INTEGRATED RESOURCE PLANNING FOR ELECTRIC AND GAS UTILITIES

Eric Hirst, Oak Ridge Na.tiona.1 LaboratoryCharles Goldman, Lawrence Berkeley Laboratory

Mary Ellen Hopkins,The Fleming Group

During the past few years, the scope and nature of resource planning at electricutilities has changed dramatically.. The scope of planning has expanded to considerenergy-efficiency and load-management programs as resources, the environmentalcosts of electricity production, and a variety of resource-selection criteria beyondelectricity price.. The nature of planning has expanded to include regulatorycommissions, nonutility energy experts, and customers, as well as utilities themselves..Similar changes are beginning to occur at gas utilities II

This paper discusses a few of the key issues related to resource planning: provisionof financial incentives to utilities for successful implementation of integrated resourceplans, incorporation of environmental factors in resource planning, bidding fordemand and supply resources, development of guidelines for preparation and reviewof utility integrated resource plans, resource planning for gas utilities, and greaterefforts by the DoS.. Department of Energy to encourage integrated resource planning.

INTRODUCTION

Integrated resource planning (IRP) helps utilitiesand state regulatory commissions consistentlyassess a variety of demand and supply resources tocost-effectively meet customer energy-service needs.Key characteristics of this planning paradigminclude: 1) explicit consideration of energy­efficiency and load-management programs asalternatives to some power plants and new suppliesof natural gas, 2) consideration of environmentalfactors as well as direct economic costs, 3) publicparticipation, and. 4) analysis of the uncertaintiesand risks posed by different resource portfolios andby external factors.

IRP differs from traditional utility planning inseveral ways, including the types of resourcesacquired, the owners of the resources, the organ­izations involved in planning, and the criteria forresource selection (Table 1). Cavanagh (1986),Electric Power Research Institute (1988), Gellings,Chamberlin and Clinton (1987), Hirst (1988a and

1988b), and the National Association of RegulatoryUtility Commissioners (1988aand 1988b) discussIRP and its development

This paper reviews recent progress in IRP andidentifies the need for additional work" Key IRPissues facing utilities and public utility commissions(PUCs), discussed in this paper, include:

.. Provision of financial incentives to utilities forsuccessful implementation of integrated resourceplans, especially acquisition of demand-sidemanagement (DSM) resources;

@ Incorporation of environmental factors in IRP;

GD Bidding for demand and supply resources;

@ Development of guidelines for preparation andreview of utility resource plans;

@ Resource planning and DSM programs for gasutilities; and

Integrated Resource Planning 5.95

Table 1~ Differences Between Traditional Utility Planning and Integrated Resource Planning

Traditional Planning

Focus on utility-owned central...station power plants

Planning internal to utility,primarily in system planning andfinancial planning departments

All resources owned by utility

Resources selected primarily tominimize electricity prices andmaintain system reliability

Integrated Resource Planning

Diversity of resources, including utility-owned plants,purchases from other organizations, conservation andload-management programs, transmission and distri­bution improvements, and pricing

Planning spread among several departments withinutility and often involves customers, public utilitycommission staff, and nonutility energy experts

Some resources owned by other utilities, by smallpower producers, by independent power producers,and by customers

Diverse resource-selection criteria, includingelectricity prices, revenue requirements, energy­service costs, utility financial condition, riskreduction, fuel and technology diversity, environ...mental quality, and economic development

@ Increased efforts the U.S. Department ofEnergy (DOE) to

... include DSM resources in PERC approval ofwholesale contracts,

o expand the DSM programs run by federalPower Marketing Agencies,

- expand DOE's Integrated Resource PlanningProgram, and

... collect data on performance of utility DSMprograms.

Many other issues related to resource planning areimportant, but are not discussed in this aper.. Suchissues include alternative ways to organize planningwithin utilities; the role of collaboration and otherforms of nonutility involvement in planning (Ellis1989; Schweitzer, Yourstone, and Hirst 1990); therelationships among competition, deregulation, and

planning; treatment of electricity pricingas a resource; fuel switching (primarily betweenelectricity and gas); treatment of uncertainty in

planning and decision making (Hobbs andMaheshwari 1990; Hirst and Schweitzer 1990); the

5.96 Hirst, Goldman, and Hopkins

appropriate economic tests for utility DSMprograms; ways to measure the performance ofDSM programs (Argonne National Laboratory1989); development and use of improved data andplanning mod.els; and transfer of information amongutilities and commissions (Goldman, Hirst, andKrause 1989).

REWARDING UTILITIES FOREFFECTIVE IRP IMPLEMENTATION

A key element of IRP is the treatment of utilityenergy-efficiency programs as a resource that cansubstitute for some power plants. Unfortunately,traditional regulation of utilities pits the interest ofutility shareholders against those of utilitycustomers. This conflict occurs because "each kWha utility sells eO<! adds to earnings [and] each kWhsaved or replaced with an energy efficiency measure$0$ reduces utility profits" (Moskovitz 1989). Thegrowing realization that effective implementationof least-cost plans may hurt utility shareholdersled the National Association of Regulatory Utility

Commissioners (1989) to adopt a resolution urgingPDCs to consider the loss of earnings associatedwith DSM programs and to adopt ratemakingmechanisms that encourage utility DSM programs..

Several proposals to reform PUC regulation havebeen made that incorporate one or more of thefollowing three factors:

• Recovery of the utility costs to operate DSMprograms,

• Utility recovery of the net lost revenue (differ­ence between revenue foregone because ofreduced electricity use and reduced operatingcosts) caused by DSM programs, and

e Provision of financial incentives to utilityshareholders for exemplary delivery of DSMservices..

The first two elements remove disincentives toutility operation of DSM programs, while the thirdelement adds positive incentives to run suchprograms. The underlying idea is that utilitiesshould operate under regulatory practices that makeit financially attractive for them to implement allaspects of their integrated resource plan, not justacquisition of supply resources (Moskovitz 1989;WieI1989)&

Simple incentive methods have been in place forseveral years in a few states, including Washingtonand Wisconsin41 In 1989, PUCs in several states,including Maine, Vermont, Massachusetts, NewYork, New Jersey, Minnesota, Nevada, California,and Washington began inquiries into the desirabilityand form of different incentive procedures& Progresshas been most rapid in california, Massachusetts,New York, and Rhode Island& The New York PublicService Commission (1989) approved proposalsfrom several utilities to test incentive schemes" Andthe Rhode Island and Massachusetts PDCs adoptedincentives. for New England ectric..

According to Moskovitz (1989), a key element of asuccessful regulatory system is the decoupling ofprofits from sales. That is, utility earnings shouldnot depend on the amount of sales achieved. TheElectric Revenue Adjustment Mechanism, used inCalifornia since 1982, accounts for the over- or

under-collection of authorized base revenues(essentially all nonfuel costs) caused by discrep­ancies between actual and forecast sales ofelectricity (Marnay and Comnes 1990).. Utilities usebalancing accounts for over... (or under-) collectionof revenues. These revenues are then returned to(or collected from) customers the following yearthrough an adjustment to the price of electricitysThis mechanism breaks the link between sales andprofits, thus eliminating a major disincentive toutility DSM programs&

Moskovitz also discusses methods that go beyondremoval of disincentives to reward utilities foreffective programs.. He groups incentive proposalsinto three groups: rate-of-return adjustments, sharedsavings, and bounty. The shared savings approach isthe most widely discussed because it seeks to rewardutilities for acquiring resources that deliver desiredenergy services at least cost (The two otherapproaches are less appealing because the utilityincentive does not depend directly on the benefitprovided by the utility DSM programs.)

New England Electric (SergeI 1989; Destribats,Lowell, and White 1990) proposed such a shared...savings incentive scheme (Fig.. 1) Rhode Island,New Hampshire, and Massachusetts<9 The proposalhas two parts. The first incentive, intended tomaximize the size of the company's DSM programs,is equal to 5% of the total benefit of the programs.The benefit is equal to the electricity savings causedby the programs multiplied by the company'savoided costs (i.e,., kWh-saved x (/,fkWh). The secondcomponent, intended to reward program efficiency,is equal to 10% of the net benefit of the programs,where net benefit is the difference between (a) theproduct of reduced electricity use and utility-systemavoided costs and (b) the costs to operate the DSMprograms. The Rhode Island PUC approved a modi­fied version of this proposal. The MassachusettsDepartment of Public Utilities (1990b) approved anincentive system with "a fixed payment for each kWand kWh saved that is verified through an after­the-fact evaluation and monitoring system,."

As part of the 1989 collaborative in California,Schultz (1990) examined alternative shared-savingsproposals for utility DSM programs<9 His analysis

Integrated Resource Planning 5.97

NE T BENEFIT

NEES CUSTOMER

\ NEES INCENTIVE- 5% OF TOTAL BENEFIT.... 10% OF NET BENEFIT

TOTAL BENEFIT OF DSM PROGRAMS:AVOIDED GENERATION AND T&D COSTS

Figure 113 The DSM Incentive Proposed by New England Electric is Proportional to the Total and Net Benefits ofthe Company's DSM Programs

focused on the purposes of the DSM programs andon the risk/reward relationships implicit in differentincentive proposalso These proposals differ in theirsensitivity to changes in total resource costs, utilitycosts, avoided costs, and electricity savingsstimulated by the DSM programs~ Schultz suggestedthat incentive mechanisms should seek to maximizethe net benefit of DSM programs, require minimumperformance, or minimize program costs"

Reforming utility ratemaking is now an importantpart of integrated resource planninge Discussionsamong utilities, commissions, and others are under­way in many states. And, in a few states, utilities aretesting su~p schemes on a trial basiss Additionalanalyses are needed to better understand the prosand cons of different regulatory reforms within thecontext of the accounting rules used by individualstates and utilities0 And careful evaluations of theeffects of these schemes are needed.. These evalua­tions should identify the effects of regulatory reformon the size and effectiveness of utility DSMprograms and. on the costs to utility customerso Inaddition, the energy savings and load reductionscaused by utility DSM programs must be carefullyand accurately measured (Argonne NationalLaboratory 1989), because these measurementsd.etermine the incentive payments to utilities..

5.98 Hirst, Goldman, and Hopkins

INCORPORATING ENVIRONMENTALFACTORS IN UTILITY PLANNING

The environmental impacts that accompany oper­ation of power plants have significant effects onsociety.. For example, electricity production accountsfor two-thirds of the 502' one-third of the NOx andone-third of the CO2 emitted in the UeSe Theseairborne pollutants are linked to secondary effects(eog., acid rain and global warming) that affectsociety (e"g.., reduced forest production and damageto coastal land.). These effects of electricityprQd\lction are externalities~ defined as any cost notreflected in the price p~Jd by customers forelectricity (Chernick and Caverhill 1989)..

External effects cease to be externalities once theircosts are paid by .the entity responsible for theirproduction and are reflected in the price charged forthe product For example, existing federal and stateregulations (e..g., pollution control rules that requiremitigation of negative environmental impacts onland and water use) internalize some of the environ­mental costs associated with electricity production"

PUC Approaches

Partly in response to increased public concern aboutacid rain and global warming, several PDCs are

insisting that environmental impacts be explicitlyaccounted for in utility resource planning andacquisitions. A recent survey found that 15 PUCShave procedures for considering environmentalexternalities in their resource planning andacquisition (Cohen et at 1990).

This internalization is typically done in one of threeways. These approaches seek to influence the choiceand relative magnitude of selected resources. Theydo not change the direct economic costs of thevarious resource alternatives to ratepayers, but mayraise the ultimate cost of electricity if environment­ally-benign resources are more expensive than otheralternatives.

The first approach relies on qualitative treatment ofenvironmental externalities by the utility in itsintegrated resource planning process. For example,the Nevada Public Service Commission has broaddiscretion to "give preference to the measures lUe

that provide the greatest economic and environ­mental benefits to the state." Under this approach,PDCs require utilities to consider environmentalcosts in resource planning but do not specify themethods to be used..

Table 2 illustrates an example of this approach, usedby Ontario Hydro (1989). Its analysis included1) development of natural and social environmentalcriteria (e.g.., land and water use, atmosphericemissions, solid waste production, and socio­economic effects such as regional employment andlocal community impacts) and 2) a comparativeanalysis of alternative demand/supply plans and theenvironmental advantages and disadvantages ofeachplan. For natural environmental criteria, OntarioHydro developed quantitative indices (e.g., totalwaste, thermal discharge, air emissions of variouspollutants, and land and water use) that wereestimated for alternative resource plans..

The second approach involves use of a percentageadder that either increases the cost of supplyresources or decreases the cost of DSM resources..For example, the Wisconsin Public ServiceCommission (1989) recently implemented a "non­combustion" credit of 15% for non-fossil supply anddemand-side resources becauseof reduced pollution..

The third approach involves direct quantification ofthe cost of the externality, which often occurs ifthe utility is developing a competitive resource

2$ Environmental Analysis Process Used by Ontario Hydro

Ie Develop criteria for environmental effectsResource use... Non-renewable resources: coal, oil, gas, and uranium.,. Water use... Land use

Emissions, effluents, and wastesom Atmospheric emissions: 802' NOx, CO2, radionuclides, and trace elements... Aquatic effluents: thermal discharges, radionuclides, uranium mining effluents, and coal

mining effluents... Wastes: coal ash, flue-gas desulphurization wastes, used nuclear fuel, low-level radioactive

wastes, and uranium mine tailings

2.. Evaluate the environmental implications of alternative plans

3" Consider mitigation/compensation to offset the potential environmental effects

4.. Determine the environmental advantages and disadvantages of the alternative plans

Integrated Resource Planning 5.99

procurement process (i.e., bidding). For example,the New York Public SelVice Commission (1989)recently approved utility bidding proposals thatassigned values to different levels of air and wateremissions and land degradation of each bid, whichcan add up to 1.4 rtlkWh. Several of the New Yorkutilities use this method to adjust the cost of eachbid, while other utilities use a "point system," whichweights environmental factors relative to the pricefactor in scoring competing supply and demandprojects.

Alternative Methods to QuantifyEnvironmental Costs

A key challenge in treating environmental exter­nalities in utility planning is deciding on the propercosting method to use in calculating external costs..Thus far, studies have adopted one of two basicapproaches.. The first approach involves calculationof damage costs imposed on society by a generatingtechnology" Costs to society are estimated by tracingimpacts through each step of the fuel cycle (i.e..,emissions, transport of pollutants, and the effects ofthese pollutants on plants, animals, people, etc..).The extent of each effect that arises from anexternality is estimated and a value is assigned tothat effect For example, 502 emissions. can .belinked to lost forest products, damage to buildings,and human respiratory problems. In direct costestimation, the challenge is to identify and quantifythe dollar cost of each effect, which yields anexplicit estimate of the social cost of the externality..However, estimating damage costs is quite difficultFor example, the technical and methodologicalissues are complicated because valuation is notpossible for some environmental-resource damages,aos;e-reSDon~;e relationships are uncertain, valuingintangible costs (e&g0' recreation facilities andendangered species of wildlife) is difficult, valuinghuman mortality is controversial, and some damagesare very site specific (Chernick and Caverhill 1989;.....,"".81...8I.A~V.l 1989)0

The second approach relies on the cost of controlmitigation) of the pollutants emitted by the

technology to estimate the value ofpollution reduction (Chernick and Caverhill 1989)"The rationale for this approach is that the cost ofcontrols provides an estimate of the price that

5~100 Hirst, Goldman! and Hopkins

society is willing to pay to reduce the pollutant Thisapproach has limitations (e.go, cannot directly beapplied. to pollutants such as CO2 which are notnow regulated& But the technique is attractivebecause the most thorny policy issues that arise indirect cost estimation (e..g., assigning dollar valuesto human life, valuing ecosystems) have implicitlybeen dealt with by the legislators and regulatorybodies that formulated the pollution control stand­ards. The disadvantage of using control costs tocalculate environmental externality costs is that theytypically bear little relation to the actual damagesimposed on society by power plant emissions"

New England Electric System developed a hybridapproach that it calls an issue-based rating andweighting index (Destribats, Lowell, and White1990)& NEBS assigns various environmental factors(e.ge, global warming, acid rain, land use, indoor airquality) a weight based on a survey of experts.. Thecompany then uses an impact index ranging fromzero to four for every contributing factor (e.g., SOxand NOx contribute to acid rain) to score resourceoptions.. NEES then assigned the highest ratedproject (i.e.., the most environmental degradation) acost adder of 15 percent; the costs of other projectswere increased based on a sliding scale (eog.., ratio oftheir score to the highest score) 0 This method iseasy to understand, but its lack of scientific basis istroubling.. Therefore, this approach may be usefulprimarily as an interim method for includingenvironmental factors in resource planning.

Clearly, valuation of external environmental costs inthe context of utility planning is an emerging field,one in which estimates and methods will evolvebased on projects underway in various states, includ­ing New York, Massachusetts, and California. Thereis significant disagreement among experts on keymethodological issueso These issues include whethercosts should be based on the payments that peopleare willing to make to avoid environmental damagesor the money that must be paid to them to acceptthese damages, appropriate discount rates,uncertainty about the effects and costs of differentpollutants, and effects that are hard to price(Ottinger et at 1989).

The complexities and consequent disagreementsabout the magnitudes of environmental costs are

shown in Fig. 2. The lower bars show the directcosts for a coal-fired baseload plant, gas-firedcombined-cycle plant, and gas- and oil-firedcombustion turbines. e low estimates ofenvironmental costs are from the New York PublicService Commission and the high estimates are fromChernick (Koarney 1990). Interestingly, the lowcosts include the environmental impacts on wateruse and land use as well as air pollution, while thehigh estimates include air pollution only.

In the future, policies that consider environmentalcosts in resource acquisition will increasingly rely onpollutant-based methods ofassessing environmentalimpacts, rather than most initial efforts which weretechnology-based. One effect of this shift will bethat bidding processes will further differentiateamong supply-side and DSM technologies based onindividual project characteristics (e.g.., expectedemissions, project size, location). New approaches,such as that adopted by the New York utilities intheir bidding systems, may well be adopted by otherstates.

In addition, increased attention to environmentalconcerns may provide an important impetus forpublic policy makers and PUCs to broaden theboundaries of IRP. For example, PUCS may ask gasand electric utilities to compare the social costs andbenefits of providing energy services (e.g., waterheating or cooking) using gas directly or throughgas-fired electric generation. Future public policyconcerns about the environmental effects of energytechnologies may force significant changes in thedemands for electricity and gas. For example, thepolicies of local air quality boards that limit vehicleemissions (and, for example, encourage electric cars)or national legislation affecting greenhouse gasemissions could affect future electricity and gas uses.

Finally, electric utilities and others are. likely toraise basic questions about the role of PUCS andutilities to address environmental externalitiesversus the roles of federal and state governmentagencies that deal with environmental quality (e.g.,the U41Se Environmental Protection Agency)..

lEVELIZED COST (cents/kWh)20 r------- '----------------------------,

15

10

5

ENVIRO COST

ENVIRO COST (lOW)

DIRECT COST

COALCOMBINED CYCLE ---COMBUSTION TURBINE---

GAS GAS Oil

2~ Alternative Estimates of the Direct and Environmental Costs ofElectricity Production from new powerpIa (Koomey 1 ~. The assumed capacity factor is 65% for the coal and combined-cycle plants and10% for the combustion turbines.

Integrated Resource Planning 5.101

NEW APPROACHES TO ACQUIRINGELECTRIC POWER RESOURCES

Nonutility power production has emerged as amajor source of new generating capacity, principallybecause of the 1978 Public Utility RegulatoryPolicies Act (PURPA). Cogenerators and smallpower producers built nearly 15,000 MW of non­utility capacity during the 19808. Under PURPA,PUCs are responsible for implementing pricingarrangements under which electricity is purchasedfrom Qualifying Facilities (QFs) at the utility'savoided cost. Avoided costs were determinedadministratively and some states, which sought toencourage QF suppliers, offered long-term contractsbased on forecasts ofavoided costs.. In several states,the response by private producers was much greaterthan expected, partly because avoided cost forecaststurned out to be high given events in world oil andgas markets. Some utilities also claimed that theobligation-to-purchase provisions of PURPA andthe open-ended nature of standard-offer contractsintroduced substantial uncertainty about how muchpower would ultimately be developed. Thus,PURPA was not an unqualified success, because thesupplier response created major planning andoperational problems for some utilities.

During the past few years, some utilities and PUCshave experimented with competitive resourceprocurements (CRPs) as one way to obtain supplyand DSM resources, partly in response to theproblems associated with PURPA Since 1986, about25 solicitations have been issued by 14 utilities..Thus far, capacity offered by private producers hastypically been 10-20 times greater than the utility'srequirements" However, some utilities have foundthat a significant fraction of bids do not meet therequirements specified in the bid package and aretherefore dropped from serious consideration.. Forexample, Central Maine Power received bids forover 2300 MW of generating capacity in response toa 1989 solicitation; only about 1000 MW remainedas realistic options after CMP's initial review of thebids$

CRPs are attractive to private producers because thepurchasing utility offers long-term contracts, whichare needed to get financing on attractive terms. Forthe utility, competitive acquisition allows it to ration

5.102 Hirst, Goldman, and Hopkins

contracts for non-utility resources in an efficientmanner. Moreover, these contracts commonly trans­fer to private developers some· of the risks associ­ated with project siting and permitting, constructioncost overruns, and environmental impacts.. Inaddition, a competitive process can reduce theburden of estimating avoided cost by providing amarket benchmark to determine value.

Despite its theoretical virtues, there are formidablepractical problems involved with developingcompetitive procurement programs.. Traditionalutility planning requires trade-ofts among financial,operating, and environmental features of resourcealternatives. Competitive bidding requires the utilityto address these issues through arms-lengthcontracting.. To assess bids from developers, a utilitymust account for and value the multiple attributesof projects.. This unbundling and explicit valuationof attributes is a new phenomenon in resourceplanning. Typically, utility bidding systemsdifferentiate projects on pricing terms, operatingcharacteristics, project status and viability(e.g.,likelihood of successful development), and insome cases environmental impacts. Determining theeconomic value of these non-price factors isprobably the most difficult problem that utilitiesconfront in designing bidding systems (Kahn et at1989)9

Two design features are particularly critical forutilities as they develop CRPs: 1) the method usedto assess or score proposals, specifically the extentto which the utility discloses assessment criteria andthe weight assigned to each feature before bidpreparation, and 2) incorporation of DSM optionsinto bidding schemes.

Bid Evalu.ation Criteria

There are two general approaches that utilities havetaken to the bid solicitation and evaluation process..In the first approach, the utility develops an explicitscoring system that clearly states the assessmentcriteria and weights for various features. Biddersself-score their projects, assigning points in variouscategories (e..g., price, level of development,dispatchability) based on project characteristics. Thisself-scoring approaCh can be considered an "open"system, in the sense that the utility's bid

evaluation process is transparent PDCs and mostutilities in Massachusetts, New Jersey, and NewYork rely on self-scoring systems.

A principal advantage ofself-scoring systems is theirtransparency. The utility's project rankings can beaudited easily' by regulators and there should belittle controversy over the utility's selection of thewinning bids. In addition, some PDCs favor self­scoring because it allows the regulators to shapeutility planning decisions in the initial stages of theresource-acquisition process rather than the morelimited role of after-the-fact prudence review ofcontracts. However, some utilities are concernedthat self-scoring denies them the flexibility neededto select the optimal mix of projects.. Anotherpotential disadvantage of self-scoring systems is thatthey assume that projects can be evaluatedindependent of their interactions with otherprojectse When the utility's resource need is smallcompared to the existing system, the independenceassumption is reasonable; however it becomesincreasingly untenable for resource procurementsthat are large relative to the existing utility system.

In the second approach, utilities reveal bid evalu­ation and project selection criteria in qualitativeterms only, providing only general guidance aboutits preferences (Kahn et· at 1989)0 Bidders submitdetailed proposals, which provide the basis for theutility's evaluation and ranking of projects~ In thisapproach, the utility retains more discretion toselect the optimal mix of projects as well as flexi­bility to negotiate with bidders in light of all offersreceived~ This approaCh can be considered "closed"because the utility has information about the evalu­ation process that is not available to bidderssProminent examples of this approach include

issued Virginia Power, FloridaPower and Light, and Public Service of Indiana"

The closed approaCh .acknowledges the inherentcomplexity in optimizing resource selection giventhat the value of proposed projects is multi­dimensional and uncertain, particularly over longtimes$ Theoretically, this approach allows the utilityto select the most efficient mix of bids, because itexplicitly recognizes the interactive effects amongindividual projects and their effects on the utilitysystem. Implicitly, PUCS that endorse this approach

trust the utility's judgment.. Utilities that wantflexibility and discretion in bid evaluation andselection often agree that their subsidiaries will notparticipate in the bidding process. This tradeoff canease concerns about abuses of market power by theutility and unfair competitive advantages (e.g.,potential impropriety and difficulty in maintainingarms-length transactions)"

Some utilities have experimented with hybridapproaches that combine elements of self-scoringsystems and closed bid systems. For example,Central Maine Power includes elements of self­scoring, although the utility retains substantialflexibility to select attractive projects for furthernegotiation. Niagara Mohawk uses a self-scoringsystem for initial screening and then negotiates withbidders in the initial award group.. TheMassachusetts Department of Public Utilities(1990a) recently proposed a similar approach. Thishybrid approach represents an attractive option thatcould successfully balance the utility's need forflexibility and discretion with the need to assurefairnesss Bid evaluation methods are an evolving artrather than a science and we expect continuedexperimentation with information requirements andrisk-sharing between utilities and private powerproducers0

Bidding for DSM Resources

Another key issue that arises in CRPs is the type ofresources and entities to include (e.ge, QFs,independently-owned generation facilities, energyservice companies [ESCOs], large commercial andindustrial customers, as well as the sponsoringutility). Among these entities and resource options,the appropriateness of including "saved kWh andkW" provided by ESCOs or individual customers hasbeen the subject ofvigorous debate (Cavanagh 1988;Cicchetti and Hogan 1989; and Joskow 1988 and1990). Much of this debate has focused ontheoretical problems with integrating DSM andsupply resources in the same "all-source" biddingprocess and the principles to use in determining theappropriate ceiling price to pay for DSM resources(Goldman and Hirst 1989)..

These debates raise interesting issues.. These concernways to measure the expected energy and demand

Integrated Resource Planning 5.103

savings, whether all sectors and demand-side optionsshould be included in a bidding program or whetherthe utility should target certain customer classes andend uses for a bidding program, and how tointegrate the DSM bidding program with a utility'sown DSM programs. A key question is whether theceiling price for DSM bids should be based onavoided cost or on the difference between avoidedcost and average revenues (to reflect lost revenues).Some of these issues are not unique to DSMbidding and arise in utility DSM programs also..

Table 3 summarizes results from utilities thatinclude DSM options in their bidding approach,including the MW offered by bidders as well as theMW selected by the utility. In addition, results areshown for recent supply-side procurements con­ducted by New England Electric and Boston Edison,along with results from their DSM performancecontracting programs involving ESCOs. Typically,there have been 5 to 15 DSM bids submitted byESCOs and individual customers.. The DSM biddershave a stronger likelihood of winning (35 to 50%)than do supply-side projects. The amount of DSMsavings proposed by winning bidders, whilesignificant (10 to 47 MW over a 3 to 5 year period),represents a small part of a utility's overall

DSM program for the same time (5...20%). Initialresults reflect current infrastructure limitations inthe ESCO industry as well as a cautious approachbeing adopted by ESCOs given the risks associated.with guaranteeing the savings and their limitedexperience with DSM bidding.

In summary, experience with incorporation of DSMoptiOns into bidding processes is limited. There area few programs nationwide, although biddingprograms are proliferating rapidly.. Initial experiencesuggests that DSM bidding may have a limited rolein a utility's overall DSM strategy but may not beappropriate for all market segments.. For example,it is difficult to imagine DSM bids for the newconstruction market. The relative immaturity of theESCO industry contrasts markedly with the strengthof private power producersll In practice, this meansthat the quantities offered under DSM biddingprograms will be small, and will not reflect the fullmarket potential of DSM. For utilities, DSMbidding programs may represent a potential businessopportunity if they establish unregulated ESCOs..Several utilities have adopted this strategy, but mostare skeptical about DSM bidding and prefer otherways to deliver DSM programs..

Table 311 Supply and DSM Resources (MW) in Utility Bidding Programs

Amount ofResource Supply Projects DSM Projects

Requested Proposed Winning Proposed Winning

central Maine Power 100 666 36 17

Central Maine Power 150-300 2338 30

& Rockland 100-150 1395 141 29 18

Public Service 200 654 210 47 47Electric & Gas

Jersey Central 270 712 235 56 26

Power 100 1251 127 28 10

auctions

New England Electric 200 4279 204 NA 14

Boston Edison 200 2800 200 NA 35

5.104 Hirst, Goldman, and Hopkins

GUIDELINES FOR PREPARATIONOF UTILITY PLANS

Many electric utilities periodically prepare long­term resource plans, often in response to require­ments from their PUC. These plans inform regula­tors and customers about the utility's analysis ofalternative ways to meet future energy-service needsand the utility's preferred mix of resources to meetthose needs. The plan is an opportunity for theutility to share its vision of the future with thepublic and to explain its plan to implement thisvision"

In principle, utility plans should be assessed on thebasis of the utility's resource-acquisition activities.But IRP is so new that insufficient implementationhas as yet resulted from these plans. Currently,

Table 4'1) Checklist for a Good Integrated Resource Plan

utility plans can be assessed only on the basis oftheir planning reports.. This section discussesguidelines for long-term resource plans, based onthe written reports (Hirst et at 1990; Schweitzer,Yourstone, and Hirst 1990). The purpose of theseguidelines is to help PUC staff who review utilityplans and utility staff who prepare such plans. Theplans prepared by Carolina Power & Light (1989),Green Mountain Power (1989), New EnglandElectric (1989), Northwest Power PlanningCouncil(1989), Pacific Power & Light (1989), PugetSound Power & Light (1989), and Seattle City Light(1989) contain many of the positive features in theguidelinese

The "goodness" of a plan can be judged by at leastfour criteria (Table 4):

Clarity of plan ... adequately inform various "groups about future electricity resource needs, resourcealternatives, and the utility's preferred strategy

• Clear writing style@ Comprehensible to different groups@ Presentation of critical issues facing utility, its preferred plan, the basis for its selection, and key

decisions to be made

Technical competence of w positively affect decisions on, and regulatory approval of,resource acquisitions

4D Comprehensive and multiple load forecasts@l Thorough consideration of demand...side optiOns and programs@l Thorough consideration of supply options• Consistent integration of demand 'and supply VIVII.AV.i..I.';'

• Thoughtful uncertainty analyses@ Full explanation of preferred and its close competitors@ Use of time horizons

Adequacy ofsl1ort...term action plan ... provide enough information to document utility's commitmentto acquire resources in long-term plan, and to collect and analyze additional data to improveplanning process

Fairness of information so that different interests can assess the plan from their ownperspectives

@ Adequate participation in plan development and review by various stakeholders@ Sufficient detail in report on effects of different plans

Integrated Resource Planning 5.. 105

.. The clarity with which the contents of the plan,the procedures used to produce it, and theexpected outcomes are presented;

• The technical competence (including thecomputer models and supporting data andanalysis) with which the plan was produced;

.. The adequacy and detail of the short-term actionplan; and

.. The extent to which the interests of variousstakeholders are addressedo

Report Clarity

The primary purpose of a utility's IRP report is tohelp utility executives decide (and PUC commission...ers review) which resources to acquire, in whatamounts, and when. Thus, the report must be usefulboth within and outside the utility. The utility's planshould be well-written and appropriately illustratedwith tables and figures. The report should discussthe goals of the utility's planning process, explainthe process used to produce the plan, present loadforecasts (both peak and annual energy), compareexisting resources with future loads to identify theneed for additional resources, document the demandand supply resources considered, describe alternativeresource portfolios, show the preferred long-termresource plan, and present the short-term actions tobe taken in line with the long-term planOl Importantdecision points should be identified, and the use ofmonitoring procedures to provide input for thosedecisions should be explainedo The most significanteffects of choosing among the available options(eog&, capital and operating costs, resourceavailability, and environmental effects) should bedis<;.ussed~ The report should also describe the dataand analytical methods used to develop the planGFinally, the plan should point the reader to moredetailed documentation on these topics..

Technical COllBPletelilce

Typically, computer models are used for a variety offunctions in developing a plan, such as load­forecasting; screening, selection, and analysis ofdemand and supply resources; and calculations ofproduction costs, revenue requirements, electricity

and financial parameters. These models are

5.106 Hirst, Goldman; and Hopkins

used to analyze a wide range of plausible futuresand available resources in developing the utility'spreferred resource portfolio. The basic structure ofthe models used, the assumptions upon which theyare based, and the inputs utilized should beexplained.

The technical competence of a utility's IRP isreflected most critically in the ways that the demandand supply resources are presented as an integratedpackage. The analytical process used to integratethese different resources should be discussed. Thecriteria used to assess different combinations ofresources (e.gG' revenue requirements, annual capitalcosts, average prices, reserve margin, and emissionsof pollutants) should be clearly stated.

Results for different combinations of supply anddemand resources should be shown explicitly. It isnot sufficient to treat demand as a subtraction fromthe load forecast and then do subsequent analysiswith supply options only" Subtracting DSM-programeffects from the forecast and using the resultant"net" forecast for resource planning eliminates DSMprograms from all integrating analysis.. Thisapproach makes it difficult to assess alternativecombinations of DSM programs and supplyresources and the uncertainties, risks, and risk­reduction benefits of DSM programs (e.gOl' smallunit size and short lead time).

Demand-side resources must be treated in a fashionthat is both substantively and analytically consistentwith the treatment of supply resources; demand andsupply resources must compete head to head. Theplan must show how the process truly integrates keyparts of the company: load forecasting, DSMresources, supply resources, finances, rates. And theimportant feedbacks among these components(especially between rates and future loads) shouldbe shown.

A thorough analysis of a variety of plausible futureconditions and the options available to deal withthem is essentiaL This analysis should consideruncertainties about the external environment (e.g.,economic growth and fossil-fuel prices) and aboutthe costs and performance of different resources.The analysis should show how utility resource­acquisition decisions are affected by these different

assumptions and show the effects of these uncertain­ties and decisions on customer and utility costs.. Theassumptions must be varied in ways that are intern­ally consistent and plausible.. Differences amongresources in unit size, construction time, capitalcost, and operating performance should be con­sidered in terms of how they affect the uncertaintiesfaced by utilities.. Finally, the links between theresults of these multiple runs and the utility'sresource-acquisition decisions must bedemonstrated..

Action Plan

The action plan, in many ways the "bottom line" ofthe utility's plan, must be consistent with the long...term resource plan.. This is necessary to assure thatwhat is presented as appropriate for the long haulis implemented, and implemented in an efficientmannerll The action plan also should be specific anddetailed.. The reader should be able to judge theutility's commitment to different actions from thisshort-term plan" Specific tasks should be identifiedfor the next two to three years, along withorganizational assignments, milestones and budgets..For example, the action plan should show thenumber of expected participants and the expectedre4uctions in annual energy use, summer peak, andwinter peak for each DSM program.. The action planalso should discuss the data and analysis activities,such as model development, data collection, andupdated resource assessments, needed to prepare forthe next integrated resource plan~

A final criterion by which a plan can be judged isthe effect of its recommended actions on variousinterested Because the interests of anstakeholders are not identical, the ways in whichthey will be affected by short- and long-term costs,power availability, and other results of utility actionswill likewise differ0

Without the involvement of customers and variousinterest groups, which requires two-way communi­cation, a plan may ignore community needs.Accordingly, the plan should show that the utilitysought ideas and advice from its customers andothers in developing the plan.. Energy experts from

a state university, state energy office, PUC,environmental groups, and organizations represent­ing industrial customers could be consulted as theplan is being developed" For example, utilities inNew England are working closely with theConservation Law Foundation to design, implement,and evaluate DSM programs (Ellis 1989)~

Additional work is needed to refine the guidelinesdiscussed here and to ensure that they are helpful toutilities and PUCS.. In particular, PUCS shouldarticulate better the reasons they want utilities toprepare such plans and how they will use the plansin their deliberations. This articulation should avoidthe "data list or cookbook approach" and focus onthe purposes of the planning report In the long...run, the success of IRP should not be measured byassessing utility reports II Rather, the level andstability of energy-service costs, the degree ofenvironmental protection, and the extent to whichconsensus is achieved on utility resource acquisitionswill be the important criteria.

RESOURCE PLANNING AT GASUTILITIES

IRP is just beginning to be applied to the naturalgas industryll At gas utilities, called local distributioncompanies (LDCs),resource planning addresses onlyoptions for purchasing and storing gas.. Traditionalplanning for LDC resource acquisition seeks topurchase the lowest cost and most reliable mix ofsupplies. This is done by determining the design daysend out, the provisions ofvarious supply contracts,and their available storage options.. (Design daysend out is the maximum amount ofgas required forthe coldest day in the coldest month.)

LDC experience with DSM programs is limited tofederally mandated programs, such as theResidential Conservation Service (created in 1978and repealed in 1989) or low-income weatherizationefforts mounted to create a positive image in thecommunity or to reduce bill arrearages..Interruptible contracts with large industrial gascustomers are also used by many LDCs to shavepeaks (e.g.. , during very cold weather).

The gas industry is concerned about declining salesand prOfits, largely because of experience during the

Integrated Resource Planning 5.. 107

19708 and 19808, when gas customers adoptedenergy-conservation actions because of priceincreases and government programso Gas consump­tion per customer fell 22% between 1974 and 1988(Energy InformationAdministration 1988)0 To manyLDCs, their focus now should be on increasing gassales (rather than encouraging conservation)becausesupplies of gas are ample and prices have beenfalling for several yearso

Gas and Electric Industry Differences

Electric utilities are vertically integrated., while gasutilities are not Production, transmission, anddistribution of electricity are regulated primarily byPUCS, with PERC involved only in wholesale con­tracts <0 The gas industry is organized. and regulateddifferentlyo In contrast, natural gas is produced,transported, and distributed by three different setsof companies.. Gas is produced by unregulatedcompanies. Pipeline companies, regulated by PERC,move gas to local distribution companies 0 AndLDCs, regulated by PUCS, distribute gas toconsumers..

The time horizon for resource planning is generallyshorter in the gas industry than in the electricindustry.. Electric utilities, which construct powerplants that last 30 to 40 years, plan accordingly.. Gasutility planning depends on e "pment lifetime andmarket conditions as well as the length of contracts(less than 20 years)..

Electric utilities meet load instantaneously andmaintain high reliability to avoid outages.. ey donot store electricity and therefore maintain extragenerating to ensure reliability.. LDCs, onthe other hand, store gas and use interruptiblecontracts to maintain for their corecustomers..

For electric procedures to determineavoided costs are reasonably well-defined because ofthe decade of experience with PURPA Avoidedcost provides a benchmark against which to assessthe value of resources offered by private producersand DSM programs.. Unlike electric utilities, gasLDCs have a limited obligation to serve dual..fuel industrial customers, which complicates thedefinition of avoided cost, The avoided-cost bench­mark for natural gas, at least for the noncore

5.108 Hirst, Goldman; and kins

market, is based on world oil prices0 This bench­mark is volatile and difficult to predict Moreover,marginal-cost pricing is much less developed in thegas utility industrye For example, the marginal costfor gas could reflect limitations in pipeline capacityand alternative uses of the gas (e.g., for generatingelectricity), upstream considerations for the LDCo

Gas Market Characteristics

The LDC market is divided into two segments, coreand noncore. The core market consists ofresidential, commercial, and small industrialcustomers that depend entirely on the LDC for gassuppliese The noncore market consists primarily oflarge industrial customers and electric utilitieseThese entities can make their own arrangements forgas transportation, and can forego purchasing gasfrom the LDC and, instead, purchase gas directlyfrom producers0 In the noncore market, the LDCoffers two products: 1) the gas as a commodity, forwhich there is competition with gas marketers andindependent prOducers, and 2) gas transportation,for which the LDC has a monopoly..

Recent reports on natural gas production capabilityshow that the gas bubble may disappear within ayear or two (American Gas Association 1990)0 As aconsequence, LDCs will rely more on long-termcontracts and less on spot-market purchases fortheir gas supplieslI Because the amount of gas willlikely remain constant, U.Slb supply and demand willbe roughlyequat Significant regional differences ingas supply and prices are likely to persist, however,because of differences in pipeline capacity anddistance between gas supplies and markets..

Regulatory Environment

Throughout the 1980s, there was little interest inapplying IRP to the gas industry for three reasons:1) resource planning for electric utilities dominatedPUC interest, 2) regulators emphasized deregula­tion of the gas industry, and 3) estimates of gasresources seemed to assure an adequate supply ofgas at low cost, reducing the importance of long­term planning.

However, regulators in a few states, especially theDistrict of Columbia, Nevada, Washington, andWisconsin, are beginning to examine IRP for the gas

industry. PUCs are interested in gas planningbecause of:

CD Benefits of electric-utility IRP, especially inimplementing DSM programs;

., Recent requests for investments in new pipeline;

• Possible environmental benefits of using gasversus electricity; and

• Interest in fuel switching, including the use ofgas air-conditioning technologies to cut electricsystem peaks in summer.

Because IRP is often viewed skeptically by gasutilities, efforts to date have been started byregulators. Gas utility experience is often limited toits RCS and low-income retrofit programs. Theseprograms are generally not based on reliable fore...casts of future gas demands, sensitivity testing viapilot programs of DSM marketing and incentivemechanisms, or evaluation of DSM programs. Toachieve the next level of program development,more rigorous analysis (e.g., end-use models toforecast gas demands for each customer class)should be conducted to quantify the DSM potentialsin specific market segments.. To date, little end-usedata are available except for residential retrofits(Ternes et at 1990) and some new technology.applications (Brodrick and Patel 1990; .Gas

eareh Institute 1989).. The next stage will be theintegration of demand and supply options to assessthe best resource mix"

The ability to move to the next stage in gas IRP isconstrained because of the lack of analysis exploringthe implications of DSM as an alternative to gassupplies4> Few methods, such as the CaliforniaStandard Practice Manual (California Commissions1 7), exist to assess the cost...effectiveness of gas...

DSM programs.. Almost all current analysesexamine least-cost purChasing, selecting the best mixof supply and storage options to achieve low' pricesfor consumers and high earnings for shareholders(McDermott 1987)~ DSM is important for com­mercial and industrial customers with interruptibleservice dual-fuel capability, when gas supplyis or gas costs become too high because ofextreme weather conditions"

To advance gas IRP, several key questions need tobe addressed. What effect will DSM programs haveon LDC supply reliability and profitability? Whatare the economic implications ofelectric-utilitycost...effectiveness tests - societal, all ratepayer,participant, and no-losers ... to LDCs, theircustomers, and shareholders? How should fuelswitching be included in gas IRP? Can gas-utilityDSM programs be used to reduce industrial bypass?What regulatory adjustments are necessary toencourage gas IRP while maintaining companyprofitability?

IRP is beginning to change the way gas DSMprograms are designed, implemented, and evaluated.Gas LDCs can learn from the IRP methodsdeveloped in the electric industry, but they must becreatively applied given the different circumstancesin the gas industry. Gas DSM programs are in theearly phases of development& Activity is expected toincrease substantially as supply reliability and energyefficiency influence more PUCs to encourage LDCadoption of IRP concepts..

FEDERAL ROLES TO PROMOTEINTEGRATED SOURCE PLANNING

Because electricity production consumes almost40% of the primary energy used in the U.S..,electricity must be a major part of national energypolicy.. In addition, concerns about environmentalquality, economic productivity, internationalcompetitiveness, and national security suggest alarger role for the Federal Government in workingwith utilities to expand their planning and toimplement DSM programs.

Improving energy efficiency through utilities maybe a particularly effective way to reach millionsof V.S$ energy consumers.. Utilities have directmonthly contact with all their customers (i.e., meterreading and billing) and are usually well respectedorganizations in their communities.. Thus, theFederal Government can work with a few hundredutilities and, through them, reach tens of millions ofhouseholds and millions of businesses..

Integrated Resource Planning 5.109

Require FERC to Incorporate DSM Programs inUtility Wholesale Contracts

PERC approves all wholesale transactions amongutilities. Currently, PERC reviews of proposedcontracts involve no consideration of energyefficiency.. PERC, in its review of long-termcontracts, could require the buying utility to showthat it has acquired all the conservation and load­management resources in its service area that costless than the proposed purchased power beforePERC approves the contract. Presentation of anintegrated resource plan, approved by the utility'sPUC, could satisfy this requirement Use of such astate-approved plan in PERC proceedings wouldeliminate concerns that PERC was pre-emptingstate regulation. Implementing such an expandedreview of wholesale contracts might requiremodification of the Federal Power Act

Require Federal Electric Utilities to ExpandDSM Programs

The federal Power Marketing Agencies (PMAs, partof DOE) and TVA (an independent federalcorporation) account for one-tenth of the electricityconsumed in the U.S. Traditionally, TVA and theBonneville Power Administration (the largest PMA)have operated large DSM programs, which savedenergy for their customers and served as examplesfor other utilities. Unfortunately, short-term budgetconsiderations forced reductions in these programsat both agencies. Indeed, TVA canceled all itsconservation programs in 1989. Bonneville, on theother hand, plans to increase its conservationbudgets over the next several years.

New legislation could require these federal ppwerauthorities to expand their DSM programs and toexplicitly consider environmental and social factorsin their benefit/cost analyses of all resourcealternatives" Such legislation would be a logicalextension of the 1980 Pacific Northwest ElectricPower Planning and Conservation Act (P.L. 96­501), which explicitly made conservation theelectricity resource of choice and gave it a 10%bonus to be used in economic analyses of alternativeresources. The 10% bonus reflects the environ­mental and social benefits ofconservation compared

5.110 Hirst, Goldman, and Hopkins

to supply resources" The other federal utilities couldemploy similar factors in their resource assessments..

Expand DOE TechnologyOlOTransfer Activities toUtilities

DOE's Integrated Resource Planning Programmanages a variety of projects aimed at improvingthe long-term resource-planning process and tools(data and analytical methods) used by utilities(Goldman, Hirst, and Krause 1989; Berry and Hirst1989). DOE sponsored conferences on utilityplanning in 1988 and 1989, and plans additionalconferences in 1990 and 1991. The DOE programcould be expanded to fund additional cooperativeprojects with utilities and PDCs. This approachfocuses on cost-sharing projects, with DOEassistance provided through the DOE nationallaboratories and other government contractors.

The underlying rationale for expanding DOE'stechnology transfer efforts on IRP is the knowledgethat many innovative and successful programsoperate throughout the country. However, informa­tion on these successes is hard to get because thesponsoring utility and/or PUC has little incentive topublish information on the program. Thus, DOEcan play a valuable role by participating in theseprograms, ensuring that the programs are carefullyevaluated, and then funding preparation of reportsand conference presentations that effectively andwidely disseminate information to other utilities andstate agencies!> The Northeast Region Demand-SideManagement Data Exchange (NORDAX), funded inpart by DOE, is a good example of such technologytransfer. The initial phase of NORDAX, a con­sortium of more than 20 utilities, yielded a database with information on 90 DSM programsoperated by 17 utilities in the region (camera,Stormont, and Sabo 1989). Another critical area forDOE attention is the transfer of planning methods,data, and processes from electric to gas utilities.

Collect More Information on Energy Use

EIA is responsible for collecting, evaluating,analyzing, and disseminating information on energyreserves, production, demand, and technologies. EIAfocuses on the supply of, rather than the demand

for, energy. For example, EIA's -:Annual EnergyReview (EIA 1989b) contains separate chapters onfossil-fuel reserves, petroleum, natural gas, coal,electricity, nuclear energy, renewable energy,financial indicators, and only one chapter on energyconsumption. -

EIA (1989a) collects detailed information fromelectric utilities on individual power plants relatedto their construction cost and capacity; annualoperations and maintenance expenses; and monthlyfuel consumption, generation, availability, andemissions. Data collected by the Federal EnergyRegulatory Commission (e.g, PERC Form 1) aresimilarly detailed with respect to electricityproduction and disposition; purchases and sales;construction costs and operations for power plants;and costs and characteristics of transmission lines,substations, and transformers. Unfortunately, EIAand PERC collect no comparable data on utilityenergy-efficiency and load-management programs.EIA and PERC could expand the data-collectionforms completed by utilities (both electric and gas)to require information on utility DSM programs.This would help to redress the imbalance betweenthe supply and demand sides in data-collectionactivities.

CONCLUSIONS

More than half of the primary energy consumed inthe UeS. flows through electric and gas utilities 0

Therefore, the economic and environmental effectsof utility actions are enormous.. Integrated resourceplanning represents a new way for utilities to meetthe energy-service needs of their customers.. BecauseIRP is a comprehensive and open process, itsimplementation is likely to yield large benefits interms of an "optimized" mix of resources and fewercontroversies over utility decisions.

Much work is needed to convert the potential bene­fits of IRP into reality. This paper dealt with a fewof the most important topics, including changes inregulation to align the interests of utility customerswith those of utility shareholders, incorporation ofenvironmental factors into resource planning,COlnpc~tltlve auctions for resources, guidelines for

review of utility plans, planning for gas utilities, andincreased activities by the Federal Government

ACKNOWLEDGMENTS

We thank Ralph caldwell, Adrian Chapman, DruCrawley, Alan Destribats, David Friedman, JamesGallagher, Richard Hahn, William Hederman, BruceHenning, Steve Herod, Jonathan Lowell,Christopher McGill, Richard Ottinger, DianePirkey, Jonathan Raab, Carol Smoots, RichardSpellman, Sam Swanson, and Ed Vine for theirhelpful comments on a draft of this paper.

REFERENCES

American Gas Association 1990, Natural GasProduction Capability GO 1990, Arlington, VA,January&

Argonne National Laboratory 1989, EnergyConservation Program Evaluation.· Conservation andResource Management, Proceedings of the 1989conference, Argonne, IL, August.

LiD Berry and E. Hirst 1989, RecentAccomplishmentsof the U.S& Department ofEnergy's Least-Cost UtilityPlanning Program, ORNL/CON-288,Oak RidgeNational Laboratory, Oak Ridge, -TN, August

J.. Brodrick and R .. Patel 1990, "Defining the Marketfor Gas Cooling,," ASHRAE Jouma132(1), January.

California Public Utilities Commission andCalifornia Energy Commission 1987, StandardPractice Manua~ Economic Analysis ofDemand-SidePrograms, P400-87-006, San Francisco andSacramento, CA, December.

R. K Camera, Do Stormont, and C.. Sabo 1989,"Developing Reliable Data on DSM Programs: TheNORDAX Experience," Chapter 37 in Demand­Side Management Strategies for the 90s, Proceedings:Fourth National Conference on- Utility DSMPrograms,EPRI CU-6367, Electric Power Research Institute,Palo Alto, CA, April.

Carolina Power & Light Company 1989, CP&LExhibits 1, 2 and 3, Before the North Carolina UtilitiesCommission, DocketNo. E-100, Sub 58, Raleigh, NC,April..

Integrated Resource Planning 5" 111

R. C. Cavanagh 1986, "Least-Cost PlanningImperatives for Electric Utilities and TheirRegulators, The Harvard Environmental Law Review10(2), 299..3440

R. C. Cavanagh 1988, "The Role of ConservationResources in Competitive Bidding Systems forElectricity Supply," Testimony before theSubcommittee on Energy and Power, U.S. House ofRepresentatives, Washington, DC, March 31.

P. Chernick and E. Caverhill 1989, The Valuation ofExternalities from Energy Production, Delivery, andUse, prepared for Boston Gas Company, December.

C. J. Cicchetti and Wo Hogan 1989, "IncludingUnbundled Demand-side OptiOns in Electric UtilityBidding Programs," Public Utilities Fortnightly123(12), June 8.

S. Do Cohen, Jo H. Eto, e.. A Goldman, J. Beldock,and Go Crandall 1990, A Survey of State pucActivities to Incorporate Environmental Externalitiesinto Electric Utility Planning and Regulation,Lawrence Berkeley Laboratory, LBL..28616, May..

A F. Destribats, J& BII Lowell, and D. SII White 1990,"Dispatches from the ront: New Concepts inIntegrated Planning," New England Power ServiceCompany,.Westborough, MA, May~

Electric Power Research Institute 1988, Status ofLeast-Cost Planning in the United States, EM..6133,prepared by Barakat, Howard & Chamberlin, Inco,for the Electric Power Research Institute, Palo Alto,CA, December$

WII B,. Ellis 1989, "The Collaborative Process inUtility Resource Planning," Public UtilitiesFortnightly 123(13), 9-12, June 22$

Information Administration NaturalGasAnnual, DOEIEIA-0131(88)/2, U,.S. Departmentof Energy, Washington, DC"

Energy Information Administration 1989a, Directoryof Energy Data Collection Forms, DOE/EIA...0249(88), U&S.. Department of Energy, Washington,

February..

Information Administration 1989b, AnnualEnergy Review 1988, DOEIEIA-0384(88), UGS~

Department of Energy, Washington, DC, May..

5. 112 Hirst, Goldman, and Hopkins

Gas Research Institute 1989, Assessment of LargeTonnage Gas Fired Cooling Technologies forthe Commercial Sector, Chicago, IL, prepared byArthur D.. Little Company, GRI Report No.88...D162, June.

C. W.. Gellings, Jo H. Chamberlin, and J.. M. Clinton1987, Moving Toward Integrated Resource Planning:Understanding the Theory and Practice ofLeast-CostPlanning and Demand-Side Management, EPRI EM­5065, Electric Power Research Institute, Palo Alto,CA, February..

C.. Goldman, E. Hirst, and F$ Krause 1989, Least­Cost Planning in the Utility Sector." Progress andChallenges, LBL-27130 and ORNL/CON-284,Lawrence Berkeley Laboratory, Berkeley, CA, andOak Ridge National Laboratory, Oak Ridge, TN,May..

C" Goldman and E.. Hirst 1989, "Key Issues inDeveloping Demand-Side Bidding Programs,"chapter 23 in Demand-Side Management,,· Partner­ships in Planning for the Next Decade, EeNENational Conference on Utility DSM programs,EPRI CD-6598, Electric Power Research Institute,Palo Alto, CA, November.

Green Mountain Power Corp.. 1989, IntegratedResource Plan, South Burlington, VT, February.

Ee Hirst 1988a, "Meeting Future Electricity Needs,"Forum for Applied Research and Public Policy 3(3),Fall.

E,. Hirst 1988b, "Integrated Resource Planning: TheRole of Regulatory Commissions," Public UtilitiesFortnightly 122(6), 34..42, September 15~

E$ Hirst et at 1990, Assessing Integrated ResourcePlans Prepared by Electric Utilities, ORNL/CON...298,Oak Ridge National Laboratory, Oak Ridge, TN,February..

Hirst and Me Schweitzer 1990, "Electric-UtilityResource Planning and Decision-Making: TheImportance of Uncertainty," Risk Analysis 10(1),137-146, 1990..

Be F. Hobbs and P.. Maheshwari 1990, "A DecisionAnalysis of the Effect of Uncertainty Upon ElectricUtility Planning," Energy, forthcoming.

P. L.. Joskow 1988, Testimony before the Sub­committee on Energy and Power, U.S. House ofRepresentatives, Washington, DC, March 31.

P.. L. Joskow 1990, "Understanding the "Unbundled"Utility Conservation Bidding Proposal," PublicUtilities Fortnightly, January 4.

E. Kahn, C.. Goldman, S. Stoft, and Do Berman1989, Evaluation Methods in Competitive BiddingforElectric Power, LBL-26924, Lawrence BerkeleyLaboratory, Berkeley, CA, June.

J. Koomey 1990, Comparative Analysis ofMonetaryEstimates ofExternalEnvironmental CostsAssociatedwith Combustion of Fossil Fuels, LBL-28313,Lawrence Berkeley Laboratory, Berkeley, CA, May..

Massachusetts Department ofPublic Utilities 1990a,Investigation by the Department on its own motioninto proposed rules to implement integrated resourcemanagement practices for electric companies in theCommonwealth, DPU 89-239, Boston, MA, January"

Massachusetts Department ofPublic Utilities 1990b,Investigation by the Department on its own motion asto the propriety of rates and charges .. ." byMassachusetts Electric Company, DPU 89-194 andDPU 89...195, Boston, MA, March..

Co> Marnay and Go> A Comnes 1990, RatemakingforConservation,,· The California ERAMExperience, finaldraft, Lawrence Berkeley Laboratory, Berkeley, CA,February"

K A McDermott 1987, Wrestling lvith Transition,,"Least-Cost Planning in the Natural Gas Industry,Center for Regulatory Studies, Chicago,October"

D" Moskovitz 1989,Proftts & Progress Through Least­Cost Planning, National Association of .KeJgul~itOJrv

Commissioners, Washington, DC, November..

National Association of Regulatory Utility Com­missioners 1988a, Least-Cost Utility Planning, AHandbook for Public Utility Commissioners,

October..

National Association of Regulatory Utility Com­missioners 1988b, Least-Cost Utility Planning, AHandbook for Public Utility Commissioners, Volume

2, The Demand-Side,: Conceptual and MethodologicalIssues, F. Krause and J. Eto, Lawrence BerkeleyLaboratory, December..

National Association of Regulatory UtilityCommissioners 1989, Resolution in Support ofIncentives for Electric Utility Least-Cost Planning,Washington, DC, July..

New England Electric 1989, Conservation and LoadManagement Annual Report, Westborough,MA, May"

New York Public Service Commission 1989,"Opinion and Order Establishing Guidelines forBidding Program," Case 88...E241, Proceeding onMotion of the Commission as to the guidelines forbidding to meet future electric capacity needs ofOrange and Rockland Utilities, Inc., Decision 89­7, April 12..

New York Public Service Commission 1989, Opinionand OrderApprovingDemand-Side Management RateIncentives and Establishing Further Proceedings,Opinion No.. 89-29, Albany, NY, September..

Northwest Power Planning Council 1989, 1989Supplement to the 1986 Northwest Conservation andElectric Power Plan, Volume II, Portland, OR..

Ontario Hydro 1990, Providing the Balance ofPower:Environmental Analysis, Ontario, Toronto..

R.. Ottinger et at.. 1989, Environmental ExternalityCosts of Energy Resources, draft, Pace University,prepared for the New York State Energy Researchand Development Agency, Decembers

Pacific Power & Light Company and Utah Power &Light Company 1989, Planning for Stable Growth,Portland, OR, November.

Puget Sound Power & Light Company 1989,Demand and Resource Evaluation: Securing FutureOpportunities, 1990-1991, Bellevue, WA, Decembero

Ds Schultz 1990, "Utility Shareholder IncentiveMechanisms: A Technical Review and Comparisonof Shared Savings Proposals for Demand-SideManagement Programs," Division of RatepayerAdvocates, california Public Utilities Commission,Sacramento, CA, Januaryo

Integrated Resource Planning 5. 113

M. Schweitzer, E. Yourstone, and E. Hirst 1990,KeyIssues in Electric Utility Integrated Resource Planning,,·Findings from a Nationwide Study, ORNL/CON-300,Oak Ridge National Laboratory, Oak Ridge, TN,April.

Seattle City Light 1989, Strategic Corporate Plan1990-1991, Seattle, WA, November.

R. P. Sergei 1989, Testimony and Exhibits Submittedon Behalf of Granite State Electric Company, NewEngland Power Service Company, Westborough,MA, September"

5.114 Hirst, Goldman, and Hopkins

Mit Ternes, P. Sit Hu, Lit S. Williams, and P. Goewey1990. The National Fuel End-Use Efficiency FieldTest: Energy Savings and Performance ofan ImprovedEnergyEfficiency Measure Selection Technique, Draft,ORNL/CON..303, Oak Ridge National Laboratory,Oak Ridge, TN.

5.. Wiel 1989, "Making Electric EfficiencyProfitable," Public Utilities Fortnightly, July 6.

Wisconsin PublicService Commission 1989,Findingsof Fact, Conclusion of Law and Order, 05-EP-5,Madison, WI, April 6"