeshavedisappeared.SheholdsaB.S.fromWashingtonStateUniversityandaJ.D.fromTheUniversityofWashingtonSchoolofLaw.

AboutKellyCrandall

KellyCrandallco-authoredthispaperwhileservingasaSeniorRates&ResearchAnalystatEQResearch.PriortojoiningEQResearchin2015,shegainedsixyearsofpublicsectorexperienceat theCityofBoulder, theColoradoPublicUtilitiesCommission,and theNationalRenewableEnergyLaboratory.WhileattheCityofBoulder,sheundertooka“bootcamp”onelectricutilityoperations, focusing on financial and resource planning issues for startup entities. As a non-practicing attorney with a deep quantitative background, she brings an interdisciplinaryperspectivetoaddresscomplexenergyproblems.

AcknowledgmentsWewishtothankKevinFoxandJasonKeyesofKeyes&FoxLLPforsupportingthisproject,aswellasJoeWiedman.ThanksalsotoChelseaBarnesforhercarefulediting,andBlakeElderforhissupportonlayoutandgraphicdesign.Anyerrorsthatremainareourown.


TABLEOFCONTENTSAcknowledgments..................................................................................................................1

TABLEOFCONTENTS..............................................................................................................2

ANOTHERPAPERONUTILITYRATES?......................................................................................3RateDesignTerminology.................................................................................................................6

RATEDESIGN:PRESENTANDPAST..........................................................................................7LegalBasisoftheUtilityCostStudy.................................................................................................9ChangingTrendsonCostandCompetition.....................................................................................10MoreComplexModels,Data,andRates........................................................................................12IncreasingNumbersandComplexityofProceedings......................................................................13WhereWeAreNow.......................................................................................................................15

FLAWSINTHECCOSS:THETHEORY.......................................................................................16WhileBetterThan“ThePublicInterest,”CostRemainsanInexactStandard..................................17UtilityCostsSendPoorPriceSignals..............................................................................................17TheMoreGranulartheRate,theMoreImpracticala“Cost-Based”Rate........................................19AGeneralRateCaseisNottheRealWorld.....................................................................................19TheTheoryInformstheTool..........................................................................................................21

FLAWSINTHECCOSS:THETOOL...........................................................................................22TheCCOSSisaBlackBox................................................................................................................22CCOSSMethodologyNeedsRefreshing..........................................................................................24TheCCOSSDoesNotMeasure“Success”........................................................................................26AnInexactToolProducesInexactResults.......................................................................................27

LOOKINGTOTHEFUTURE:REPAIRORREPLACE?..................................................................28QuestionsRegulatorsCanAsktoMakeRateStructureProposalsUseful....................................... 30Conclusion.....................................................................................................................................31APPENDIX1:StepsintheCostofServiceStudy..............................................................................32APPENDIX2:PrinciplesofPublicUtilityRates................................................................................34APPENDIX3:TheThreeClassificationMethodologies....................................................................35

ENDNOTES...........................................................................................................................36


ANOTHERPAPERONUTILITYRATES?They’reback.No,notghosts—ratecases:specifically,electricutilityratecases,andthecostofservicestudiesthatcomewiththem.Afteralullinthe1990sandearly2000s,thenumberofgeneralratecasesbegantoescalateandcontinuestosurgetoday.Aginginfrastructureandthedigital economy have put pressure on investor-owned utilities (“IOUs”) and public powerutilities to increase investments, particularly in their distribution systems and software.Concurrently,thegrowthindemandforelectricity—formanyutilities—hasstagnated.Increased investmentand lowor flat loadgrowth isacommonrecipe for ratecases.What isnew in this decade’s rate cases is a rising interest in changing how residential and smallcommercialelectricityuserspayforutilityservices.Utilitiesareproposing,andregulatorsaregrapplingwith,newratedesigns,sometimesforthefirsttimeindecades.Frequently,utilitiessupport theirnewpricingproposalsby referencinga textwrittennearly60yearsago: JamesBonbright’s1961PrinciplesofPublicUtilityRates.1In1961,manyquestionsofregulatoryeconomicsremainedunsettledacrosstheUnitedStates,rangingfromthepropertreatmentofacquisitionadjustmentstomethodsofcalculatingarateofreturn.Whatisnowalmostuniversallycalled“revenuerequirement”wascalledthe“costofservice,” and utility cost studies were high-level, if they were conducted at all. Most ofBonbright’sworkfocusedoncomplexrevenuerequirementandrateofreturnquestions,intheprocesssummarizinga robustconversationbetweeneconomists thathadspannedalmost50years. Bonbright is best known, however, for developing a series of high-level criteria forregulators to use in determining if utility rate proposalsmet the abstract legal standards ofbeing“justandreasonable”andpreventing“unduediscrimination.”Whileacknowledgingthatratedesignisanart,hearguedthatitcouldbeanartthatwasatleastpracticedconsistently,withaclearlyarticulatedeconomicfoundation.In the three decades that followed Bonbright’s book, the concept of the monopoly electricutilityproviding serviceatdeclining costswas flippedon itshead,publicutilities commission(“commission”) proceedings expanded in scope and number of participants, and personalcomputerscameintowideruse.Thedesignation“costofservice”cametobeusedtorefertotheprocessofallocatingcoststocustomerclasses,whereasthefundstheutilitymustcollectare now called the “revenue requirement.” In 1992, the National Association of RegulatoryUtility Commissioners (NARUC) issued the Electric Utility Cost Allocation Manual (NARUCManual),whichcataloguedaseriesof“generallyaccepted”methodologiesregulatorsandtheirstaffsshouldconsiderwhendecidinghowtoallocateanapprovedrevenuerequirementamongbroad classes of utility ratepayers.2The NARUC Manual catalogues the evolution of thoseaspects of utility cost studies that became at least somewhat settled in the years sinceBonbrightarguedthattheirmeritswere“sodubiousthattheyfullyjustifytheskepticismwithwhich utility cost analysis has been received by public utility companies and public servicecommissions.”3 These aspects were the methods for functionalizing, classifying, and then


allocating costs among the various rate classes4or—if a rate class had more than one rateschedule—thevariousrateschedules.Wearenowatanotherstageintheevolutionofutilitycostanalysis.Utilitiesandregulatorsaregrapplingwithcomplexquestionsaboutthefutureoftheenergyutility industry,arisingfromforces of technology, economy, demographics, and policy changes. With better and moredetaileddatanowavailable,manyregulatoryparticipantshavebeguntoadvocateusingresultsoftheclasscostofservicestudy(“CCOSS”)directlytosetthelevelofvariousrateparts,suchasfixedcustomerchargesanddemandcharges.Thisisabreakfromtradition:inthepast,CCOSSswere informational, but not dispositive -- useful as a framework for discussing class revenuerequirementsandraterelationships,butnotfordeterminingthedirectpricesthatcustomerspayforselectedpartsofelectricityservice.TheCCOSSisamathematicalmodelpackedwithassumptionsabouthowtoallocateautility’soverall base revenue requirement among the various services it provides and the types ofaccountsithasforeachofthoseservices.Appendix1outlinesthekeystepsintheCCOSS.Themodelgenerallyfocusesoncustomerclasses—thetypesofaccountswhoseusageprofilesareassumedtobesimilarenoughthatthecharacteristicsoftheirelectricityusecanberepresentedasagroup—suchasresidential,commercial,andindustrial.TheCCOSSseparatestherevenuerequirementintocoststhatcorrespondwiththesegroupcharacteristics.Yourarely, ifever,seeautilityratecasewithoutaCCOSSanymore.Frequently, theCCOSS isthe basis for proposals to increase the fixed monthly charge for residential and smallcommercial customer accounts, or to add a demand charge to those rate schedules that fordecades have not had one. A common rationale for using CCOSSs in rate cases is that suchstudiesquantitativelyanswerthreeinterrelatedratestructurequestions:

• Whichutilitycustomershavecaused,andarecausing,whichutilitycosts?• Whichutilitycustomersshouldpaythosecosts?• Whatchargeswillensurethatthe“right”customeraccountspaycoststhey“cause?”

In so believing, stakeholders often take the assumptions underlying a CCOSS for granted,arguing about inputs rather than purposes, and expertise rather than outcomes. But thequestions include a leap of logic that the methodologies of CCOSS do not support. In thispaper, we address why the CCOSS is ill-suited to the new demands being made of it --specifically,whyaCCOSSdoesn’t“answer”questionsrelatedtowhocauseswhichcosts,whoshouldpay,andhowtheyshouldpay.First,weexplaintheprominenceofratedesignquestionsandCCOSSstoday,andilluminatethehistory that led to that prominence. Evolving economics, technology, demographics, andpoliciesthroughoututilityserviceterritoriesarepressuringoldratestructuredecisionsthat,insomecases,datebackdecades.Understanding thehistoryand thecontextof costof serviceandtheCCOSStooliscriticaltoreflectingonitsuseandusefulnesstoday.


Second,wedescribehowthemodernCCOSSispoorlysuitedforansweringthechallengingratedesignquestions it isbeinggiven. Indoingso,we lookatbothregulatoryeconomic theory—withafocusonBonbright’stexts—andthetoolitself.Wehighlightthemanyframing,data,andmethodology choices and challenges facing those preparing and questioning the tool and itsresults.Inshort,theregulatoryarenaisdueforaconversationabouthowandwhycircumstancesarechanging, and what actions will create more desirable outcomes. Our suspicion is that theCCOSS as currently practiced will not enlighten that conversation. Rather, fueled by anabundanceofdata that is stillnotnecessarily completeoruseful,CCOSSsmaybeanelectricindustry version of the recently coined term “weapon of math destruction:” a tool that isopaque,scalable,andharmful.5But if the CCOSS is not the focus of discussion about rate design, and if it cannot serve asjustification for decisions aboutwhich households and small businesses should paywhat forelectricityservice,whatshouldstakeholdersdo?Toaddressthis,weofferrecommendationsonhowtothinkaboutratestructureandtheroleofCCOSSsinthefuture.Oldquestionsarenewagain:Whichratestructuresarefair?Whichservicesdoesandshouldapublicutilityprovide?Whichservicesshouldhaveapriceandwhoshouldpay?Ifyouronlytool isahammer,everyproblemlookslikeanail.NoteveryquestioncanbeansweredwithaCCOSS,soregulatorsandstakeholders needmore than just a hammer in their toolbox.With an understanding of thehistoryofcostofserviceanditsexpressioninaCCOSS,weoutlinewaysthatregulatorscanputtheCCOSStooltogooduseinordertoreachbetteroutcomesovertime.


RATEDESIGN:PRESENTANDPASTRatedesignisakeyissueforelectricutilitiesacrosstheU.S.,andthreebigquestionsdominatetoday’sdebates:

• Should a given utility change its residential or small commercial rate classsegmentation?Thismightappearasaquestionaboutwhethertheutilityshouldcreateanewclassforaccountswithrooftopsolarorforlow-incomecustomers.Ithasemergedout of an increasing knowledge about the characteristics and usage patterns withinthese mass-market classes. Regulators and practitioners, however, have rarely setparametersaroundwhatdegreeofdifferencejustifiesestablishinganewrateclass.

• Should a given utility increase the customer charge applied to accounts on its

residentialandsmallcommercialrateschedules?Outof126generalratecasesfiledbyinvestor-ownedelectricutilitiesthatEQResearchhastrackedsince late2014,88havesoughtincreasesinthemonthlyresidentialcustomerchargeof25%ormore.6Generally,utilitiesrequesttoincreasecustomerchargesforresidentialandcommercialcustomerswiththeargumentthatahighlevelofcostsforservingsuchaccountsareembeddedor“fixed.” For example,Wisconsin Public Service Corporation recently claimed that thefixed, embedded cost to serve residential customers was $68.81 per month andproposed to increase theirmonthly charges from$19 to $25.7However, commissionshave often rejected large fixed-charge increases, citing negative impacts on customerfinancialincentivestoinvestinenergyefficiency.8

• Should some or all of the accounts in the residential and small commercial classes

have demand meters and pay a demand charge? Around the middle of the 20thcentury,meters andpricinghad generally settled into thepattern that prevails todayexcept where utilities have invested in advanced metering infrastructure (AMI):residential and small commercial accounts have energy meters, and other accounts9have meters that are capable of measuring peak demand during the billing period,whichisthebasisofademandchargeforthoseaccounts.ForutilitieswithAMI,theonlyhurdlethatremainsbeforeexpandingtheuseofdemandchargesispoliticalfeasibility.While severalutilitieshaveproposeddemandcharges specifically for solar customers,Oklahoma Gas & Electric and Arizona Public Service Company proposed universaldemand charges for residential customers.10Their stated justification is that a three-partrate“providesamoreaccuratepricesignaltocustomers,whichhelpstopromoteefficientuseofenergy.”11

These questions have emerged because four important stakeholders in the utility regulatoryarena—utilities,newmarketentrants, regulators,andmass-market customers—have reasonstobedissatisfiedwithcurrentratestructuresandratedesignconversations.


• Utilities are concerned about revenue erosion and lost investment opportunities.Customers have new opportunities to meet at least some of their electricity needswithout their electric utility, through energy efficiency, conservation, customer-sitedgeneratingresources,batterystorage,energymanagementsoftware,smartappliances,or a combination thereof. This threatensutilities’ ability to collectenough revenue tocover increasingcosts,andmaycurtail future investmentopportunities consideredbyinvestorsinvaluingthecompany.

• New entrants want opportunities. Rate structure plays a significant role in the

economic case that new entrants make to potential customers when they offersubstitutesforutilityservices.Economicsisnot,ofcourse,theonlyreasonpeoplemaydecidetoinvestinthesesubstitutes,butitisasignificantone.

• Regulators are under pressure. The growing number of rate case intervenersmeans

growing demands on regulators, including demands to lessen environmental impacts,reduce pressure on low-income ratepayers, or increase economic competitiveness.Thesedemandstypicallyfocusandcollideonratestructurechanges.

• Mass-market ratepayersmay notwant new rate structures. Some customerswould

likedifferentserviceattributes,suchasmorerenewableelectricityorincentivestouseenergy off-peak, and everyone wants more affordable bills, but few customersaffirmativelywantdifferentratestructures.

As commissions grapple with these complex issues and mediate between these differentstakeholders,CCOSSshavegainedunprecedentedimportance.Whilecommissionsstilltendtoacknowledgethatratemakingisanart,manystakeholdersarguethatratesshouldcomeclosertothe“ideal”asquantifiedintheCCOSS,andthatviewisgainingtraction.RecentyearshaveseenratecasetestimonysuggestingthattheCCOSSshouldbethefoundationforratedesignbecauseofthreeinterrelated,foundationalassumptions:

• RatesdesignedusingCCOSSallocationsarecost-based;• Cost-basedratesdonotundulydiscriminate;and• Cost-basedrateswillsendthebestpricesignals.12

While these three high-level assumptions are not wholly wrong, they are also not preciselycorrect. To explain why, we first explore the confluence of electric industry trends that arecreatingsuchaheavyrelianceontheCCOSStoolfordecidingbothratespreadandratedesignquestions.


LegalBasisoftheUtilityCostStudy

Statesformedcommissionstosetpricesorratesforcriticalservices(includingelectricity)whentheirmonopolynaturebecameclearandthemonopolypropensityforpricemanipulation,suchas that of Standard Oil, became known.13Early legal cases regarding economic regulationestablishedastructurethatbalancedthetensionbetweenthestate’spolicepowerstoensuresafety and public welfare under the Tenth Amendment with prohibitions on governmenttakings of private property under the Fifth Amendment.14In the absence of sophisticatedaccountingsoftware,earlycommissiondecisionsandcourtreviewsfocusedonestablishingtheoverallcostoftheutility’sserviceandinstitutedambiguouslegalstandardsthatpricesbe“justandreasonable,”not“undulydiscriminatory,”and“inthepublicinterest.”Identifyingaratestructurethatmetthoserequirementswasnotatrivialtask.Underconditions“whenacompany’searningspowerissohighthatanynumberofavarietyoftariffscouldbemadetoyielda fair return,”15commissionssoughtsomerationale toselectamongstpossibleratestructures,leadingtotheCCOSS.Priortothe1980s,thesestudieswerelimitedbythelackofelectronicdataandcomputerprocessingcapability.Commissionsdidnotblindlyadheretotheirresults,andcourtsrepeatedlyacknowledgedcommissionauthoritytoengageinthe“art”ofutilityregulation.16Moreover,somestatesexplicitlyrequiredtheircommissionstoconsiderobjectivesofratestructureotherthanreflectingthecostofservice.

In recent years, however, the temptation to rely on a CCOSS has strengthened. If a utility’srevenuerequirementshouldbe“cost-based,”itseemsonlylogicalthat“cost”shouldguiderate

“Q:Ifa...cost-of-servicestudyistoassistwithratemakingandtheadequacyofrates,shouldresultsfromthestudybeusedtohelpguideratedesign?”

“A:Yes.Thecost-of-servicestudywillrevealtherateclasses’revenuerequirementsandtheunitcostsforuseinthedesignofeachoftherates.Byadheringascloseasfeasibletothesecosts,subsidiescanbeminimizedandefficientpricesignalswillbesent.”

FloridaPublic Service Commission,DocketNo. 160186-El,Direct TestimonyofMichaelT.O’Sheasy,witnessforGulfPowerCompany,October12,2016,atpp.5-6.

“Bywhatbasicstandards...shallregulationpassjudgmentonasystemofelectric-utilityrates which allows liberal discounts for incremental blocks of energy; or which levieshigher charges, per kilowatt-hour, on residential consumers than on industrialconsumers; or which concedes lower rates for off-peak consumption than forconsumptionatpeaktimesorseasons?”

BonbrightontheCommissioner’sConundrum,1961,atp.288.


spread and rate design as well. The last several decades have seen changing cost andcompetition trends, increasing granularity of data and rates, and greater complexity of ratecaseproceedings.Thenext sections summarize these threekeyhistorical trends,whichhavecombinedtoelevatetheCCOSStothekeytoolofmodernelectricratedesign.

ChangingTrendsonCostandCompetition

The design of rates—not just their overall level—becomes a key issue every time a newchallenge to the electric industry’s competitiveness emerges. This is because competitioncomes at the individual customer level, whether that customer is the energy manager of amanufacturing facilityor thebillpayer ina small townhouse.Electricbillsare thebasisuponwhicheachcustomerwilldecidewhether tochangesomeaspectof thewaytheymeet theirelectricityneeds.Onlyrarelydoesthisdecisioninvolveacompletealternativetoelectricutilityservice (i.e., going off the grid). Because of this, it is the variable or per-unit cost of utilityelectricity that becomes the point of comparison and choice. The specter of competition forindividual customer energy choices has risen about every twenty years over the last severaldecades,withfocalpointsinthe1970s,1990s,and2010s.The 1970s challenged natural monopoly assumptions. In the aftermath of World War II,electricutilitiesbyandlargedemonstratedthecharacteristicsofanaturalmonopoly:decliningper-unit costs. Demand soared, with annual growth rates for electricity consumption in thedouble digits.17The costs per unit of power delivered decreased as technology (particularlypower generation) improved and costs were spread over more customers and moreconsumption.Utilities freely encouraged consumption through all-electric homedisplays andpromotionalrates.18The1970s,however,turnedthenaturalmonopolyassumptiononitshead.New generation, including nuclear plants, became expensive. The Energy Crisis dramaticallyincreasedthecostoftheoilstillinwidespreaduseforgeneratingelectricity.Inflationonlaborandcommoditiesandhighinterestratescausedratestorise,oftenbytriple-digitpercentages.As higher rates depressed the amount of electricity demandedby spurring conservation and(sometimes) energy efficiency investments, the rate increases came faster.19Marginal costsexceeded the average costs onwhich rateswerebased, leading to revenue erosion.Utilitiesfiledfrequentratecasesandsoughtfuelandpurchasedpoweradjustmentridersthatallowedthem to pass on some or all of their fuel risk to consumers. Hard-hit industrial consumersprotestedvehementlytorisingrates,andsomeofthefirstratestructurefightswereoverratespread:wastheindustrialclasssubsidizingtheresidentialclassorviceversa?Regulators responded, often under extreme public outcry and utility financial pressure.20Disallowednuclear andother new generating plant expenditures drove a fewutilities to thebrinkofbankruptcy;interimrateincreasessavedothers.WiththepassageofthePublicUtilitiesRegulatoryPoliciesActof1978andtheNationalEnergyConservationPolicyActof1978,someutilities began testing time-of-use rates, including for mass-market customers; othersguaranteedqualifyingfacilitieshighpricesforsupplyingpowertoutilities,openingthesupplyofelectricitycommoditytooutsidersforthefirsttimesincetheearly20thcentury.


Almostallregulators,occasionallyunderstatemandate,beganrequiringthatutilitiesengageinproviding basic energy efficiency programs. Several commissions opened proceedings toexploremodifyingratedesignstoencourageconservationandpeakdemandreduction.Thoughdramatic rate design changes rarely ensued, this period saw the first inverted rate designs,underwhichgreateruseinagivenbillingperiodtriggeredhigherratesandhigherbills.Intheprocess of these responses, commissions reviewed utility cost studies, but rarely consideredthemasconclusiveinmakingratestructure—eitherallocationordesign—decisions.

The movement toward retail competition in the 1990s brought fears of significant loaderosion. The 1980s saw competition become a worldwide movement, with airlines,telecommunications, and thennatural gasutilities all experiencing thepartial or total lossofmonopolypower.Bythe1990s,severalstateswereexploringretailelectriccompetition,underthe expectation that customers would experience both cost savings and innovation. Theprospect of retail electric competition, however, raised fears amongmany stakeholders thatnew, unregulated market entrants would cherry-pick the choicest customers—particularlylarge,industrialcustomerswithhighloadfactors(i.e.,fairlyconsistentelectricitydemand).Following in the footsteps of natural gas utilities, most electric utilities during this decaderevisedtheirratestructuressothatthelargercommercialandindustrialaccounts,whichweremost attractive to competitors, paid rates much more aligned with their individual usagepatterns. The energy component of electricity rates, billed per kilowatt-hour, increasinglyincluded riders thatpassed throughchanges in variable costs, like fuel andpurchasedpowercost fluctuations.Commissions institutedratefreezestoensurethatthechoiceofalternativeelectricityprovidersby somecustomersdidnot raiseutilityprices forallof those remaining.Economicdevelopmentratesre-appearedasafavoredtooltokeeplargeloads intheutility’sserviceterritoryandon itssystem,preserving largecustomers’contributiontofixedcostsforthebenefitofallratepayersandoftenpreservinglocaljobs.21Onlymass-marketcustomersdidnot receive a major rate-structure makeover. As computing power and the amount ofcomputerizeddataincreased,theCCOSSbecamemoregranular,butoftentookabackseattomorepressingpoliticalissues.The 2010s herald increasing capital investments but stagnant load. Because equipmentinstalledintheearlytomiddledecadesofthe1900sisnowreachingitsendoflifeandtheneedto ensure reliability is rising in an increasingly digital economy, utilities are steadily boostingtheir investments in thedistribution system, including through roll-outsofmultimillion-dollar

“Practical considerations also militate against making cost of service the exclusivecriterion inratesetting.Virtuallyeverycourtconsideringthematterhasrejectedoutofhandarulethatwouldreduceratemakingtoanexerciseincostaccounting...”

TheFloridaSupremeCourt in InternationalChemicalCorp.v.Mayo,336So.2d548,1976,atpp.551-552.


AMIandsoftwareupgrades.22Whilegenerationandtransmissionwilllikelyremainthelargestcategory of future capital investment, distribution investment is anticipated to risesignificantly.23 Inflation also remains a factor for most utility non-capital costs.24However,loads—and,thus,revenues—arenotrisingasquicklyasinflationor,forsomeutilities,areflator falling.25 Energy efficiency, particularly that gained through building codes and federalequipmentstandards,continuestodepressloads,alongwithchangesindemographicsandtheeconomy. From a high of 9.8% annual growth from 1949 to 1959, according to the EnergyInformation Administration, national average annual load growth has been 0.5% from 2000through2015.26Theprospectof long-term loadstagnation is causinga fewcommissions to renew interest in1980sconceptsofperformance-basedratemakingandnewalternativeutilitybusinessmodels,but the overriding issue for the last few years has been the treatment of distributed energyresources (“DERs”), including solar energy and battery storage. As was the case with thecompetition from conservation in the 1970s and retail electric providers in the 1990s, thehigher the rates for using electricity, the more attractive alternatives like DERs are tocustomers.Whatisnewistheavailabilityandattractivenessofalternativestomass-marketcustomers.Atahigh level, some believe the existing utility regulatory model is flexible enough to promoteincreasedcustomerchoiceandrenewableenergy;othersbelievethemodelwillrequirealesseror greater amount of adaptation. These opposing beliefs confront each other in complexgeneral rate cases or revenue-neutral rate design proceedings that are ill suited to resolvethem. Instead, experts put forward complex and competing CCOSS models as the basis forensuringthatratesdirectedtowardthesenewmarketentrantsare“fair”andservethepublicinterest.

MoreComplexModels,Data,andRates

Complicating the context for rate design issues now is the several-decade regulatory trendtowardcomplexmodelsfedbyincreasinglylargeamountsofdata.Almostallutilitydatanowexists in electronic form and new technologies, including AMI, dramatically increase theamountofdataavailable.Whilethissmartmeteringtechnologyisfarfromnew,theAmericanRecovery and Reinvestment Act (“ARRA”) kicked offwaves of investment. Computing powerhas increasedto the levelwhere thousandsof simulationscanberun inminutes rather thandays.Thegrowingimportanceofcomputermodelsassignificantifnotdeterminativeevidenceforregulatoryeconomicdecisionsledtoanewterm:“regulationbysimulation.”27The CCOSS has participated fully in this trend. The calculation-intense nature of utility costanalysesprovedchallenging inpre-computeryears,and fullCCOSSswereuncommoneven inthe1960s.28Thatchangedinthe1990s:withtheincreasingpowerofdesktopcomputers,least-costplanning(nowcalledintegratedresourceplanning)andothermodelsandmethodologies,includingtheCCOSS,begantoadddetailandscopetoutilityregulation.Business intelligencesoftwarenowpromises improvements to operations throughbetter tracking ofwork orders,


accounting,anddistributionsystemreliability—allrelyingonmoregranulardatathatcouldbedrawnon to populate the CCOSS.Utilities are still struggling to understand the full range ofopportunitiesthispresents.Another source of increasing complexity for rate design issues is the nature of electric ratesthemselves. Inpastdecades, thebillwas the resultofapplyingbillingdeterminant rates toaparticular customer account. The rateswere set in general rate cases, so-called because theratecasecomprehensivelydealtwithautility’scostofservice.This isnolongerthecase.Thebillscustomerspaygenerallyincludenotjusta“base”rate,butalsoanynumberofadjustmentsthatriseandfallonseparateschedules.Thefuelandpurchasedpowerautomaticadjustmentclauses approved in the 1970s were just the beginning. In recent years, commissions haveapproved pass-through adjustments for costs including generating plant retrofits forenvironmental compliance, vegetation management, smart grid investments, demand-sidemanagementcosts,andratecaseexpenses.29Generally,automaticadjustmentsaffectonlythevariable part of the rate, particularly formass-market customers. Thus, the “price signal’” acustomer’sratesandbillssendcannolongerbecomprehensivelyidentifiedintheratedesigncontextofageneralratecase.

IncreasingNumbersandComplexityofProceedings

Ontopofallthis,thesheernumberofelectricutilityregulatoryproceedingshasskyrocketed.Sofar,the2010shavebeenthedecadeofthegeneralratecase.Theaveragenumberofratecasesfiledinthe2010sexceedstheaveragenumberofthosefiledinpriordecadesforwhichwehaveconcretedata—outofabout190investor-ownedutilitiesinthecountry,afourtharefiling rate caseseach year (seeFigure 1).30Data fromEdisonElectric Institute (“EEI”) andEQResearchshowanaverageof38ratecasesfiledperyearsincefrom1990-2016.In2016,moreratecaseswerefiledthananyotheryearonrecord.Generalratecasesarenolongertheonlysourceofchangestoratestructureorindividualrates,however.Inadditiontotheadjustmentclauses mentioned above, many states are addressing rate structure questions outside ratecases,inproceedingsabout“revenueneutral”costallocationandratedesign.Thefrequencywithwhichratecasesandotherproceedingsoccurtaxstakeholders’resources.Whileutilitiescanpassthroughthereasonablecostsoftheirlegalfilingstotheircustomers—oftennowexceeding$1million31—otherpartiesbecomeresource-constrained.Inaddition,anunintended consequence of the overall paucity of general rate cases in the 1990s and early2000s was the attrition of industry experts, an issue that continues to affect regulatoryproceedingstoday.32


Figure1:Numberofratecasesfiledbyinvestor-ownedutilitiesbyyear33Meanwhile,evermorestakeholdersareappearingintheincreasingnumberofproceedings.Inthefirstfewdecadesofthe1900s,ratecaseparticipantsmighthaveincludedautility’slargestcustomers or the cities with which it had franchises, and those parties were primarilyconcernedwith cost and reliability. In the1970s, consumers’ counsels emerged to representthe interests of residential and small business ratepayers in the face of steadily increasingcosts.34Moreover, environmental nonprofits like the Environmental Defense Fund began toengage on issues such as the role of rate design in energy efficiency and conservation. Theinflux of retail competition and the emerging secondarymarket of energy services providersmeanthatmorepeoplewanttobeatthetable,particularlyinhigh-stakesproceedingslikeratecases. By the early 2000s and continuing today, a rate case can have twenty or moreinterveners, representing a range of interests including small customers, multistatecorporations,manufacturers,energyefficiencynonprofits, localgovernments,anddevelopersofproductscompetingforcustomers’attention.35A settlement among the parties is a frequent result of numerous rate-related proceedings,many stakeholderswith constrained resources, and a lack of experts to hire for help on thetechnicalissuesthatoftendominate.Indeed,settlementshaveresolvedandendedasignificantportionofratecasesinrecentyears.Aconsequenceofthisisthatratestructurepolicymaybelostinblack-boxsettlementsthatresolvecompetinginterests.Totheextentthatcommissions

35

40

57

44

27

40

33

2016 15

1114

2320

2824

34

4846

42

66

55

5053

51

58

48

70


relyontheirownpriordecisionsandthedecisionsmadebycommissionsinotherstates,black-box settlements considerably narrow the tools available to regulatory participants in otherjurisdictions or policymakers to help decide tough issues of economic regulation.Many suchsettlements also leave big parts of the decades-old rate structure untouched—such as themass-market residential and small commercial rates—and consequently, there may be littlerecentguidanceforthedecisionscommissionsaremakingtoday.

WhereWeAreNow

Pastisprologue,andwherewearenowcomesfromwhereelectricutilityregulationhasbeen.Theincreasingdiversityoftypesofrates,andtheincreasingnumbersofintervenersengagedinthelitigatedprocess(eachwithdifferentvaluesandpriorities)furtherencouragescommissionsto prioritize a tool that provides quantitative support for engaging in the art of ratemaking.LongstandingtheoreticalandpracticalflawstousingtheCCOSStodesignrates,however,makethis quantitative support dubious. Whether a CCOSS supports decisions that a given ratestructure is just and reasonable, in the public interest, and does not unduly discriminate isquestionable.


FLAWSINTHECCOSS:THETHEORYThe CCOSS grew out of a history of robust economic debate on the nature of regulatedmonopolies and how to set prices for their services. Economists like Kahn, Pigou, andmoremodernthinkerslikeBorensteinexploredregulationandcompetitionissuesindepth.ButwhenutilityratecasestakeholdersarguethataCCOSSdefinesthecoststhatratesmustrecover,theymostcommonly invokeJamesBonbright.Bonbright isbestknownforthe“criteriaofasoundratestructure”thatheputforwardasasummaryofadviceforregulatorsandtheregulatedastheycraftedratesoutofrevenuerequirements(wecompareBonbright’s1961and1988ratecriteriainAppendix2).Yet,Bonbrightwasalsorenownedbecausehisworksummarizedthen-currentthinkingaboutcontroversialissuesthathavesincebecomemoresettled—rangingfromthe calculation of a return on equity to the calculation of rate base using original cost orreplacementvalue.Themajority of electric rate cases filed in recent years either explicitly invoke Bonbright orreferenceoneormoreofhiscriteria.ReferencestoBonbrighttypicallystart—andstop—withreferencetoratestructure,andassert thatrelyingonhiscriteriawillprovidecustomerswith“better”price signals.Yeteven lookingonlyat the residential rateclass, thediversityof rateproposals that utilities claim meet Bonbright’s guidelines is staggering. They include highermonthly customer charges,36tiered customer charges,37tiered “grid use charges” meant torecoverdistributiondemandcosts,38residentialnon-coincidentdemandcharges,39andtime-of-usecharges.40Bonbright’s rate criteria are but one part of a broaderwork that explores themeaning andusefulness of cost of service as a regulatory concept. The two editions of Bonbrightdemonstrateathoughtfulapproachtoaconstantlychanging,inexactindustry.Wesummarizethehighlightsbelow,quotingextensivelyfrombotheditions.There are four important theoretical concepts that are often overlooked or omittedwhen astakeholderadvocateinvokesCCOSSasthebesttoolfordesigningrates:

• “Cost”maybebetterthan“thepublicinterest”asastandard,butit’sstillinexact;• Utilitycostssendpoorpricesignals;• Themoregranulartherate,themoreimpracticalisa“cost-based”rate;and• Ageneralratecaseisnottherealworld.

“Youkeepusingthatword.Idonotthinkitmeanswhatyouthinkitmeans.”InigoMontoya,ThePrincessBride


WhileBetterThan“ThePublicInterest,”CostRemainsanInexactStandard

OneofBonbright’sprimarygoalswastodisplaceapublicinterestandsocial-welfarestandard41forratemaking,whichhesawastheepitomeof“extremevagueness”and“notarealstandardatall.”42Rather, thepublic interest standardwas “amere formofwordsofhighlyemotionalcontent, invoked as an instrument of persuasion by people who have at heart much moreimmediate interests in public utility tariffs—interests often, but not always, of a self-seekingnature.” While attempting to add meat to a vague standard, Bonbright recognized that hisrecommendationthatcommissionssetcost-basedrateswasnosilverbullet.Intheabsenceofatruly competitive market, costs must be constructed. What costs to include?What time orgeographic selection to make? Should a commission consider only costs incurred, orhypothetical costs avoided, or both?Why are avoided costs and externalities considered fordemand-side management program design but not for rate design? How can customerelasticity of use be effectively measured? While the term “public interest” is noticeablysubjective,costintheratemakingsenseisalsoaconstructionthatdependsonperspective.

UtilityCostsSendPoorPriceSignals

Bonbright roughly categorized his rate design criteria into three functions that ratemakingshouldideallyaccomplish43:

• TheCapitalAttractionFunction,bywhichitprovidessufficientmonetaryincentivesforeconomicagentstodevotetheirresourcestoprovidingservicesthatpeoplearewillingandabletopurchase;

• TheEfficiencyIncentiveFunction,bywhichitensuresthatsuchservicesareprovidedatthelowestpossiblecost;and

“[T]heverymeaningtobeattachedtoambiguous,proposedmeasuressuchasthoseof‘cost’ or ‘value’—an ambiguity not completely removed by the addition of familiaradjuncts,suchas‘out-of-pocket’costs,or‘marginalcosts,’or‘averagecosts’—mustbedetermined in the light of the purposes to be served by the public utility rates asinstrumentsofeconomicpolicy.”

Bonbright1961,atp.290.“Oneofthereasonsforthepopularityofacost-of-servicestandardofratemakingnodoubt liesintheflexibilityofthestandarditself. ‘Cost,’ like‘value’isawordofmanymeanings,withtheresult thatpeoplewhodisagree,not justonminordetailsbutonmajor principles of ratemaking policy, all may subscribe to some version of theprincipleofserviceatcost.”

Bonbright1988,atp.109.


• TheConsumer-RationingFunction,bywhichitpreventswaste,ormoreprecisely,limitsconsumptiontolevelssuchthattheincrementalbenefitsreceivedbyeachindividualatthemarginequalorexceedtheincrementalcostsofprovidingtheservices.

He isclear inbotheditionsthat forratestructurequestions (spreadanddesign), theprimaryratemaking function should be consumer rationing. This recommendation has two practicalcomponents.First,thosesettingutilitypricesshouldexplorethe likelyconsequencesofthoseprices—howwillthepeoplepayingthemactinresponseandwhatwillthoseactionsmeanfortheutility’scosts?Second,therate-makershoulddeterminewhethertheresultingbehavioralchangesandconsequencesfortheutilityarewastefuloreconomicallyjustified,fromasocietalperspective.Discouraging wasteful use means establishing rates that will avoid wasteful utility costs.Bonbright focusedon theconsequencesofpricing thatwould reduceutility costs in the longrun. He identifies the most important costs for the purpose of developing prices as those“anticipated costs that can still be escaped orminimized by a control of output.”44In otherwords, rates should cause utilities to avoid costs thatwould be needlessly expended, ratherthan to recover costs that were already incurred. Yet CCOSSs rarely look at marginal—oravoidable—costs, and even less rarely look at long-run marginal costs, which economistsgenerally prefer.45Bonbright acknowledges why ratemaking is rarely done on the basis ofmarginal costs: such rates would inevitably recover more or less than the utility’s revenuerequirement, setting the conditions for either utility extraction of monopoly profits or theconfiscationofautility’sprivatepropertyprohibitedbytheConstitution.Determiningwhich costs resulting from the use of electricity arewasteful—as opposed toeconomically justified—requirestakingasocietalperspective.Bonbrightopinedthatwhenitcomes to lookingat efficiency froma consumer-rationingperspective, full costs andexternalcostsmatter.BothBonbrighteditionsassertthatdifferentconcernspredominatewhensettinga revenue requirement versus establishing a rate structure. With respect to revenuerequirement, “the relevant costs are enterpriser costs and not social costs [and] . . . theimportantequationisbetweentotalrevenuesandtotalcostsratherthanbetweentherateforanyspecificserviceandthecostofrenderingthisparticularservice...”46Incontrast,withregardto rate structure, social costsmattergreatly.Utility costsare inadequateunless they roughlyinternalizeexternalcosts.47ThismeansthatpricesignalsbasedonaCCOSSthatfailtoaccountforexternalities,suchascarboncosts,areincomplete.Importantly,internalizingexternalitiesisnotthesameasthesocialengineeringthatBonbrightopposed.Bonbrightsuggeststhatthemoreimportantquestionis:howcanratesbedesignedoveralltopreventwaste?To that,wewouldadd:whatelse shouldandcaneconomic regulationdo toprevent“wasteful”utilitycosts,asBonbrightdefinesthem?DecadesafterBonbright’sseminalthinking, it is apparent that utilities plan and build their systemsof generation, transmissionanddistributionbasedon“representative”customers,particularlyforthemassmarket.Thisisunderstandable: individual mass-market customers come and go, along with the electricity-usage affecting behaviors peculiar to them. Utilities cannot adjust the physical system in


responsetoeachandeverysuchindividualcustomer.Theycanonlyobservewhatishappeningatamacro-scale,andputinmotionthechangestothesystemthatwouldreducewasteifthosechangespersist.Regulatorscanmosteffectively influencepotentiallywastefulutility costsbyinterjectingthemselvesfirmlyintheseplanningandbuildingdecision-makingprocesses,notbyhopingthatelectricityuserspayattentiontopricesignalsthatratessupposedlysendandthatutilitiesadjusttheirinvestmentsaccordingtouserresponse.

TheMoreGranulartheRate,theMoreImpracticala“Cost-Based”Rate

Bonbrightrecognizedthattheproblemsofdesigningratesbasedonsunkcostsgotlargerastherates at issue got more precise. In other words, general rate levels are easier to set thanindividualrateschedulesortheirbillingdeterminantcomponents.48Botheditionsagreeon,anddiscussatlength,thedifficulty—ifnotoutrightimpossibility—ofdeterminingcostofserviceforindividual rate schedules or, certainly, individual customers. His objection was partiallyphilosophical, as the idea that “the rates charged to any single customer within that classshould cover the costs of supplying this one customer” would reduce the “theory of ratestructurestoameretheoryofcostdeterminationthroughtheaidofmoderntechniquesofcostaccounting and cost analysis.”49Indeed, “the problem of measuring the costs of separateclassesandamountsofserviceisoneofthemostcontroversialproblemsinpublicutilityratetheory.” 50 Rates inherently create subsidies. 51 Even with better data and more powerfulcomputingcapabilities,thisproblemhasnotchangedoverthedecades.52

AGeneralRateCaseisNottheRealWorld

Perhaps themost refreshingpartofbotheditionsof thisseminalworkonratemaking ishowBonbrightplaceseconomic regulationand ratemakingwithin the largercontext.General ratecase assumptions do not reflect the real world, and utility costs are not the only ones that

“In a very real sense, . . . the supply of any one type of service to thousands ofratepayersatdifferent locationsconstitutesthesupplyofadifferentproducttoeachcustomer. Similarly, service rendered at any one time is not the sameproduct as anotherwisecomparableservicerenderedatanothertime.Butthesemillionsofdifferentservicedeliveriesbyasinglepublicutilitycompanyareproducedincombinationandattotalcosts,mostofwhicharejointorcommoneithertotheentirebusinessorelsetosome major branch of the business. Under these circumstances, the attempt toestimatewhatpartofthetotalcostofoperatingautilitybusinessconstitutesthecostof serving each individual ratepayer or class of ratepayerwould involve a hopelesslyelaborateandexpensivetypeofcostanalysis…themostthatcanbehopedfor isthedevelopmentoftechniquesofcostallocationthatreflectonlythemajor,morestable,andmorepredictablecostrelationships.”

Bonbright1988,atpp.391-392.


matter forpurposesof settingprices.Moreover, neither regulators reviewing rate filingsnorutilitiesmaking them are unbiased, evenwhen applying cost of servicemetrics. In this vein,Bonbrightgaveregulatorsthreeimportantpiecesofguidance.First,use the right tool for the right task.Utilitypricesdonotoperateonpeople’sdecisionsinvolvingelectricityinavacuum.Indeed,thestewofincentives,barriers,policies,andpracticesapplicabletoanygivendecisionthatmightaffectthecostofelectricitysupplyoruseismind-boggling. Similarly, regulators do not make decisions in a vacuum. Applying principles andpracticesderived fromhistorical situations thatno longer applywill not servemodernutilityratepayers. Themost important question to ask is:What are the objectives that a regulatorwantstoaccomplish?Answeringthisquestionthenpromptsanother:Giventhelargercontextthat the regulator cannot control, what tools of economic regulation are appropriate toinfluencetheachievementofthoseobjectives?

Second, unintended consequences are probable. It is rare to find anymention of potentialunintendedconsequencesintestimonyproposingratedesignchanges.Yet,itwouldbeequallyrarefornonetooccur.Bonbrightadvisedthatthe“administrationofanystandardorsystemofratemakinghasconsequences,someofwhicharecostlyorotherwiseharmful,”andthereforemight necessitate making less economically based choices.53Any rate design changes thatcommissionsultimatelyapprovewill land ina volatilemixofnew technologies,demographicchanges, and changes in values, beliefs, and feelings. New rate designswill almost certainlyaffect the economics and timing of end-user decisions to adopt or reject particulartechnologies.Thesechoices,inturn,willlikelyaffecttheenvironmentaloutcomesofelectricityuse. Somemay alter the role that electricity prices play in the economic vitality of a givenserviceterritory.Othersmayaffect therelativeburdensofelectricitybillscomparedtootherhouseholdneeds.Thesedifficultquestionscontinuetoproliferate.Finally, forget about perfect competition, optimums, or the ideal. One of the wordsmostfrequently used in economic regulation—from resource planning to pricing—is “optimal.”

“A mixed economy has at its disposal a variety of economic controls ancillary oralternativetothatofthepricesystem,withtheresult that, intheperformanceofanygivenfunction,pricemaymerelyassumesomeshareingettingthedesiredresult.Whatthisshareshallbeisaquestionofpolicythatcannotbedecidedinthesamewayforalltypesofprices,oreven forallpublicutilityprices…On thecontrary,wemustassumeresponsibilityforanattempttoresolvescontroversiesastohowprogramsofratemakingshouldcooperatewithotherpoliciesofpublicutilityoperationandregulationinsecuringbasicobjectivesofpublicpolicy.”



Bonbrightcautionsagainstitsuseinconnectionwithpricing:“Satisfactoryresults,notidealoroptimumresults,areallthatcanbeexpectedoftheablestgroupofratemakers.”54

TheTheoryInformstheTool

ACCOSScannotanswerwhatwillcreateordeferautilitycostinthefuture;itattemptsonlytoallocate costs the utility has incurred to serve its entire expected customer base at varioustimes in the past. Thus, while some assert that a CCOSS can answer today’s rate structurequestions, Bonbright himself would likely disagree. But he would just as likely agree that aCCOSS can provide one perspective on these questions, along with other evidence andconsiderations.Todothatwell,however,itmustbedesignedappropriately,withassumptionsthat are both reasonable and logically consistent. The next section looks at some of thesespecificmethodologicalissues.

“[The] task of ratemaking or rate regulation is that of adapting utility rates to a largereconomic environment, including a universe of nonutility prices and wages on whichtheserateshaveonlyalimitedrepercussion...”



FLAWSINTHECCOSS:THETOOLEvokingtheCCOSSastheoreticalsupportformass-marketratedesignisdubious.Butthetoolis also problematic in practice. In theprocess of assessing the cost causationof various rateclassesandspreadingautility’srevenuerequirementtothem,theCCOSSanalystmakesmanyassumptions,includingthat:

• Allofautility’sserviceshavebeen identified,andeachrateclass’susagepatternsarefullyunderstood.

• In states using embedded CCOSSmethodologies, the only costs that are relevant arethose found in the utility’s accounting system, because those are the costs thatcomprise the revenue requirement. In states applying marginal cost CCOSSmethodologies,costsfromtheutility’smostrecentintegratedresourceplanandfuturecostsforutilitydistributionequipmentarerelevant,althoughtheresultingstudymustthenbetranslatedintoembeddedcostterms.

• Theimportanteffectsofagivenratepayerusingornotusingutilityserviceswillappearintheutility’saccountingsystemorotherdataitcommonlytracks.

• Customerclasses,astheyarecurrentlyconstructed,containaccountswithsimilarusagepatterns, such that applying the same rate to everyone in the class does not undulydiscriminateagainstanyparticularaccount.

• Coststhatwerejointlyincurred—likedistributioncostsandsoftware—canbepreciselyallocatedtocustomers.

It iscriticalthattheseassumptionsareexplicitandthatcommissionsstructureprocessesthatallowforquestionsabouttheiraccuracyandapplicability.WhenBonbrightwrotePrinciplesofPublic Utility Rates (both editions), commissions considered CCOSSs informative, but notdispositive.Herightlycriticizedanymodelthatattemptedtoallocatejointandcommoncostsdowntothelastdecimalandaccountasameanstodecideratespreadanddesign.Asutilitiescollect more data, the CCOSS is being tasked with doing precisely that. This section looksspecificallyattheflawsinherentinthetoolitself,ofwhichthreeareparticularlyimportant:

• TheCCOSSastraditionallyprogrammediseffectivelya“blackbox”forwhichtheutilityistheonlypartywithfullaccesstodata;

• Thetool’sassumptionsforallocatingjointandcommoncostsmaynotsyncwithcurrentelectricindustrypractices,eitherastheyareorastheymaybe;and

• TheCCOSS,byitself,doesnotprovideaneffectiveframeworktomeasureanymetricof“success”besideswhetheraparticularcustomerclasshasmetitsrevenuerequirement.

TheCCOSSisaBlackBox

Utilitiesunquestionablyprovideasignificantamountofdatathroughthediscoveryprocessinaratecase.Thisdoesnotnecessarilymean,however,thatallpartiesintheratecasehaveaccessto the information sufficient to evaluate the CCOSS and its support for any rate design


proposalsmadeintheratecase,bytheutilityorothers.ThedevelopmentofrateclassesfromloadresearchisafoundationalsteptoeveryotheraspectofeachCCOSSdevelopedforaratecase,butitcanalsobethestepforwhichtheleastinformationisavailable.ACCOSSrequiresload data from individual customers at a daily, hourly, or sub-hourly basis. This data isfrequently confidential, and it may be neither recent nor robust in terms of the variety ofaccountswithinagivenrateclassthatwereincluded.The original rate classes, still used bymany utilities, were based on assumptions about thetypes of accounts that were “similarly situated.” But who is similarly situated? Rate classesincludewithinthemaccountswithdifferentservice,voltage,load,orothercharacteristicsandthe broader the rate class—such as the residential and small commercial—the more thedifferencesandthewidertheirrange.Arecustomerssimilarlysituatedwithrespecttotheiron-siteequipment(electricheatornaturalgas),orhowtheirbuildingsorequipmentrelatetotheutility systems of distribution, transmission, and generation (voltage, usage times)? Or whatabout an area that regulators have generally feared to tread in—the purposes for whichcustomers use energy (dry cleaners or corner grocery, hospital or office building, renting orvacationing)? Some have proposed addressing only “obvious” exceptions to the notion thatcustomersare similarly situated—suchaswhether theaccounthas rooftopsolarornot—butthequestionofwhatanobviousexceptionishasrarelybeenapproachedwithmethodologicalvigor.55Thesequestionsareparticularlypertinenttotheresidentialclass,whosediversityinthehow andwhen of using utility services can be large, let alone the cost implications of thosedifferences. For example, residential accounts may include vacation homes, small studioapartments,mansions,anddwellingsthatheatwithnaturalgasoroil(seeFigure2).56

Figure 2: Variations in hourly demand across a single day for five residential customers inColorado57


Historicalloadresearchcompiledtoshowtrendsovertimecouldprovidesignificantinsightastothevalidityofrateclassconstructions,butthismaynotbeavailable,eitherbecause ithasbeen years since the last rate case or because load research is conducted infrequently orinformally.58Many utilities are still in the process of installing AMI or have only recentlycompleted its installation,meaning that for certain customers—and particularly those in theresidential and smallbusiness classes—utilitiesmaynotevenhavea full test year’sworthofhourlyorsub-hourlydata.UtilitiesthatlackAMImayinstallspecialintervalmeters,butthisisusually only on a small sample of customers.59Furthermore, the historical purpose of loadresearchhasbeentoupdatedataaboutexistingrateclasses; thepurposehasrarelybeentoexamine whether customers within a given class remain “similarly situated” or explore theadditionorsubtractionofrateclasses.Designingtheloadresearchtosupporttheconstructionorde-constructionofthetraditionalrateclassesmightrequiredifferentloadresearch.An additional load research issue is that to varying degrees, all states consider individualelectricityconsumptiondatatobeprivateandrequirethatitbekeptconfidential.Sometimesratecasestakeholderscanreceivethedatasolongasaccountdetailsareremoved,butthiscanbe time-consuming for utilities to produce and for stakeholders to evaluate. To avoid theseproblems,someutilitiesconductstatisticalsamplesofdifferentcustomers’electricitydemandor consumption. Lacking the base data, stakeholders cannot assess whether the statisticalelementsofthedataarecompleteorcorrect—animportantconceptgiventhatdataareoftenprovided as averages, which can mask significant non-normal data distributions and createcross-subsidies between accounts that lie on either side of the average.60The gap betweenutilitiesandstakeholderswillcontinuetowidenas“bigdata”becomesmoreprominentintheelectricindustry.

CCOSSMethodologyNeedsRefreshing

There are three core steps in a CCOSS: functionalization, classification, and allocation (andmultistate utilities may also jurisdictionalize costs to the state’s service territory their filingfocuseson).Appendix1providesmoredetail.Naturally,CCOSSsrequiremanymethodologicalchoices, including the type, granularity, andvintageofdata touse in constructingallocators.TheNARUCManualclarifiedwhichmethodologiesregulatorsandtheirstaffsconsideredmorevalid (in 1992); however, it provides a series of alternatives rather than presenting a single,unified approach.61The decades since have brought, and continue to bring, upheavals inelectricindustryapproachestoplanning,engineering,andoperations.Ultimately,asignificantchunkoftheCCOSSiscomprisedofcoststhatarejointlyandcommonlyincurred. Unlike a utility-owned streetlight, the costs associated with a substation, atransformer, a billing system, a transmission line, or a senior executive’s salarymay not bereadilyassignedtoaparticularcustomerclass.Analystsarethusforcedtocreateassumptionsabout how to divvy up those costs. Sometimes these assumptions are fully explained and


argued inaratecase,andsometimestheyarenot.Amongtheassumptionsananalystmightmake:

• Functionalizingadministrativeandgeneralcostsassociatedwithofficespace.Autilitymayownorleaseandofficebuildingtocentralizemuchofitsstaff.TheNARUCManualproposesthatanalystsfunctionalizethecostsforanofficebuildingbasedonthesquarefootageofspacededicatedtotheproduction,transmission,distribution,andcustomerservicefunctions.However,acompetingmethodistofunctionalizethesecommoncostsinproportiontohowmucheachcomprisesoftheoverallrevenuerequirement.62

• Classifyingthecostsassociatedwiththedistributionsystemascustomer-related.The

secondary distribution system includes the lower-voltage circuits, poles, wires, andservice drops that connect to individual meters. Distribution costs are sometimesclassified as a mix of “customer-related” costs and “demand-related” costs, and thisdesignation can influence ratedesign. For example, a utilitymightuse the “minimumsystem”approachtofindthatthereisabare-bonessetofdistributionequipmentthatisnecessary to serve customers within that utility’s service territory. Because thismethodology classifies that bare-bones system as customer-related, its costs areallocatedbasedonthenumberofcustomers—whichtendstoplacemorecostsontheresidentialcustomerclass,which isusuallythe largestcustomerclass.SeeAppendix3formoreinformationonalternativeapproachestoclassification.

• Allocating the costs associated with meter reading and billing between customer

classes.Meter-readingexpensescanincludemanualreads,thecostsofsendingtruckstocollectmeter reads,or thecostsofestablishingasoftwaresystemtocollectmeterintervaldata.Autilitymayallocatethecostsassociatedwithreadingcustomers’metersbased on the cost of the meters used by each customer class, the proportion ofmetering costs as a share of the overall revenue requirement, or some othersimplification. Often these weighting factors are applied across customer-relatedexpenses, like the costs associated with maintaining call centers. For residentialcustomers, theseapproachesmightdouble theexpensesallocated to thosewhohaveproduction meters for DERs, even as it averages out the higher costs required formanualmeterreadswiththelowercostsrequiredformeter-readingtrucks.

New technologies like DERs are changing not only when, how, and why customers useelectricity, but also how utilities design, build andmaintain their systems. For example, thevalidityofmethodsusedtoallocatedistributioncostsdependsonhowdistributionsystemsareplanned—anareathatisincreasinglybeingreevaluatedinstateslikeCalifornia,NewYork,andHawaii (and under discussion inMinnesota,Maryland, and New Hampshire, among others).Whereasdistributionplannersmighthaveoriginallyreliedonhistoriccustomerusagedatatoassessnew lineextensions, theynowmayneed to considerhow to integratemicrogridsandnet-zero neighborhoods. Stakeholders considering the utility of the future also considerwhether to unbundle, price, and provide choice around such utility services as reliability orvoltage,andthemoreesotericconceptsofcomfortorconvenience.


Allocationmethodologies that spread revenue requirementsanddesign ratesbasedonwhatthe systemwas rather than on what it could become will hinder efforts to influence whatregulatorswant the system tobecome.CCOSSmethodologiesmust reflect conscious choicesabout the future of the system rather than past intended and unintended actions thatproducedwhatthesystemistoday.TheCCOSSDoesNotMeasure“Success”

AfterCCOSSanalystsfunctionalize,classify,andallocatecosts,theyreconstitutethetotal intorateclass-specificandschedule-specificrevenuerequirements.63Comparingtheserateclassorschedulerevenuerequirementstotheactualrevenuescollected(ortobecollected, infuturetestyearjurisdictions)atcurrentandproposedratesinformstheanalyst’soverallratespreadand rate design recommendations. A final test is a demonstration that the proposed andapprovedratesproducetheproposedandapprovedrevenuerequirementsundertheassumedlevelsofbillingdeterminants.The critical question rarely addressed in a rate case, however, concerns the effects of ratespreadandratedesigndecisions.There isnoprecisewaytomeasurewhether rates thatareadoptedunderthepretextthattheyare“justandreasonable”actuallyturnouttobethecase.What first- and second-order effects arise from a particular rate design? What shouldcommissions look for? How will they observe what happens? Many commissions haveprocessesinplacetoobservetherevenuesandearningstheratesproduceovertime,checkingtomakesuretheyarenottoohigh.Isthisutility-centricmeasuretheonlymeasureofsuccessneeded?Ordoweneedmore?

“Public utility counsel have sometimes argued that once a company’s total revenueentitlementshavebeendeterminedbya commission,thechoiceofapatternofratesthatwill yield the allowed revenues should be left to the discretion ofmanagement,whichwill then be in an impartial position tomake a fair apportionment of burdensamongitsdifferentclassesofcustomers.Thisisonlyahalf-truthbecause,amongotherreasons,autilitycompanyisconcernednotjusttosecureratesthatwillpresentlyyieldthe approved fair rate of return, but to develop a pattern of rates thatwill promotegrowth of earnings and thatwill protect these earnings against business depressions.Thebettertheutilitymanagement,thegreateraretheseconcerns.”



AnInexactToolProducesInexactResults

Experiencedmodelersoftensay:“Allmodelsarewrong;somemodelsareuseful.”ACCOSSisnoexception.Applyingthemethodologiesthatareusedtoallocatecosts intheCCOSStothedesignofratescreatesdistinctionsdowntothethousandthofacent.Theresultisinformation,but it is not “true.” Even though the informationonwhich theCCOSS is based is improving,fundamentalflawsinthetoolitselfanditsuseremain.ACCOSSisoftenablackbox.Alitigatedprocessmaynotbethemostconstructivewaytoidentifyallofthedataandchoicesnecessarytomakeausefulmodelandresult.ItisnotusuallypossibletoredotheCCOSSwithchoicesitspreparerdiscarded.UsingtheCCOSSmodelinaratecaseproducesabattleamongadwindlingpoolofexperts,withcomplicatedargumentsthatexcludesomepartiesandalmostallgeneralstakeholders—nonprofits, trade associations, and local governments, who have neither theabilitytohireoutsideexpertsnorlong-lastingrelationshipswiththeexpertswhoremainactive.Members of the public can play virtually no role in advocating on CCOSS disputes or therecommendationsbasedonthem.Inthefinalsection,we’llproposesomepossiblesolutionstothedilemmasregulatorsfindthemselvesinwhenthinkingaboutelectricrates.


LOOKINGTOTHEFUTURE:REPAIRORREPLACE?WhethertheCCOSStoolprovidesananswerdependsonthequestiononeisasking.Whetherthe CCOSS provides a good answer depends on questions and considerations outside thecapabilities of the tool. Whether the CCOSS produces a desirable outcome, with as fewunintendedandadverseconsequencesaspossible,dependsonseveralfactors:

• Havedecision-makersclearlyarticulatedthedesirableoutcomeandwaysthatitcanbeobservedandmeasured?

• DothoseusingtheCCOSSunderstandthefactorsthatwouldcontributetoadesirableoutcome?

• Aredecision-makersreadyandabletotakeactionifthatoutcomedoesnotappear,orifunintended,adverseconsequencesdoappear?

Considering the flaws inherent to the concept of the CCOSS and its execution, the questionemerges:isthisthetoolwewanttousetodefinedesirableoutcomes?Isthisthetoolthatwillhelp regulators make decisions that keep utilities whole while allowing competitive energymarketstothrive?Isthisthetoolthatwillleadtocleanerenergyandreducecarbonemissions?Is this the tool thatwill helpmore customersmake goodenergy choices, orwill it eliminatechoices?Theanswersdependonwhichquestionsaregulatorybodyiswrestlingwith.TheCCOSS,asitiscurrentlyused,istetheredtothepast.Projectionsofelectricitysalesrelyonpast experience and identifiable relationships (e.g., construction, gross domestic product) toestimatethefuture.Accountingdata isbasedoncoststhathavebeenincurred.EvenCCOSSsthatapplymarginalcostprinciplesmayshiftbetweenshort-orlong-termmarginalcost,andbevagueabouttheassumptionsmadetocalculatethenextunit’scost.Becausetheindustryisinthemiddleofashift,notallofthesepastexperiencesordatawillberelevantintenorevenfiveyears.Ratedesignisanimportantdimensionoftheenvironmentinwhichutilitiesandtheircustomersand competitors interface. Tonavigate through this environment, rate-settersmust focusonwhatishappeningnowandwhattheythinkwillhappensoon,adjustingasrealitydiffersfromwhatwasexpected.Lookingbackwards,tothepast,hindersratherthanhelpsthenavigation.IftheCCOSSistohaveaplaceinthisevolution,stakeholders—includingregulators,utilities,andinterveners—will need to refresh both the theory and the practices behind it.Wehave foursuggestionsforthesestakeholders,butparticularlyforregulators:

• Bespecificabouttheroleofpricinginthequestionsyou’retryingtoanswer;• Prioritizeloadresearch;• MakesureaCCOSSistransparentandusable;and• Refresh the methodologies in the CCOSS to reflect current—and possibly future—

practices.


Regulators should be specific about the role of pricing in the questions they’re trying toanswer. Regulators are again being asked to set the lines of interaction between electricutilitiesandnewmarketentrants.Inthefuture—andcurrentlyinstateslikeHawaii—theymaybe asked to define or oversee new electric utility services, like voltage control or enhancedreliability. Thesequestionshave implications for rate structure, and certain ratedesignsmayreducecompetition inorevendemandforneworexistingservices.Outcomesofratesettingcouldbelesscustomerchoiceandfeweropportunitiesforthirdpartiesorutilitiestoinnovateandelectricityuserstoobtaindesiredservices.Asinthe1970sand1980s,regulatorsmaylooktoeconomistsforadviceonwhatsortsofratestructuresareappropriatetopreventwaste(andhowtomeasurethatoutcome).Theymayalsowanttoexaminewhetherothertoolsthattheyalready use are relevant to rate structure. Integrated resource planning (IRP), for example,offersinsightintheformofitslong-termperspective.IRPsmaycontaingranularassumptions,buttheyarefundamentallyaboutmakingchoicesthatcouldmeetmultipledifferentalternativefutures,abenefitthatisincreasinglyasrelevanttodistributionplanningasitistopowersupplyplanning.Ultimately, it is important to consider what is and is not within the purview of utilities andregulators.Bonbrighturgedregulatorstousetherighttoolsfortherightjob.Sometimes,thatinvolvessettingprices,andothertimes, it involvessteppingbacktoallowotherpolicymakersandthemarket to takea leadingrole.Utility ratedesign isablunt tool,andother toolsmayoffermuchbetterprecisionandtheabilitytocontrolunintendedconsequences.Forexample,ifutilitiesareexperiencing revenueerosion—aquestionof fact—then therearemultiplepolicyalternatives toconsider, includingwell-designed revenuedecouplingadjustments64thatallowformorestablerevenuecollectionregardlessofelectricitysales.Loadresearchshouldbeapriority.Ifpatternsofelectricityuseandtheirrelationshipstokeyindicatorslikegrossdomesticproductarechanging,thentraditionalforecastingmethodsmaynoteffectivelypredict the future. Identifying thesepatterns requiresnew load research, andthat takes time. Utilities report on the same data and metrics that they did decades ago,despitetheincreasingavailabilityoftoolsthatcanmeasureandrecordhowelectricityisused.Weneedtoaskbroader,deeper,betterquestionsaboutloadifwehopetoinfluencehowtheproductionandapplicationof electricityunfoldsover thenextdecades. This research shouldinformnotonlypricingbutalsoplanningacrossthegeneration,transmission,anddistributionsystems. Ideally, the questions and sources for this research would be developedcollaboratively, in a regulatory proceeding that invited as many different perspectives aspossible.TheCCOSS shouldbe transparentandusable.Traditionally, theutilitypreparesandfilestheCCOSS,andthenintervenerscritiqueit.Theunderlyingdatamaynotbeavailable,becauseitisnotinaformatthatmakesreviewpractical,hasnotbeenaskedforcorrectly,orisconfidential.The Arizona Corporation Commission took steps to address this by requiring Arizona PublicServiceCompanytoprovideatransparentcostofservicemodelaspartof itsratecasefiling,although parties could debate how well this worked.65We recommend taking this a stepfurther by having a third party oversee the creation of the CCOSS. This third party would


transparently document all data and methodologies. Having a third party whose role is toindependently oversee the CCOSS would help in multiple ways. Because they would haveresponsibilityforthecalibrationofthemodel,theycouldcreatemultiplealternativescenariosandidentifywhichassumptionsaresensitivetotheoutcomes.Theycouldensurethatdata isstatisticallymeaningfulandlogicallyconsistentthroughoutthemodel.Theycouldalsoidentifydatagaps.Webelievethatacollaborativeprocesscouldprovideabetterframeworkforaskingquestionsandexploringdifferentopportunities.CCOSSmethodologiesneedrefreshing.TheNARUCManualhasbeenadoptedinmost,butnotall, states—but itwas lastupdated in1992. Justas theNARUCManual refinedconcepts thatwerenotfullysettledinBonbright’stime,itistimetorefineconceptsthatwereappropriatein1992,butmightnotbetoday.ThismeansreexaminingutilitypracticesandassumptionsaroundsystemplanningandoperationstoseeiftheCCOSScorrectlyreflectsthemintheircurrentformorhowtheymaychangeinthenear-orlonger-term,basedonnewutilitysoftwareandrelatedprocess changes. For example, as battery storage and advanced inverters improve the loadfactorofrooftopsolarinthenearterm,CCOSSloadassumptionsandmethodologychoicesmayneedtoshifttoreflectthatchange.


Conclusion

The growing complexity of energymarkets and utility regulation has fueled a desire to relymoreandmoreontheCCOSSasatoolnot justtoconstructclassrevenuerequirements,butalsotodesignrates.Thisisnotthefirsttimethatratestructurehasbeenfrontandcentral—itoccurseverytimethereisachallengetotheelectricutilityindustry’scompetitiveness.Boostingthe CCOSS into its new role are two trends: big data, and the hope for quantitative andobjectivedecision-makingsupport.ButtheCCOSSisnotinnatelyobjective,andparties’relianceonitcallsintoquestionnotjustwhethertheassumptionsandmethodologieswithintheCCOSSarevalid,butalsowhethertheCCOSSisstillthecorrecttooltobeusingtomakethedifficultdecisionsthatregulatorsface.Ultimately,there isno innately“good”or“bad”ratestructure.Thereareonlyratestructuresthat, over time,move you closer to your goals and those thatmove you further away fromthem.Regulatorsshouldaskthequestionsdescribedbelowtohelpensuretheyunderstandtheobjectivesandrealitiesofratedecisions.Thedesiretomakeratestructure,andparticularlythedesignofcertainrateschedules,aproblemthatissolvedbyapplyingtheCCOSSis,infact,partoftheproblem.


APPENDIX1:StepsintheCostofServiceStudy(ActualStudyStepsinBlue)

Theamountofmoneythatthatautilitymustcollectfromitscustomerstopayforcapitalcosts,manageexpenses likesalaries,andprovideareturnto its investors iscalledarevenuerequirement(andmorespecifically, a “base” revenue requirement, which is distinct from costs collected through riders oradjustments).ACCOSSmathematicallyderivesthe“costtoserve”particulartypesofcustomers—itusesengineering,load,andaccountingdatatoallocatetheutility’srevenuerequirementbetweencustomersbased on how they use the utility’s system and services. This concept is known as “cost causation.”Importantly,this isnotthecosttoserveanyoneuniqueindividual,butasimplified(typicallyaverage)resultbasedonhowthattypeofcustomertendstousetheutility’soverallsystem.ACCOSShasfourkeysteps—functionalizing, classifying, allocating, and constructing class revenue requirements—but thereareeightoverallstepsinthelifecycle,fromcreatingunderlyingdatatoverifyingthattheproposedrateswillleadtotherightamountofrevenue.

ConductLoadResearchMany of the important steps in the CCOSS require load data (generally for a twelve-monthperiod)andatadaily,hourly,or sub-hourlybasis.Utilitiesmayneed to install special intervalmetersoncustomers’premisesiftheirmetersdonotordinarilyprovidethislevelofgranularity.Loadresearchgenerallysamplesacross-sectionofestablishedcustomerclasses.Thisloaddatamaybenormalized,or“smoothed”,toremoveanomalousweatherinaparticularyear.

SettheRevenueRequirementTherevenuerequirement(whichmaybe“jurisdictionalized”toautility’sstateserviceterritory)istheamountoffundingautilityneedstocoveritscapitalandoperatingexpenses.Utilitiesandregulators make numerous key decisions in setting the revenue requirement, includingestablishingthehistoricorforecast“testyear”usedtocheckwhethercostestimatesarevalid.Some states treat rate cases in phases, startingwith the setting of the revenue requirement,followedby a secondphase for regulatory approval of rate structure.Autilitymayprepare aCCOSSforeitherphase,orboth.Aftertherevenuerequirementhasbeenset, thesubsequentstepsoftheCCOSS,aswellastheratedesignprocess,becomeazero-sumgame:coststhatarenotallocatedtosomecustomersmustbeallocatedtoothers.

FunctionalizeCostsFunctionalization is thefirststep inbreakingdowntheapprovedrevenuerequirement into itscomponentpartstodrawconclusionsaboutcostcausation.Ingeneral,functionalizationfollowsthecategoriessetintheFederalEnergyRegulatoryCommission’sUniformSystemofAccounts.66Thismeansthatcapitalandoperatingcostsarecategorizedbasedonwhethertheyrelatetothepowergeneration(“production”),transmission,distribution,customerservice,oradministrativeandgeneral(“A&G”)functions.ClassifyCostsNext,aCCOSSanalystmustanalyzeeachfunctionalizedsetofcostsanddecidewhetherthecostwascreatedbecausecustomersusedelectricity,neededelectriccapacity,orsimplyconnectedtotheelectricsystem.Thesecostsarethenclassifiedas“energy-related,”“demand-related,”or“customer-related.” This step informs how those costs are allocated in the next step—forexample,bykWh,kW,ornumberofcustomers.NotethatA&Gdoesnotobviouslyfitintoanyoftheseclassifications.


DevelopAllocatorsIn the allocation step, the analyst uses load research and historic billing determinants (andsometimesprojections) todevelopcustomerclass-specificmathematical“allocators”basedonthefunctionalizedandclassifiedcosts.SpreadCoststoClassesAfter costs are functionalized, classified, and allocated, they are reconstituted into customerclass-andschedule-specificrevenuerequirements.Thesecustomerclassrevenuerequirementsare compared to actual revenues collected from the customer class at current and proposedrates in order to provide perspectives on how to allocate changes in the overall revenuerequirement. If a given rate schedule does not generate enough revenue to cover what theCCOSS shows to be its allocated revenue requirement, and another rate schedule generatesmorethanwhatitshouldhave,partiesmayproposethataratedecreasebeassignedtotherateschedulerecoveringmoreandarateincreasebeassignedtotherateschedulerecoveringless.DesignRatesGiventherevenuerequirementallocatedtoeachrateclass,thequestionthenturnstothetypeofchargesbywhich thosecustomersshouldpay for theservice they take.Themostcommontypes,shownbelow,arecalled“billingdeterminants”:

CustomerorAccountChargeAfixeddollaramountpaideverybillingperiod.

DemandChargeA rate charged for the kilowatts (“kW”) of demand that the customer needs to useelectricity. The charge is generally based on the customer’s demand at either theirindividualpeak(“non-coincident”)orthesystem’soverallpeak(“coincident”).ThepriceforakWmaydifferdependingonthetimeofdayorseasoninwhichthepeakoccurs.

EnergyChargeA rate per kWh for the electricity used during the billing period. The ratemay differdependingonthetimeofdayortheseason inwhich itwasused,orboth. Itmayalsoincrease themoreelectricity theaccountuses inagivenbillingperiod (incliningblockrates)ordecreasethemoreelectricityisused(decliningblockrates).

ProvetheRevenueTheapprovedratesmultipliedbybillingdeterminantsmustroughlyequaltheapprovedrevenuerequirement.Billingdeterminantsmustbesomethingtheutilitycancount(suchasaccountsonagivenrateschedule)ormeasure(suchaskWhorkW)andeitherkeeparecordof(iftheratecaseisbeingdoneonahistorictestyearbasis)orforecast(ifafuturetestyearisbeingused)—inotherwords,theyhavetobeabletodemonstratethatthebillingdeterminantstheyareusingareappropriate.

1

Principles of Public Utility Rates 1961 Bonbright Criteria

The related, "practical" attributes of simplicity, understandability, public acceptability, and feasibility of application.

Freedom from controversies as to proper interpretation.

Effectiveness in yielding total revenue requirements under the fair-return standard.

Revenue stability from year to year.

Stability of the rates themselves, with a minimum of unexpected changes seriously adverse to existing customers. {Compare ''The best tax is an old tax.")

Fairness of the specific rates in the apportionment of total costs of service among the different consumers.

Avoidance of •undue discrimination• in rate relationship.

Efficiency of the rate classes and rate blocks in discouraging wasteful use of service while promoting all justified types and amounts of use:

(a) in the control of the total amounts ofservice supplied by the company;

(bl in the control of the relative uses of alternative types of service (on-peak versus off-peak electricity, Pullman travel versus coach travel, single-party telephone service versus service from a multi-party line, etc.).

1988 Bonbright Criteria (Differences Italicized)

The related "practical" attributes of simplicity, certainty, convenience of payment, economy in collection, understandability, public acceptability, and feasibility of application.

Freedom from controversies as to proper interpretation.

Effectiveness in yielding total revenue requirements under the fair-return standard without any socially undesirable expansion of the rate base or socially undesirable level of product quality and safety.

Revenue stability from year to year with a minimum of unexpected changes seriously adverse to utility companies.

Stability of the rates themselves, with a minimum of unexpected changes seriously adverse to ratepayers and with a sense of historical continuity. {Compare "The best tax is an old tax.)

Fairness of the specific rates in the apportionment of total costs of service among the different ratepayers so as to avoid arbitrariness and capriciousness and to attain equity in three dimensions: (1) horizontal {i.e .• equals treated equally); (2) vertical {i.e., unequals treated unequally); and (3) anonymous (i.e., no ratepayer's demands can be diverted away uneconomically from an incumbent by a potential entrant).

Avoidance of •undue discrimination• in rate relationship so as to be, if possible, compensatory (i.e .• subsidy free with no intercustomer burdens).

Static efficiency of the rate classes and rate blocks in discouraging wasteful use of service while promoting all justified types and amounts of use:

(a) in the control of the total amounts ofservice supplied by the company;

(bl in the control of the relative uses of alternative types of service (on-peak versus off-peak service or higher quality versus lower quality service).

10

Reflection of all the present and future private and social costs and benefits occasioned by a service's provision (i.e., all internalities and externalities.)

Dynamic efficiency in promoting innovation and responding economically to changing demand and supply programs.

34

Appendix 2:


APPENDIX3:TheThreeClassificationMethodologies

Theclassificationofcustomer-relatedcosts isparticularlycontroversial,asutilitiesmaybetemptedtorecover those costs through uniform fixed charges that are applied to each customer in a class. TheNARUCManualexplainsconsiderationsrelatedtoselectingbetweencertainmethodologies,butitdoesnottakeapositionontheuseofoneoveranother.67

Minimum-SizeMethod(“MinimumSystem”)An analyst applying this method attempts to construct a bare-bones version of the electricsystembasedonthenumberand locationofutilitycustomers,andsaysthatthecostsof thatminimumsystemarecustomer-relatedratherthandemand-related.

Zero-InterceptMethodAnanalystapplyingthisstatisticalmethodologyinterpolatesthesizeofthesystemwherethereiszerocustomerload,andcallsthecostsassociatedwiththatsizeofsystemcustomer-related.

BasicCustomerMethodAnanalystapplyingthismethodassertsthattheonlycostsdirectlyattributabletocustomersarethoseonthebasicserviceline,andthatallothercostsrelatetodemandorenergyuse.68(ThisapproachwasnotputforwardintheNARUCManual,buthasbeenacceptedinsomestates.)

There is no national consensus on the “best” classification methodology to use, and the approvedmethodologiesmayvarybystate:

Kentucky(2000)“As it has stated in numerous orders over the last decade, the Commission believes that thezero-interceptmethodology is themoreacceptableway todividedistributionmain costs intodemand-relatedandcustomer-relatedcomponents.”69Minnesota(2016)“TheCommissionispersuaded,onvalidtheoreticalgrounds,thattheminimum-systemstudiesover-allocatedistributioncoststothecustomercomponent.”70Oregon(1998)“Thezero-interceptapproachcanbeusedinlimitedcircumstances,butisnotsufficientlyrobusttobeusedforalldistributioncostcalculations.”71Indiana(2011)“Inaddition,wefindtheminimumdistributionsystemcalculationtobelesssubjectivethanMr.Heid'szero-interceptmethod.”72


ENDNOTES1JamesC.Bonbright,PrinciplesofPublicUtilityRates(1961)[“Bonbright1961”];JamesC.Bonbright,AlbertL.Danielsen,DavidR.Kamerschen,PrinciplesofPublicUtilityRates(1988)[“Bonbright1988”].2Outofits198pages,theNARUCManualincludesonlyoneparagraphaddressingtheuseofcostofservicestudiesindesigningrates.Seepage13.Thatparagraphstates“Regulatorsdesignrates,thepriceschargedtocustomerclasses,usingthecostincurredbyeachclassasamajordeterminant.Othernon-costattributesconsideredbyregulatorsindesigningratesincluderevenue-relatedconsiderationsofeffectivenessinyieldingtotalrevenuerequirements,revenuestabilityforthecompanyandratecontinuityforthecustomer,aswellassuchpracticalcriteriaassimplicityandpublicacceptance.”3Bonbright1961atp.367.4Thecommonlyreferencedcustomerclassesareresidential,smallcommercial,largecommercial,andindustrial.Thelattertwo,inparticular,generallyhavemorethanoneratescheduleassociatedwiththem.Whatcharacterizesresidentialandsmallcommercialratedesignsmorethananyotherfeatureistheabsenceofademandcharge.5SeegenerallyCathyO’Neil,WeaponsofMathDestruction(2016).6BasedonEQResearchgeneralratecasedataaccessedFebruary9,2017.7PublicServiceCommissionofWisconsin,6690-UR-124,DirectTestimonyofRondaL.Ferguson(May15,2015)atpp.8-11,availableat:http://psc.wi.gov/apps40/dockets/content/detail.aspx?dockt_id=6690-UR-124[citesBonbrightatp.2](WPSC“installsstandardsizefacilitiesforresidentialcustomers…[and]hasnocontroloverequipmentthatcustomerschoosetoinstall…”).8Thistrade-offhaslongkeptsomecommissionsfromapprovingcustomerchargeincreases.See,e.g.,WashingtonUtilitiesandTransportationCommission,DocketUE-140762consolidated,Order08,March25,2015,p.91(“TheCommissionisnotpreparedtomoveawayfromthelong-acceptedprinciplethatbasicchargesshouldreflectonly“directcustomercosts”suchasmeterreadingandbilling.Includingdistributioncostsinthebasicchargeandincreasingit81percent,astheCompanyproposesinthiscase,doesnotpromote,andmaybeantitheticalto,therealizationofconservationgoals”).9Largecommercialandindustrial,butgenerallynotstreetlightingaccounts.10OklahomaGas&Electric,DocketNo.PUD201500273(December18,2015);ArizonaPublicServiceCompany,DocketNo.E-01345A-16-0036(June1,2016).11OklahomaCorporationCommission,PUD201500273,DirectTestimonyofAhmadFaruqui(December18,2015)atp.4,availableat:http://imaging.occeweb.com/AP/CaseFiles/occ5248973.pdf.12See,e.g.,DirectTestimonyofH.EdwinOvercastonbehalfofTucsonElectricPowerCompanyandUNSElectric,Inc.,DocketNo.E-00000J-14-0023.13ThissectionisdrawnfromKahn(TheEconomicsofRegulation:PrinciplesandInstitutions);CharlesF.Phillips,Jr.,TheRegulationofPublicUtilities(1988);andPaulJ.GarfieldandWallaceF.Lovejoy,PublicUtilityEconomics(1964).14SeeSmythv.Ames,169U.S.466(1898);FederalPowerCommissionv.HopeNaturalGasCo.,320U.S.591(1944).15Bonbright1961atp.287.


16See1stDistrictAppellateCourtofIL,Chicagov.IllinoisCommerceCom,133Ill.App.3d435,5/13/85,atp.444(“Moreover,althoughclasscostofserviceistheprimecriteriontheCommissionlooksatindeterminingtheproperallocationamongcustomerclasses,itisnottheonlycriteriontheCommissionmayconsider.Section32ofthePublicUtilitiesActallowstheCommissiontoconsider‘otherrelevantfactors.’(Ill.Rev.Stat.1983,ch.1112/3,par.32.)Oursupremecourthasheldthatthoseotherrelevantfactorsarenotexclusivelyrelatedtocostfactors…Considerationsofratecontinuityorthedesirabilityofmakingchangesgraduallycanproperlylimitchangesinratedesign.”);seealsoInternationalMinerals&ChemicalCorp.v.Mayo,336So.2d548(1976Fla.LEXIS4463)atp.551-552(“Practicalconsiderationsalsomilitateagainstmakingcostofservicetheexclusivecriterioninratesetting.Virtuallyeverycourtconsideringthematterhasrejectedoutofhandarulethatwouldreduceratemakingtoanexerciseincostaccounting…FactorslikethosetestifiedtobyTECO'sexpert,viz.,‘ratehistoryandexperienceoftheutility,theconsumptionandloadcharacteristicsofthevariousclassesofcustomers,valueofservice,publicacceptanceofratestructureswhichhavebeenineffectinthepastwithoutseriousdissatisfaction,andconservationofenergy’wereallproperlyconsideredbythePSC,eventhoughthesefactorsarenotspecifiedbystatute”);1975Mich.PSCLEXIS3,Nov.24,1975,U-4698&U-4718(“TheCommissionhasrepeatedlystatedthatcostofserviceisnotthesolecriterionforthestructuringofrates…SuchrigidanddoctrinaireadherencetocostofservicedoesnotprovideareasonablebasisfortheCommissiontodesignrates.”);1976Wash.UTCLEXIS4,Sept.27,1976,U-76-1(“Insofarasrequiredrelianceonacostofservice,arecentdecisionbytheFloridaSupremeCourt,InternationalMineralsv.FloridaCommission,decidedJune4,1976,P.U.R.No.22,351,page50,933,statedthatpracticalconsiderationsmilitateagainstmakingcostofserviceanexclusivecriterioninsettingrates;thatmosteverycourtconsideringthematterhasrejectedoutofhandtherulethatwouldreduceratemakingtoanexerciseincostaccounting.”).17RichardHirsh,TechnologyandTransformationintheAmericanElectricUtilityIndustry(2003)atp.56.18See,e.g.,JamesE.Suelflow,PublicUtilityAccounting:TheoryandApplication(1973)atp.59-62;Bonbright.19JosephEto,ThePast,Present,andFutureofU.S.UtilityDemand-SideManagementPrograms(1996)LBNL-39931,availableat:https://emp.lbl.gov/sites/all/files/39931_0.pdf.20RichardHirsch.EnergyCrisisofthe1970s(2002)SmithsonianInstitution,availableat:http://americanhistory.si.edu/powering/past/history3.htm.21See,e.g.,RichardL.CarlsonandDonaldJ.Sipe,GuidelinesforGrantingIndustrial“Distress”RateDiscounts,PublicUtilitiesFortnightly(January15,1995)atp.18(“Increasinglyweseeelectricutilitiesofferingdiscountedratestotargetedcustomersorclassestooffsetrisingcompetition.”);DiscountsDefined,PublicUtilitiesFortnightly(July15,1996)atp.29(“Todate,atleast41stateshaveallowedeconomicdevelopmentrates,34stateshaveallowedloadretentionrates,4stateshaveestablishedflexrates,and5haveok’dsomeformofbroad-basedPBR[performance-basedrates].”).22See,e.g.,AmericanSocietyofCivilEngineers,2013ReportCardforAmerica’sInfrastructure,http://www.infrastructurereportcard.org/a/#p/energy/overview,andU.S.EnergyInformationAdministration,Electricitydistributioninvestmentsroseoverthepasttwodecades(2014),availableat:http://www.eia.gov/todayinenergy/detail.cfm?id=18531.23EEI,ElectricUtilityIndustryFinancialDataandTrendAnalysis,availableat:http://www.eei.org/resourcesandmedia/industrydataanalysis/industryfinancialanalysis/QtrlyFinancialUpdates/Pages/default.aspx;andEEI,ElectricPowerIndustryCapitalInvestmentExpectedtoRemainatRecordLevel,availableat:http://www.eei.org/resourcesandmedia/energynews/Pages/Electric%20Power%20Industry%20Capital%20Investment%20Expected%20to%20Remain%20at%20Record%20Level.aspx24Non-capitalcostsarelargelyoperatingandmaintenanceexpendituresandadministrativeandgeneralexpenditures.Thefuel/purchasedpowercomponentoftheseisusuallyseparatelyhandledunderandautomaticadjustmentclause.


25DepartmentofEnergy,LargePowerTransformersandtheU.S.ElectricGrid(2012)atp.28,availableat:https://energy.gov/sites/prod/files/Large%20Power%20Transformer%20Study%20-%20June%202012_0.pdf;ASCE,FailuretoAct:TheEconomicImpactofCurrentInvestmentTrendsinElectricityInfrastructure(2011)atp.37,availableat:http://www.asce.org/uploadedFiles/Issues_and_Advocacy/Our_Initiatives/Infrastructure/Content_Pieces/failure-to-act-electricity-report.pdf;EIA,Investmentinelectricitytransmissioninfrastructureshowssteadyincrease(2014),availableat:http://www.eia.gov/todayinenergy/detail.php?id=17711;EIA,Electricitydistributioninvestmentsroseoverthepasttwodecades(2014),availableat:http://www.eia.gov/todayinenergy/detail.php?id=18531.26EIA,AnnualEnergyOutlook2015withprojectsto2040(2015),atp.8,availableat:https://www.nrc.gov/docs/ML1617/ML16172A121.pdf.27E.Kahn,RegulationbySimulation:TheRoleofProductionCostModelsinElectricityPlanningandPricing,OperationsResearch,Vol.43No.3(May-June1995),atpp.388-398.28Bonbright1961atp.339(“[b]utcomprehensiveapportionmentshavebeenpresentedofficiallyinonlyatinyfractionoftheratecases.”).29NationalRegulatoryResearchInstitute,AlternativeRateMechanisms(March1,2014)availableat:http://nrri.org/download/nrri-14-03-alternative-rate-mechanisms/;AARP,Inc.,IncreasingUseofSurchargesonConsumerUtilityBills(May2012)availableat:http://www.aarp.org/content/dam/aarp/aarp_foundation/2012-06/increasing-use-of-surcharges-on-consumer-utility-bills-aarp.pdf.30EEIQ42016FinancialUpdate[EEIQ42016],availableathttp://www.eei.org/resourcesandmedia/industrydataanalysis/industryfinancialanalysis/QtrlyFinancialUpdates/Pages/default.aspx;AmericanPublicPowerAssociation,U.S.ElectricUtilityIndustryStatistics(2015-2016),availableat:http://www.publicpower.org/files/PDFs/USElectricUtilityIndustryStatistics.pdf.31NewsarticlesstatethatFloridaPower&Lightwillspend$4.9milliononits2016ratecase,e.g.,PalmBeachPost,FPLcustomerspaying$4.9millionforratecase(August23,2016),availableat:http://www.mypalmbeachpost.com/news/business/fpl-customers-paying-49-million-for-rate-case-cost/nsK4B/.Initsfinalorder,thePublicUtilitiesCommissionofTexasapprovedover$3millioninratecaseexpensesforElPasoElectricanditsmunicipalities,minus$600,000.DocketNo.44941,Order(August25,2016)atp.8-9.32HethieParmesanoandJeffD.Makholm,TheElectricityJournal,TheThaw:TheEndoftheIceAgeforAmericanUtilityRateCases–AreYouReady?(2004),availableat:http://www.nera.com/content/dam/nera/publications/archive1/The%20Thaw%20End%20of%20Ice%20Age%20July%202004.pdf;EnergyTimes,CraftingyourNextRateCase(January1,2005)availableat:http://tdworld.com/business/crafting-your-next-rate-case.33EEIQ42016.34NationalAssociationofStateUtilityConsumerAdvocates,availableat:http://nasuca.org/about-us/.35Forexample,theservicelistforPG&E’s2017ratecase(A.15-09-001)includesacertificateofservicewith29namedparties.36See,e.g.,monthlyresidentialfixedchargeincreaseproposalsfromAlliantEnergyinWisconsin($7.67to$18.00in660-UR-120);DelmarvaPowerinDelaware($11.70to$17.47in16-0649);UnitilinNewHampshire($10.27to$15.00inDE16-384);andPenelecinPennsylvania($9.99to$17.10inR-2016-2537352),amongmanyothers.37See,e.g.,NationalGrid’sfilingsinRhodeIsland(Docket4568,filedJuly1,2015)andMassachusetts(Docket15-155,filedNovember6,2015).38ColoradoPublicUtilitiesCommission,ProceedingNo.16AL-0048E(PublicServiceCompanyofColorado,filedJanuary25,2016).


39See,e.g.,TucsonElectricPower(Arizona,E-01933A-15-0322);ElPasoElectric(Texas,44941);andMontana-DakotaUtilities(Montana,D2015.6.51),amongothers.40See,e.g.,UnisourceEnergyServices(Arizona,E-04204A-15-0142),amongothers.41Otherfoundationsmonopoliesandquasi-monopoliesusedtopriceserviceinthefirsthalfofthe20thcenturyincludedvalueofserviceandpromotionoftheservice.42Bonbright1961atpp.27-28.43Afourthfunction,“compensatoryincometransferfrombuyertoseller,”isusuallysubsumedwithinthecapitalattractionfunction.44Bonbright1961atp.301.45EvenCCOSSsperformedonamarginalcostbasistendtoassumethesystem,andusagepatterns,astheyare,ratherthanhowtheycouldbe.46Bonbright1961atpp.80-81.47Id.;Bonbright1988atp.114.48Thefirsteditionaddsthis,reflectingoneofthemajorargumentsofthe1950swithrespecttoratemaking:“Idonotbelievethattheargumentsonbehalfofreplacement-costpricingofutilityservicesarewelltaken.Onthecontrary,Iamconvincedthatsomeversionofanactual-costbasisofratemaking,withitsfrankacceptanceofasunk-costpricephilosophy,ispreferableforpracticalreasonsandisbynomeansinferiorforreasonsofpricetheory.Butthedeficienciesofasunk-coststandardofratemakingshouldberecognized;andtheyareespeciallyseriousinthedeterminationoftheindividualrateschedulesasdistinctfromthegeneralratelevels.”Bonbright1961,atp.76.49Bonbright1961atpp.295-296.50Bonbright1988atp.116.51SeanCastenandJoshuaMeyer,PublicUtilitiesFortnightly,Cross-Subsidies:GettingtheSignalsRight(2004)availableat:http://www.fortnightly.com/fortnightly/2004/12/cross-subsidies-getting-signals-right.52See,e.g.,DirectTestimonyofScottJ.RubinonbehalfofArizonaUtilityRatepayerAlliance(February3,2017)DocketNo.E-01345A-16-0123.53Bonbright1961atpp.37-38(emphasisinoriginal);Bonbright1988atp.79.54Bonbright1961atp.35(emphasisinoriginal).55Asanotherexample,ifnon-coincidentpeakdemandisimportantforcustomersatthecircuitlevel—andmanyutilitiesarguethatitis—shouldratesconsidergeographicdiversity,insteadofbeingassessedlikepostagestamps?See,e.g.,ColoradoPUC,ProceedingNo.16AL-0048E,DirectTestimonyofDoloresR.Basquezatpp.30-32(January25,2015)(discussingloaddiversity).Incontrast,someAlaskanutilitieshavegeographicallydiverserates,suchasAlaskaPowerCompany(see,e.g.,DocketTA857-2).56RhodeIslandPUCDocket4568,WorkpaperNG-3,atp.3;OklahomaCorporationCommission,PUD201100087,DirectTestimonyofGregTillman(July28,2011)atp.21;P.Chernicketal.,ChargeWithoutaCause?AssessingElectricUtilityDemandChargesonSmallConsumers(2016).57P.Chernicketal.,ChargeWithoutaCause?AssessingElectricUtilityDemandChargesonSmallConsumers(2016)atp.19.58HethieParmesanoandJeffD.Makholm,TheElectricityJournal,TheThaw:TheEndoftheIceAgeforAmericanUtilityRateCases–AreYouReady?(2004)availableat:http://www.nera.com/content/dam/nera/publications/archive1/The%20Thaw%20End%20of%20Ice%20Age%20July%202004.pdf


59See,e.g.,DirectTestimonyofGeorgeNovelaonbehalfofElPasoElectric(May11,2015)DocketNo.NMPRC15-00127-UT,atpp.8-10.60See,e.g.,DirectTestimonyofWilliamA.MonsenonbehalfoftheAllianceforSolarChoice(October27,2015)DocketNo.15-07041,atp.10,availableathttp://pucweb1.state.nv.us/PDF/AxImages/DOCKETS_2015_THRU_PRESENT/2015-7/6655.pdf.61NARUC,ElectricUtilityCostAllocationManual(1992)atp.67[“NARUCManual”](“nomethodisprescribedbyregulatorstobefollowedexactly”and“Thepossibilitiesforvaryingthemethodsarenumerousandshouldsuittheanalysts’assessmentofallocationobjectives.”).62NARUCManualatp.105.63Thesumoftheserateclass/schedulerevenuerequirementsmaynotbetheutility’stotalrevenuerequirement,particularlyiftheCommissionhasadoptedapolicyofbasingratespreadanddesignonmarginalcostofserviceratherthanembeddedcostofservice.64Decouplingisdefinedas“arateadjustmentmechanismthatbreaksthelinkbetweentheamountofenergyautilitysellsandtherevenueitcollectstorecoverthefixedcostsofprovidingservicetocustomers…decouplingallowsautomaticorsemi-automaticpriceadjustments,whichensuresrecoveryoftheallowedrevenueamountaspricesareadjustedsothattheallowedrevenueisrecovered…andthisremovestheincentiveforutilitiestoincreasesalesasameansofincreasingrevenueandprofits…”NationalRenewableEnergyLaboratory,DecouplingPolicies:OptionstoEncourageEnergyEfficiencyPoliciesforUtilities(December2009)atp.1,availableat:http://www.nrel.gov/docs/fy10osti/46606.pdf.65ArizonaCorporationCommission,DocketNo.E-00000J-14-0023,DecisionNo.75859(January3,2017),atpp.21-22,143-145.66Becausemany,ifnotmost,utilitiesmaintainbooksthatconformwiththeFederalEnergyRegulatoryCommission’sUniformSystemofAccounts,functionalizationcanrelyonquestionsofaccountinginterpretationratherthanmodelingorengineeringdata.67NARUCManualatpp.90-95.68See,e.g.,1993Wash.UTCLEXIS65atpp.16-17(“ThecompanyproposedtoclassifydistributioncostsusingtheBasicCustomermethod,whichtreatssubstations,poles,towers,fixtures,conduit,andtransformersasdemand-related.Servicedropsandmetersareclassifiedascustomer-related…TheCommissionfindsthattheBasicCustomermethodrepresentsareasonableapproach.Thismethodshouldbeusedtoanalyzedistributioncosts,regardlessofthepresenceorabsenceofadecouplingmechanism.WeagreewithCommissionStaffthatproponentsoftheMinimumSystemapproachhaveonceagainfailedtoanswercriticismsthathaveledustorejectthisapproachinthepast.”).Alaskaappearstogenerallyprohibitanyclassificationmethodologyotherthanbasiccustomer,accordingto3A.A.C.48.540onCostofServiceMethods(“Customercostsmayinclude(A)carryingcostsassociatedwithservicelinesfromthetransformertothemeter,meters,andinstallationoncustomerpremises;and(B)meter-readingexpense,customerinstallationexpense,metermaintenance,andcustomeraccountingandbillingexpense.(2)Customercostsmaynotinclude(A)anyportionofthedistributionsystemcosts,whichwillbeconsideredandclassifiedasdemand-relatedcosts;(B)anyportionofthetransmissionsystem;or(C)anyportionofthegenerationsystem.”);2016Minn.PUCLEXIS254,pp.71-72,80(findingforanaturalgasutilitythat“TheBasicCustomermethodreflectsthepremisethatthedistributionsystemisasharedassetdesignedandbuilttoprovidethecapacitytoservecustomersduringperiodsofpeakdemand,andthusshouldgenerallyberegardedascapacitycost.Onlycoststhatcanbetracedbacktoaspecificcustomer--suchasthecostsofservicelines,meters,billing,andcollection--areclassifiedascustomercosts…Inaddition,theCommissionwillalsodirectGreatPlainstofileinitsnextratecaseanalternativeCCOSSusingtheBasicCustomermethod.Havingbothstudieswillbetterilluminatehowthechoiceofanalyticalmethodinfluencesthecostsassignedtoeachcustomerclass.”).69KentuckyPublicServiceCommission,CaseNo.2000-080onLouisvilleGas&Electric’snaturalgasratecase(2000Ky.PUCLEXIS1356,atpp.92-93).


70MinnesotaPublicUtilitiesCommission,G-008/GR-15-424onCenterPointEnergy’snaturalgasratecase(2016Minn.PUCLEXIS163,atp.127).71OregonPublicUtilitiesCommission,UM827onmarginalcostmethodologiesforelectricutilities(1998Ore.PUCLEXIS246,atp.8).72IndianaUtilitiesRegulatoryCommission,CauseNo.43839onSouthernIndianaGas&Electric’selectricratecase(2011Ind.PUCLEXIS115,atpp.200-201).

new uses for an old tool - final...new uses for an old tool 1 about eq research eq research, llc...

Documents