Integrated esourcRevie

lannin at as Utiliti s:f Initial Efforts

Charles Ae Goldman e lawrence Berkeley laboratoryMary Ellen F0 Hopkinsa The Fleming Group

Several local gas distribution companies (LDCs) have initiated integrated resource planning (IRP)processes in response to public utility commission (PUCs) directives.. In this study, we review the initialIRP plans and planning processes of four gas utilities .. These case studies illustrate the state-of-the-art ingas resource planning and identify technical and analytic issues in which additional work or improveddata are needed.. These areas include integration and linkage of detailed models used in various steps ofthe IRP analytic process, appropriateness of end-use forecasting methods, treatment of uncertainty offuture loads and load growth, system reliability criteria to be used in gas resource planning, the marketpotential for gas DSM programs, methods used to estimate gas avoided costs, and institutional andfinancial barriers to gas DSM ..

Introduction

Until recently, state public utility commissions (PUes)have focused most of their attention on development of. rated resource planning processes for electric utilities ..However, several PUCs are now looking closely at theplannJm~ processes of gas local distribution companies

in because of the increased control andreS'pOllSlt)l!11tv that LDCs have for their purchased gascosts and because of questions surrounding the potentialrole for gas end-use efficiency options .. A recent survey ofstate commissions identified 15 PUCs that are activelyde\relolDU'llSI or considering IRP for gas utilities (Goldmanand aU accounts, gas IRP is stiB in itsinfancy, as gas utilities have filed initial IRP in only fivestates (District of Columbia, Illinois, Nevada, Oregon,W~lSnm2~[OI1J in response to formal regulations. However,gas utilities in many more states believe that they win bereaUlr~~ to IRP or DSM For 41out of 85 gas expect to have an IRPprogram within the next two years, to theAmerican Gas Association (1992) ..

prelpart~ by four. """"'''' ... '''... '''''''''' ... of Columbia)

Gas & Coke CompanySouthwest Gas Corporation (SWG 1991a

and Washington Water Power CompanyOUf objectives are: (1) to assess progress

among gas utilities in responding to the challenge posed bytheir state regulators, (2) discuss how the regulatory andmarket context affects plan objectives, development andemlphaSlS, and (3) identify key technical, analytical, andinstitutional issues that arise in a comparative assessmentof these gas IRP

Approach

We reviewed the resource plans and related documents offour gas utilities.. This in itself was a significantundertaking as the longest IRP plan (Le*, DCNG)consisted of 15 volumes featuring an executive summary,main report, and 22 technicsJ appendices.. In addition, weconducted telephone interviews with staff involvedin the preparation of IRP plans& We used the generalcriteria and checklist developed by Hirst et aL (1990) as astarting place for our review: clarity, technicalcompetence, adequacy of the short-term action plan, andfairness of the plan.. However, Hirst et al .. developed theseguidelines based on a review of over 30 electric utilityplans, many of which were second and third-generationefforts .. Experience with electric utilities suggests that IRPis an iterative and evolving process, and thus expectationsregarding these initial plans should take account of therelative newness of gas IRP" we view the analysisas primarily exploratory, because of limited gas industryexperience and our small sample size"

tate egulatory equirements

Differences in state regulatory or legal requirements andpractices had a significant impact on the development ofinitial IRP plans of these four gas utilities. For example,the District of Columbia Public Service Commission's(PSC) order and subsequent regulations were quitedetailed, specific, and often prescriptive.. The PSCestablished ambitious conservation goals for natural gasutilities.. These targets set 1998 goals of 25 % and 35 %usage reduction in residential and multifamily sectorsrespectively and 18-25 % reductions in the commercial

Inl~8lJra~:ea Resource fll8'nnifnD at Gas Utilities: A Review of Initial Efforts - B667

LDC Responses to IndustryRestructuring and IncreasedCompetition

Each of the four utilities has been greatly affected by therestructuring of the federally-regulated segments of thegas industry, specifically, wellhead production andinterstate transportation. Prior to these changes, a LDCsleast-cost gas supply process was primarilylimited to daily cost optilnization decisions between a few

(e.g., one or more and storageoptions). The combination of comprehensive regulation ofaU segments, cou.pled with contractsbetween pipelines and that gas supply

was dominated the interstate pipelines.Ho~we"er_ during the last decade, wellhead gas sales havebeen deregulated and the Federal Energy RegulatoryCommission (FERC), in a series of orders, establishedconditions which made open access service for interstateDloeUIle transportation available to endusers, producers,and marketers. LDCs were then able to purchase gassupplies and firm transportation from various points asseparate unbundled services.

Southwest Gas (SWG) filed its first resource plan for thesouthern portion of its service territory in July 1990* TheNevada regulations require the Public Service Commission(PSC) to review IRP plans in formal proceedings and ulti­mately approve or disapprove of the utility's filed plan.The commission rejected the DSM component of SWG'splan and required SWG to refile its DSM plan in April1991* Based on a stipulation, the commission orderedSWG to evaluate an expanded list of DSM programs usingthe total resource cost test as the primary entenon.. InNevada, an explicit, desired outcome of LCP regulationsis an approved utility IRP plan"

The dramatic pace of these structural changes has led to amuch less predictable business environment for LDCs.For example, within 1-2 years after the advent of openaccess on interstate pipelines and the availability oftransportation for customer-owned gas supplies,Washington Water Power found that nearly aU of itsindustrial customers had moved from sales to transpor­tation customers (see Figure 1). Transportation serviceaccounts for roughly one-third of People's Gas annual gasload. In their plans, several gas utilities emphasize thenew operational and supply planning problems posed bysubstantial transportation volumes. These range fromprovision of short-term supply balancing services for end­use transporters to the longer-term planning issue oftransportation customers that want to

sector (with 70% reduction in commercial cooling). ThePSC did not formally impose these targets as require­ments; however, the PSC placed an implicit burden on the

to show why certain targets were either unachiev­able or lmeconomic. In addition, the PSC strongly encour­aged DCNG to develop its IRP plan in close conjunctionwith "collaborative ff working groups involving representa­tives from DCNG, the PSC staff, Office of People'sCounsel, and the DC Energy Office. The collaborativecreated several working groups on different topics whichmet over 70 times during a two year period (April 1988-­August 1990) to review and discuss virtually all aspects ofthe company's IRP activities. The working groups focusedprincipally on the development, design, and evaluationplans for a comprehensive set of DSM pHot programs.DCNG was also required to develop end-use load forecast­ing models, and commence data collection and analysisefforts so that it could properly assess the DSM potentialin commercial sector.

In Illinois, multiple state agencies are involved as thePublic Act of 1987 mandated that the IllinoisCommerce Commission establish administrative rulesimplementing least-cost and that the .»-JP''''"'''iJ .... Jll, ... &R.J1,"'""........

of Energy and Natural Resources (DENR) prepare a state­wide least-cost In terms of process, the statewide gas

£I~'J~I~:1nf~i first and established an overall policywhich included 20 recommendations on

various of gas resource planning * Representativesfrom an of the gas industry, public interestgroups, and commission staff reviewed drafts of theand recommendations through partici-

in a Natural Gas Plan Advisory Onewas to use DSM as an initial and

n~'M"'IIO~r'" source of new gas although its practicaleffect was muted because most utilities forecast noincremental need for the next 20 years in the faceof very slow or nonexistent load Utilities thenae,!elc)De~ individual which were filed in

1991 and had to be consistent with the state

were enacted in 1987 andreclullrea gas utilities to prepare an IRP in consul­tation with Commission staff and major stakeholders. InW~lShln11~tO]rn~ the commission has emphasized the Pia.nn:lngprocess and mechanisms which facilitate public involve­nlent. For WWP created a Technical AdvisoryCommittee which reviewed and commented on drafts ofthe WWP's least-cost plan; the commission held public

aU()WJlDil interested parties to comment on theThe utility's plan is not formally

the although the utility's actionsmust be consistent with its submitted least-cost

8" 68 - Goldman and HOJ'}kiflS

Industrial

75%

50%

25%[..O%l!iJiI

_ Gas Sales

1980 1983 1988 1990

1~ wastung~tonWater Power Annual Gas Sales

Table 1 illustrates some differences among gasutilities in terms of firm size indicated gasre<;luu~errlents).. the structure of gas demand intheir service the relative of gas

and the overall balance.Residential customers account for between 40-77 % offirm sales among the four utilities .. In terms of customer

(jV\l'eUln ~!s represent a significantof the residential gas market for PGLC and DeNG (63 %and 34% of residential gas sales respectively). Commer­cial sales over 30 % of total gas sales for threeof the four utilities. comparisons across utilitiesin the ell sector are more because of defIni-tional inconsistencies and forcustomers to their own gas and use the LDC

for tra]i1St.orltatlon~

Southwest Gas' Southern Nevada service issituated in a very with forecasted load

of 7.8% per year between 1990-2000. A signifi-cant fraction of its increase is drivenincreased use of gas in electric generation.. SWG and

to a lesser extent, are facing significant capitalinvestments in their pipeline and distribution system tomeet demand over the planning horizofie In.contrast, utilities such as DeNG and PGLe project thatgas demand will remain flat or decline over thenext decade. These projections are driven by macro­economic and factors in each utility"s service

shift back to LDC salescustomers lack alternate fuel CaT)alJllIH:Y

aU four utilities have been buffeted.res;tnICtllrung, it is to note differencesin their institutional and market which are

of the the gas ForWWP is a while the other

three utilities are 'WWP has -n1l"r~nA"cuil

an natural gas conversion program which wininstallation of efficient gas to electric

heat customers with central forced air furnaces (be~al..1sehave duct and gas heat customers with

electric water These gas conversion programsaccount for a fraction of the prcllectedin gas sales over the next decade 0 In contrast, the

the three utilities reflect the intensethat often occurs with the local

and the that fuel substitutionpose in this situation for inany pueso ForaU three utilities were

interested in new markets for gas that would theirload factor , increase summer gas theinitial of DeNG and PGLC focus on gas efficiencyprograms, while the importance of fuel substitutionprograms. SWG both gas efficiency and fuelsubstitution programs in its IRP Of these, thecommission two of the gas programs.

Uti1ities: A Review of Initial Efforts -

......

•••••.•.••••••••••••••••••••••••••.•••••••••••••••••••;~~

61,806

-0.26% ~O.36:%,> .•... ····•·· ... ·· •.•.•·•.•.•.•.•. ·· •.•.·.•.•·....•.•.•..•

..

,~($/Mbtu). ...

.

7~50

7·.12

..............

:.

:

... . ...

. ..

:

:.

: : .

...

:::: ::::::::<:>:> <.................. . .. . ... .....

that WWP's service is close to Canadian andRocky Mountain gas producing regions and two interstatepipelines (Northwest Pipeline Corporation and Pacific GasTransmission). SWG is located relatively near the majorgas producing regions of the U.S. southwest Illinois isparticularly well-situated in terms of access to gastransportation and storage as it is served by nine interstatepipeline and ranks second in underground gas storagecapacity. This facilitates competition among producingregions and pipelines on commodity costs, but also meansthat it is relatively easy for end-use customers to bypassthe existing LDC and buy gas directly from a nearbypipeline. Other such as Washington Gas Light,have historically been served by a few major pipelines andare more remote from major regions.

aec:Hnll1f industrial base forin older urban areas

as well asa result of

In residential

Because the gas is not integrated, it isalso to understand the historic relationships withgas and as they affect the strategic

r\nll"\A1If"'hl!1i"181t'H::lI.C! of LDCs. Access to multipleand pipelines, the relative costs of

h-r1lnn'IV1l€1lr gas to market from interstate pipelines, as well asthe costs to serve various customer classes, account formuch of the variation in end-user gas costs* For example,WWP's residential gas prices are significantly lower thanthe other three gas utilities, in attributable to the fact

70 - Goldman and HOJ'JkiI7S

Ian bjectives

Table 2 summarizes major objectives listed by the fourgas utilities in their initial IRP plans. Several utilitiesframed the planning exercise in terms of the ways inwhich their traditional gas supply planning process had to

••••••••••••••••••••••••••••••••••'...... '." ~

be modified to accommodate additional regulatoryrequirements. Ofteo, utilities provided specific objectivesfor both gas supply planning and DSM resource planning,which reflects to some extent the more limited degree ofintegration.

..

uy

,-,raS COmPly' fully withIUil1?isstateWide: Gas:Utility Pla.n'&requirements of ICC' least' costplanhi'ngregulations

@r........ .......... '

@.;,y-.

VIa J

@.',r.,;.

""v:),_t- 'V

@

In "@ e-

Li

erursueoest., •. --- " .. '- ~ ---- -'" . ,~;" >..

auul)JJ.l..nam. w

.1.'•., LV ;'1

r,

.

.~

:

•••••••••

It is clear that the turmoil in the gas industry has had itseffect on the planning horizons of LDCs as several of theutilities noted that their planning environment was domi­nated by the need for "transition strategies" to cope withfast-paced changes brought about by industry restructuringand new regulatory requirements. Not surprisingly, theplans reflect the fact that few gas utilities believe that theycan afford to focus much analytic effort on the long-term.In some cases, objectives are offered that appear to beprimarily reactive rather than stated as strategic goals forthe utility. For example, DCNG develops alternativescenarios which comply to different degrees with PSCDSM goals, indicative of the way in which commissiondirectives explicitly shaped the IRP plan. Some utilities(e.g., WWP) framed their objectives in terms of gettingstarted with an IRP process.

Public Involvement in PlanDevelopment and Review

Plan ontents: echnical andnaiytic Issues

Gas Utilities Rely on SeriesModels for IRP Analysis

Electric utilities typically rely on one of two generalanalytical approaches in performing integrated resourceplanning: (1) linked, detailed models and/or (2) integratedplanning models (Eto 1990). In the first method, utilitiesmust link inputs and outputs of individual, detailed modelsfor each step of the IRP process (e.g., load forecasting,generation planning, production costs, financial analysis)into an integrated process. In the second method, electricutilities use commercially-available integrated planningmodels which incorporate important elements necessaryfor comprehensive treatment of DSM and supply-sideoptions and where the major linkages are embedded in thesimulation model and are made transparent to the user.

Adequate participation in development and review byvarious stakeholders is one of the major criteria used byHirst et al. (1990) in assessing the fairness of an IRPplan. The four gas utilities employed a range ofapproaches in terms of the degree and stages of publicinvolvement As indicated previously, DeNG relied

on the expertise of formally constituted workinggroups of stakeholders that were involved in the

of a comprehensive set of pilot DSM programs andreview of its IRP plan. DeNG's approach essentially canbe characterized as a collaborative process, driven to a

extent by commission policy goals and withsubstantial involvement of in bothde'vel~oPineJlt and review.

In our small sample, gas utilities utilized the linked,model approach. This is not too surprising because theIRP analytic effort for most of these utilities involvedbootstrapping models and tools that were already availablewithin the appropriate departments into an integratedanalysis procedure. Some utilities (e.g., DeNG) had todevelop new models for the various steps in the IRPanalytic process. The apparent lack of stand-aloneintegrated planning models might also be a by-product ofgas industry structure (Le., vertically disintegrated) whichmeant that it wasn't particularly relevant for LDCs toconsider all strategic resource options from wellhead toend-user customer.

Figure 2 summarizes the analytic framework developed byPeople's Gas. The company notes that the most importantchange brought about by the LCP requirements was thecomplete integration of its to demandforecasting, demand-side planning, and supply-sideplanning.

The major steps in PGLe's process included: (1) developforecast of annual and design peak day loads underseveral alternative economic scenarios and futures; (2)using demand forecasts, develop Baseline Resource planbased on results of screening available supply-sideresource options in company-developed monthly gassupply optimization model (LINDO--Linear, Interactive,and Discrete Optimizer); this produces supply-only least­cost plan; (3) identify available DSM options; developsavings and cost estimates for each option; (4) conducteconomic analysis of 13 DSM program options usingestimates of avoided costs developed from Baseline

In contrast, SWG did not have a formal collaborativeprocess, but did hold three workshops which involved

SWG wasthe

review process. This is exemplified the stipulationthat SWG reached with the other parties. In Washington,involvement of non-utility occurred duringae'vel~OPine]lt and with input in the form

nrll'l'''C''~''1ll'''1&1 groups than the formal coHaborativesAs in Nevada, review of the initial IRP

commission staff and other parties resulted inrevisions (and additional

in the DSM planning portion of its filing.Public involvement took a different form in Illinois,

because of the two-stage resource planningprocess. Major stakeholders were involved in review and

consensus-building activities in the statewide gasplannJlng framework, with much less explicit involvement

development for the individual utility plans.

72 - Goldman and HOA'Jklf.lS

Demand­Forecasting

Demand...

Forecast otIndependent

Economic Variables

IntegrativeAspects

20 People's Gas IRP Analysis Frarnework

which

limitedModels

SHznltlCaJllt Q'f"110 11.,f" 1 00 and data gaps for eachare briefly discussed in the next sections.

Except for gas utilities relied oneconometric models to develop their long-term salesforecasts for residential and commercial customers. Forexample, WWP's econometric model for these two sectorshas 95 equations relating gas consumption (by class, rateschedule, and state jurisdiction) to weather, the economy,and price variables and is built on a historical data basebej~lrunlng in 1978. The model does not include cross-priceelasticities and assumes that new customers will choosegas water and space heating based on current patterns(i.e., 60% of aU new customers use natural gas). Typi­cally, gas utilities then forecast future loads for largeinterruptible and transport customers using customersurveys and analysis of future gas use by utility marketingaccount executives ..

Resource screening of DSM options conducted usingLOADCALC (a DSM screening software system devel-

a consulting options had to provide positivenet benefits using societal, participant, and utilityperspectives; (5) combine viable DSM and supply-sideoptions for further detailed analysis in LINDO model;model selects combination of supply-side and DSMresources; (6) utility considers several secondary issues

barriers, rateconsideration) and determines financial impact oncOlm!)at]lV in terms of earnings, and rates; and (7), select

Least-cost based on

One clear benefit of the IRP process is that gas utilitieshave had to with analytical and modeling tech-niques to integrate DSM options into the tradi-tional process* Prior to implementing

not consider DSM programs asin its least-cost supply

The various in the IRP process--forecasting,Oni:ImJlzatlon of the gas supply plan, and development andscxoeelnUJl2; of DSM options--were performed with relative

of sophistication among the four utilities. Theplanning process, and the requirements to

consider DSM measures as resource options, revealed

At the PSC's direction, DCNG developed end-use fore­casting models for residential and commercial sectors.The residential model estimates gas usage across sixprimary end-uses and three customer classes and attemptsto model and incorporate appliance choice decisions andchanges in appliance saturations. In thiscapability, DCNG conducted detailed and statistically

in1'eo:ral~e(J Resource PlannJrng at Gas Utilities,," A Review of Initial Efforts - 8" ]3

AU four utilities estimate usage recenthistorical data on the relationship between annual gasCOIllSUlnotlon and day demand. System load factors

range between 30-35 % and are defined as:

Concern over the and economicconsequences of of gas service severeweather for customers without short-term alternatives togas has meant that utilities traditionally place the highest

on in gas supply planning. Gasutilities seek to ensure that their gas supply isdiversified. and can meet the usage requirements of corecustomers under adverse weather conditions.

Several of the gas utilities used regressionestimates (base-load and weather-sensitive) to calculatepeak day usage under extremely adverse weather condi­tions ("design day") .. Typically, the design day peakdemand requirement was then increased upward anadditional reserve margin, which ranges between 5-15%.Rationales offered to support reserve levelstypically rest on Initial analysis thatthere appears to be substantial variation in "reservemargins It that are deemed appropriate among utilities. It isdifficult to discern if these differences are solelyattributable to unique characteristics of individual gas

, company's gas configuration oftransmission and distribution system, ofpeaking facilities) or reflect lack of ~eJler;aH~v-alcce~pte~

industry standards on supply planning reliability criteria.In an IRP context, reliability planning criteria assumeincreased importance because it has a direct bearing onthe relative mix of firm vs. non-firm gas supplies as wenas the comparative evaluation of the benefits of DSM vs ..supply resource options (Jensen 1991).

Peak day demand estimation methods vary in their degreeof sophistication among utilities. One utility assumed a33 % load factor for the entire forecast period. Anotherutility conducted an econometric analysis of the relation­ship between daily firm sales and weather by day type(Le., weekday vs.. weekend) during five recent winterperiods and then adjusted peak day demand over theforecast to account for efficiency impacts ..

where, average gas use equals annual gas requirements(Le., firm sales plus transportation volumes)/365~

Table 3 summarizes the approach and results from variousstages of each utility's DSM planning process: assessmentof DSM technical and market potential, proposed DSMprograms, projected DSM program expenditures, savings,and relative impacts .. We make the following observations:(1) Several utilities' assessment in their IRP plans of theDSM technical and market potential was somewhatnarrow initially, generally confined to a limited set ofresidential DSM options.. However, these utilities havequickly taken steps to remedy deficiencies and haveconducted more comprehensive level assessmentsof DSM resource opportunities, most often at the urgingof commission staf[ (2) Several utilities indicated thatthey do not have a of confidence in the keyinput assumptions (i.e., savings, incremental costs, and

(1)gas use/peak demand

representative surveys of single-family, multifamily, andcommercial buildings (disaggregated into 16 SIC codes) ..These surveys were also used to assess the level of energyefficiency in the building stock and constituted a majordata collection and analysis effort involving almost 4,000commercial and multifamily buildings and over 1,500single-family dwellings.. In addition, DCNG developed adatabase of gas consumption data using historic meteredbilling data at both the building and rate class level (e.. g .. ,commercial and apartment heating, non-heating, andcentral heating), which involved aggregating gas consump­tion on a whole building level from meter or accountingrecords..

These efforts illustrate the substantial baseline datacollection and model development which are required inorder for gas utilities to conduct an integrated andcomprehensive analysis of demand-side options in an IRP

One advantage of end-use-based models is that DSMresource opportunities and impacts can be more readilyaccounted for in forecasts of future loads and factors thataffect gas usage, such as thermal integrity of buildings,appliance efficiency and saturation, and gas consumingactivity can be accounted for explicitly. Another side­benefit is that this information is invaluable for marketresearch" Based on our smaU sample, few gas utilitieshave ,this at the present time. However, we

that the data requirements of along withbenefits competitive pressures),

will encourage more gas utilities to undertake theseactivities ..

LF ==

8~ 74 - Goldman and H04'J/(117S

. ':':' .:<:.-:.: .. :-: .. ,

@' S4.0ht·(92f:· .. ···

.. ," .. ".. " :-:.::.-:-:: ":::.>:.:< ,' .

.<.~..:.. ,~.::-J.;:-::'e.··:-·:·-::'Z:::·.V:.... ·(-f.·.'l-A.:-.··· :·.R·"~.';~o···-: U'::'r':~j;;<p:-:.·:·;l:j:.;:;;n·:::~g'.: ...:: ... ::.;.;:;:-:::., . .. ..4UUf,i J ~ "'iY.l.: ~.,. "''"' WTS "'." ::::<:<::::<:::::::::::: <::<:::::::<:::::>:/::::.: -:>;:.';:-:';-:;. .' :::::::>:::.::::>::::::::/

. . <:-";-:,.: '. . .: ::.: ; :: ::: j <. :: ::;:; :;.:;: .'::; (:(:;::" .'. .' .' :: :::,::.-: .:.:.:':.::::<.::;:' . ." .' ::::::<. :.::. :.:.:.:.: :.:.' : .:.:.:. . ;:.';':':':'::::::;:;';:::::;:;';:

:': ..>..,<>:. :'p~(jpie!s:'G~:s .;.·.:::::Y>j >vv~Shill}rtriri::W~t~r>P6W~r·· ..... .' -:: '.:'" .-:: DeNG::::::-:-:' . -: :::-:.:::::::.::::: '::.:-:::::::...•. i4pfugra~/ •..<~5te$lde~tiatetl,i¢i~Y ••~•••.•••"4v~gfflm~U$& .•..••••.....•...•........•.....•.•.•...•

.' :::.:. tlO··reside·nd~I;)rQ/lt/ :\:: :·;::·:))ijpt19ns?/H?\?(;:::··;://»:> )«";11 :Pf()i#~J#*:#i¥ted .....':' .. .' ::. <:::-::: :::-: :::':'<:.':::':'::;::::: ::~·::;\f~o·us:-Qtitue~ ::~ubstitu~:·\.··>·.,-: ': ..>'::: .' :::- :<. .. .

::: y}Ji~Irprt>gra~{-::::/r:):;: >:<:..:.::: :.:: :.?:(: '::\::::::>:'-:\«\>:\.:::< :'. :.' :..':::::«::: <:::>:·~::·:·:::~i~ng::~~Y~r~~$.~~s~;·.-:·:···::-:::: :.:; :;:.::::::: .:: ..::\::/:;.~:.:\<::::<;.::::.. . ."::::::::;:;::::::::::?:j:;:j:j:;j.::!.

: . ;:/:::::::ili~nt:~t~~::~1:#i9#:::~~{;;·. ::. ':':'::.' :'-: '" :::;: :::: ::I:>::);::/:.::.:.:.:. '.. .... '.' ':':'.' '.':':':'.':':':

..... SOOie~I'~~i~i~a~;w~;i~ii~fr~~il......... .. . ./.·utility testS> ... .. ···.···.i~ct,~*~~~¥#c()~f.· •.·.:· •. ·:;:·•••.••:...•..••.•:.•.:· .•.r..·.: ...m:.n::.:.~.•.·::pt.•.·:e.::••..•Q.::.•:.:.•~:..:ti.::..:..~.•.:.:ta:..s:·:.:.•::f!::'p.:l:::••:i:·a.:.~::.n.•:l·.:..e:.o..:m::.·.•::·..·.~.:.•c::.·:•. a::a.~•.~•.e.•.~:U.•.:d'.a.:.·::'::.:.~~:.b:$.:.;:.•t•.~..t:.::.reae:••::'n.•. :.·.da....•~...:.••.e.ng...•:t...•.I.•.:.:••.••.•.•••.••:.•:..•••.••:•.•.•.•••..•.••••.\

.1i:;;::~:;,~~~~:·f ~V', r". . .

........................................ :.: ::>:' ::<:;:):jn~ntill)nwact¥~:\.<::.:-:.::- ." ·:>:·riOn~parii~~p~~t~#s.f·.· .: :::.:::. : '.>::::.. '.' .:.::/9~risistencS; :w~ili.:>·: .':" .. 0 .' .:·Quali~ti~t~9:~t§~r: :;: :':";::..' ":':;' ::«

.::~orpoiite>~bjectiv.es·, . '. ~ctept#ilci~·~··:~~t#~#.j~&·: .:'., :.:: :: .. :.: '.:.::.:':. proven: market···· . :a¢ceptance/~e.~~~ilty~·: '.' .. .performance.' . ·.·nurilbet:o(r~~:::rydets~··.. .. .,

...... ::.:i.mm.~.••·.~.i.~.~.i..e..•~.;'.••. c.••.••.re. a.m.:: : ..:."<:.:., ::':::.-:.:':-:-:'

.. ~ ··~:'~:e~lr~6be~· .. < .. :'1988)<,':-:-:':::':-:': . .'.' : :<.,:: .. ' . :':.::.: .:.:.: :·.·· .. ::::.. ·.::·····:··::::····;···:·:·;·::·:·::::·;(i:::::

'@:., $3 Jllvf{~l).::. :: :". )(>:: :.>/:\\Ho .. ·Proposed·:·:SO.36rvf(92) .. ~ ::-::S:t6M (92): -: ..:'@ Pf<jpo~d.:::$·3.~20Nf(93)· "'$ ·····:·SS·.2M '(93)': .

" ' . . . '.. ':'.:.- : .::' .. -: ;.'. . ::: .: . . . ....:. .. .' . . .. '..' .. -:->:.: ''':'. :./::: :.,::. ::::.;<::.::::>:. :: -:> :,-:< .;:. «<. »::<::

. 0' :- O..~7'%·of :an.tual . of ·a~l.al gas . :.•. ·:·.;¥e:et d<>.ijj®:~lQt{t8rgets::<: ....

• .~~~~t::u~ltI9:U9b.5):)::/::U?>::·::;:••~~:foq~:r~entsrivet nexf ·@~:;l:Ji~~~~:n .reqiJiremenf(2000). . "@.-- "::~f8% ~. 9;6 %.. of. .'systemwid~,'("'·:--

.................... ' . @ 0.'04%~. Ojj5.%. of '. .' .xevenue:s '(in 'l992~

rev~nue$.(iii'.t9·92y:199.3- .:::·.J~93: tespe:¢tivelyJ. .'t'eSp~cti~~ly) .. .

~ .... .

1991; WWP 1991b;'DCNG 1990 t .'

.·••• i·..·.· ..

s . ....( progr~m~···

.:

. <

'.: '~"

.....

....... :-<.:<:::.

.... .• :>

because are not drawn from theirown actual Their own In iI""Qlr'll~'Ilo"!l""'"

DSM programs is limited to energy audit for aUbut PGLC6 Some utilities stated that their estimates ofnlarket and program features were illus-trative in an initial indication of the relativecost-effectiveness and contribution of DSM in a least-cost

In estimates of the aggregate DSMin their service territory were often

limited data on characteristicsOUl.lOIIUZ and stock~ (3) Based on this

there are fewer gas efficiency options thatn'lr'.....~'ll!1r~.Q Sll1l;DlUl'CaIlt resource savings at much lower costs

aUt~mcaU\'es'& which is a distinctive feature ofby many electric utilities (eego,

sector and water heating

dominate residential and comm.ercial gas andsome efficiency opportunities have been and will continueto be realized through appliance efficiency standards asweU as comprehensive weatherization efforts initiated bygovernment, utilities, or customers .. In addition, analysisof industrial DSM options is technically more complex(and often constrained by proprietary concerns related toprocesses), while analysis of the economic benefits to gasutilities are complicated by industry structural changes(e.. g., customer-owned gas, end-user transportation)e (4)For two utilities, estimated savings from gas DSM pro­grams are quite small (eog., 1-2%) relative to annual gasrequirements, while the systemwide effects are minimal atthe other two utilities given current DSM programs (seeTable 3) ..

In1'eara1"eCi Resource ffJlCJ'nnJfng at Gas Utilities." A Review of Initial Efforts - 8.75

Analyzing Economics of DSM: What Testto Use, to Estimate Benefits?

In screening DSM options, aU four utilities reliedprimarily on the results of the total resource cost testresults in the economic analysis of DSM programs, oftenat the commission's direction. In Nevada, SWG wasordered to obtain better utility and rate impact measure(RIM) values for one of its DSM programs prior toimplementation, despite a satisfactory TRC test value. Atthe other utilities, DSM options were not eliminatedbecause they failed the ratepayer impact measure (RIM) ornon-participants test, although several of the gas utilitiesnoted their serious concerns about potential rate impactsassociated with gas efficiency programs..

The gas industry has also raised practical and methodo­logical concerns with regard to estimating the long-termbenefits of gas efficiency programs and utilized divergentapproaches in developing gas avoided costs. These includedifficulties in estimating gas commodity costs over theplanning horizon, disagreements regarding the extent towhich demand or capacity-related costs can beavoided by DSM, and the use of simplified proxyaPtPro~lch(~s to estimate avoided gas costs 1991).

promoted (e.g., package cogeneration, gas chillers).Comprehensive program evaluations are underway forthese pilot programs and DeNG win not proceed to fuU­scale implementation until evaluations demonstratebenefits of programs in their own service territory.

Southwest Gas experience in DSM program implementa­tion included the Residential Conservation Service (ReS)energy audits during the active period of the federal Resprogram and residential weatherization activities. Thecompany's revised DSM program Plan consisted of 13DSM programs, four of which were fuel switching pro­grams. Of the 13 programs, the PSC approved two.Objections to the other programs included RIM testvalues, the sensitivity of the fuel-switching issue, and thehigh cost of the Plan (estimated at $3.5 million). The twoDSM programs, which only recently received approval,are: (1) weatherization retrofit plan, and (2) boiler retrofitwith heat recovery program. Both are expected to beimplemented in 1992, with estimated costs of $335,000.

and Strategic loadUtility's wvGolden

As it DeNG was required to

ImlpleJmeJt1t a set of DSM pilot programs,which it did in late 1988. Initially, DeNGoffered 17 DSM programs targeted to all majorcustomer groups: residential (9), multifamily (2),commercial (2), and multi-family/commercial (4) (seeTable In each sector, programs build off of an initialenergy which then are complemented by rebate orloan programs for high-efficiency equipment, controls,and weatherization measures. Several the pilot programsare innovative either in terms of program design (e.g.,

rehabilitation) as weB as technologies

As indicated earlier, of the four utilities, only WWP andSWG proposed fuel substitution programs in their initialIRP plans. These utilities proposed that fuel substitutionefforts would account for a significant portion of theiroverall DSM activities. This is not to suggest that fuelsubstitution and strategic load building were not a majorconcern for the other two utilities. In fact, fuelsubstitution and load building were extremely important toboth PGLC and DeNG. It appears that differences amongutilities have to do more with timing rather than sub­stance. For example, PGLC identifies barriers to strategicload growth as one of the two potential barriers toimplementation of its Least-cost plan. Specifically, PGLCis interested in pursuing compressed natural gas poweredvehicles, gas-fired cogeneration, and gas air conditioning.The utility calls upon the Illinois commission to re­examine policies that provide electric utility competitorswith an advantage in certain end-use markets (e.g.,promotional allowances for electric heat pumps), allowPGLC to recover expenses related to promotion of thesenew market opportunities, and support the utility's ratedesign and promotional proposals where gas utilizationcould reduce overall consumer energy costs (PGLC 1991).

Initial evidence suggests that utility desires in the fuelsubstitution area have been thwarted somewhat by reluc­tant PUCs. For example, the Nevada commissionexplicitly deferred a decision on SWG's proposed fuelsubstitution programs and ordered that a special

DSM

the four actual experience implementingman,ye...~,ealle gas DSM programs is limited.. PGLe had the

experience implementing gas efficiencyprograms at the time its was filed. Beginning in

the offered six DSM programs, whichincluded loan programs for single and multi-familyDUllOlruz owners, incentives to purchase high-efficiency gaseQtJuprnenlt.» a program at religious andseveral informational programs. PGCL completed an

evaluation of its program, which ledto modifications in program PGCL offered twoDSM programs ,multi-family andeQluplmeltlt 'Il~"'''',Q'njh'l1~ 11 at the time its IRP was fiJe,L

8 .. 76 - Goldman and H04'JklflS

investigatory docket be opened to address 'the fuel­switching programs and other fuel substitution issues..Other commissions have also not been particularly anxiousto confront this issue head-on .. However, the issue isfundamental for gas only and combination utilities andappears to be driven by the underlying system economics..Annual load factors for gas utilities tend to be low (30­35 %) compared to typical values in the electric utilityindustry (50-60%) .. Not surprisingly, many gas utilitiesseek to develop new gas loads, in off-peak periods, thathave load factors greater than their average load factor.. Inso doing, gas utilities seek to reduce systemwide averagegas costs on a per unit basis, essentially spreading fixedcosts over larger volumes ..

to transportation customers.. However, a variety oftraditional and innovative cost recovery approaches havebeen suggested by the industry (AGA 1992).. It is athreshold question for serious utility involvement on thedemand-side.. Mechanisms that seek to overcome financialdisincentives for the utility to pursue gas efficiencythrough recovery of "net lost revenues" are more difficultto implement. They will typically involve the utilitydocumenting gas savings and corresponding "lostrevenues, " which will involve a significant commitment toongoing DSM program evaluation, an area in which fewgas utilities have much experience..

Conclusion

Regulators often cite the policy goal of comparable andconsistent ground rules for electric and gas utility planningas a primary motivation for initiating IRP .. However, ourreview of first-generation gas IRP plans suggests thatthese processes have to be tailored to the conditions,circumstances, and structure of the gas industry.. Itappears that the most successful processes have occurredin working environments that are relatively "non­threatening" to the utilities: workshops, collaborativeprocesses to design DSM programs, or joint researchprojects which investigate controversial or technical topics

, fuel substitution or DSM potential) ..

Based on discussions with gas utility staff, it is apparentthat IRP processes significant staff resources, are

time-consuming, and involve major infrastructureinvestments, not in pipe and compressors, but principallyin human resources. Moreover, the IRP process requires abroad interdisciplinary team consisting of staff fromvarious departments (eGg., planning, engineering, market­ing, rates, affairs)G In reviewing IRP plans,PUCs are insisting that a gas utility demonstrate that aserious effort has been made to analyze supply and DSMresource opportunities in a consistent and comprehensivefashion. In establishing a gas IRP process, explicit policyguidance from regulators is most needed in the followingareas: (1) balancing of various economic tests, (2)ensuring comparable earnings opportunities for DSM, and(3) interfuel competition and promotional practices.. Mostcommissions have adopted a flexible approach in terms ofbalancing plan contents with the actual experience base ofgas utilities.. However, utilities will be expected toproceed up the IRP learning curve quickly in terms ofanalytical sophistication and data requirements.. For gasutilities, potential benefits of an IRP process includeestablishing a framework for utilities and regulators toaddress and reconcile short- and long-term resourceplanning objectives, ensuring fairly-structured competition

Institutional and Financial Barriersto DSM

of traditional load management programsallowed electric utilities and their regulators toexperience in developing cost recovery mechanisms fordemand-side interventions that preceded large-scale DSMprograms" In contrast, despite the fact that many gasutilities offered informational audit and weatherizationprograms to residential customers in the late 1970s and

1980s, this appears not to have producedand standardized procedures for cost

recovery for these of activitiesG in addition to'IlflrhlC't1lMl1 concerns "lost revenues" and poltenltlal

of fixed costs that arise from gasprograms, several gas utilities mention cost

recovery issues in their initial IRP For example,PGLe states: "The recovery of incurredcosts associated with the implementation of DSMprograms is critical if utilities are to have aincentive to pursue DSM resource "U'1V1l.·llVA.Jl."'GTo remove a disincentive associated with aor on recovery, the is acost recovery mechanism which illustrates the Company's

for direct costs of DSM progralllSGlost

established rate

Issues related to direct cost recovery may initially besomewhat more difficult for gas utilities compared toelectric utilities because of more limited experience on thedemand-side and because of cost allocation issues related

In its PGLe stressed that Im]Jleluerltatlon of its two-action which DSM capability

upon the commission'saPIJropnate DSM cost recovery and margin

In1'eara1~ed Resource Plann~rna at Gas Utilitiess" A Review of Initial Efforts - 77

among fuels in end-use and creating new marketopportunities for gas&

Acknowledgments

Hirst, E~, M. Schweitzer, E .. Yourstone, and J. Eto.1990~ "Assessing Integrated Resource Plans prepared byElectric Utilities .. " Oak Ridge National Laboratory,ORNL/CON-298, Oak Ridge, Tennessee.

The work described in this paper was funded by theAssistant Secretary for Conservation and RenewableEnergy, Office of Utility Technologies, Office of EnergyManagement Division of the U&S. Department of Energyunder Contract No. DE AC03-76SF00098. The authorsthank Adrian Wally Kolberg, Vivian Scott and

Dombrowski for their helpful comments&

Jensen, V. 1991. "Illinois Experience with IntegratedResource Planning for LDCs." Proceedings of 3rdNational Conference on IRP, NARUC, Washington,D.C., ApriL

Peoples Gas and Coke Company (POLe) 1991."Integrated Least Cost Plan. If Volume I, Chicago, Illinois,January.

References Southwest Gas Corporation. 1991a.. "1990 SouthernNevada Resource Plan. 'f January.

Southwest Gas Corporation 1991b. "1990 SouthernNevada Resource Plan: Revised DSM Plan Technical

Washington Water Power Company (WWP). 1991a."Managing Natural Gas Resources: Options for the

Least Cost Plan." Vols. January.

Water Power (WWP). 1991b.*'Evaluation of Demand-side Management forWashington Water Power. Vols.

of 1~ools for~lannlng." Lawrence

California.

American Gas Association. 1992. USurvey of Gas UtilityIRP Activities. it Market Brief 1992-2, Virginia.

District of Columbia Natural Gas 1990.Least. Cost Plan." Formal Case No.

Phase Volumes

American Gas I\S~;OC:latlon'!l I!Js~ru"lno and _11;.t~'lI"'.8'"

1991. DSM Evaluation and CostKe<~overv Issues for Gas Distribution Utilities." V1"~~n~'r'~t1

Inc.,

UOllOman'll C. A., and M. E. 1991.State Activities on Least Cost ~ll)nnjBnO'

Utilities." La\vrenceCalifornia.

8" 78 - Goldman and HOA'J/(1f.1S