Statewide Joint IOU Study. of Permanent Load Shifting

Transcription

1 Statewide Jint IOU Study f Permanent Lad Shifting Prepared fr: Suthern Califrnia Edisn, Pacific Gas and Electric, San Dieg Gas and Electric December 1, 2010 CALMAC Study ID SCE Page 1

2 Table f Cntents 1. Executive Summary Definitin f PLS PLS Cst-Effectiveness Value Prpsitin t End-User PLS Market Assessment PLS Prgram Recmmendatins Intrductin and Purpse f Study Plicy Backgrund CPUC Regulatry Backgrund Other Plicy Backgrund PLS Pilts SCE PG&E SDG&E Definitin f PLS Study Methdlgy Cst effectiveness PLS Mdeling Inputs Cst-effectiveness Tests PLS Avided Cst Benefits Key Sensitivities Brad Scenari Analysis ( Matrix ) Cst effectiveness f individual cases Specific Case Studies: Simulated Installatins Specific Case Studies: IOU PLS Pilts and Recent Installatins End-user Value Prpsitin Market Assessment Mdeling Results Cst-effectiveness Avided Cst Benefit Matrix Bill Savings and Rate Payer Neutral Incentive Sensitivity Analyses Specific Case Results: Simulated Scenaris Specific Case Results: PLS Pilts and Recent Installatins Case Studies: Generatin Capacity Value Sensitivities End User Impacts Prject Paybacks befre Incentives Incentive Level Requirements Sensitivities t Incentive Level Requirements Market Issues and Stakehlder Feedback State f the Industry Permanent Lad Shifting Prgrams Stakehlder Feedback Prgram Design Recmmendatins

3 6.1. Overall Cst-effectiveness f PLS Ttal Resurce Cst Test Cnsideratins fr a Market Transfrming PLS Prgram Cnclusins Based n the TRC Test PLS Prgram Design Framewrk Standard Offer PLS Prgram Design Characteristics System Design Build and Perfrmance Test Operatins Establishing Incentive Levels fr Standard Offer Ratepayer Neutral Incentive Levels Incentive Levels based n Expected Payback Cnsideratins fr RFP- based Prgram Designs Cnsideratins n Retail Rate Design Perfrmance Based Incentives Bibligraphy Appendix A: Avided Csts.. A-1 Appendix B: Stakehlder Feedback...B-1 Energy and Envirnmental Ecnmics, Inc Page 2 2

4 Glssary CE CEC CPUC DR DSIRE EM&V IMIOC ISAC PAC PG&E PLS RIM SCE SCHW SDG&E TES Tn-hur TOU TRC Cst effectiveness Califrnia Energy Cmmissin Califrnia Public Utility Cmmissin Demand Respnse Database f State Incentives fr Renewable Energy Evaluatin, Measurement and Verificatin Internal melt ice n cil Ice strage air cnditining Prgram Administratr Cst Test Pacific Gas and Electric Cmpany Permanent lad shifting Ratepayer impact measure test Suthern Califrnia Edisn Stratified chilled water San Dieg Gas and Electric Thermal energy strage Unit f cling energy (equivalent t 12,000 BTUs) Time f use Ttal resurce cst test Energy and Envirnmental Ecnmics, Inc Page 3 3

5 1. Executive Summary The purpse f this study is t investigate cst-effectiveness and prgram design t expand the use f permanent lad shifting (PLS) within the SCE, PG&E, and SDG&E service territries ( Jint Utilities ). PLS refers t a brad set f technlgies that shift electricity use frm peak t ff-peak perids. This reprt is an utcme f the Califrnia Public Utility Cmmissin (CPUC) Order D Decisin adptin demand respnse activities and budgets fr 2009 thrugh 2011 and will prvide mre infrmatin t the Jint Utilities n PLS fr use in preparing prpsed Demand Respnse prgrams, including PLS, t the CPUC. Energy and Envirnmental Ecnmics, Inc. (E3) and StrateGen Cnsulting were selected by the Jint Utilities and the CPUC t cnduct this study. E3 and StrateGen Cnsulting (the prject team ) used a cllabrative stakehlder prcess with tw wrkshps, numerus stakehlder interviews and meetings, and the release f a publicly available cst-effectiveness tl t develp the study results. The prject team als gathered and used data frm each f the utility PLS Pilt Prgrams, and technlgy vendr data in the public dmain and under Nn-Disclsure Agreements (NDAs). As described in this reprt, the study addresses the fllwing areas: Definitin f Permanent Lad Shifting Cst-effectiveness f PLS PLS Prgram Best Practices and Stakehlder Input Prpsed PLS Prgram Design Elements, including Standard Offer 1.1. Definitin f PLS Fr the purpses f this study, the prject team prpsed and uses a brad definitin f PLS. With supprt f the stakehlder grup, a technlgy neutral Energy and Envirnmental Ecnmics, Inc Page 4 4

6 definitin was prpsed based n the impact f the electricity usage prfile, rather than the technlgy used t create the impact. Additinal guiding principles include business/wnership neutrality, and the measurable shift at prgram level fr EM&V. PLS is defined with the verarching gal f rutine shifting frm ne time perid t anther during the curse f a day t help meet peak lads during perids when energy use is typically high and imprve grid peratins in ding s (ecnmics, efficiency, and/r reliability). The type f lad shape impact that meets the PLS definitin can be delivered by technlgies in three brad categries; electrical energy strage, thermal energy strage, and prcess shifting (see Table 1). Each technlgy categry and individual technlgies within each class have their wn unique csts, benefits, strengths and limitatins. Fr example, sme f the technlgies are mature and in wide use, such as thermal strage systems fr building cling systems, and sme are still emerging such as electric battery strage; sme prvide a static set shift in lad pattern, while thers can prvide a dynamic respnse based n electric system cnditins. There are als prcess shifting effrts that invlve rescheduling the use f electricity. Fr all f these categries, it will be extremely challenging t create a single, simple, technlgy neutral PLS prgram design that apprpriately addresses the differences in the csts and benefits f the technlgies t establish a cmmn design framewrk. Table 1: PLS technlgy applicatins, categries and examples Applicatin Categry Primary characteristics/ examples Statinary Thermal strage Generate ice r chilled water at night, then use this stred ice r chilled water t prvide cling during the day. Statinary Nn-thermal strage Chemical batteries, mechanical strage e.g., fly wheels, mdular cmpressed air Statinary Mbile Facility prcess shifting Plug-in electric vehicles (CAES) Prcesses cnducted within a facility that are shifted frm ne time perid f the day t anther Nt in scpe (Because mbile strage has a cncurrent prceeding at the CPUC) Energy and Envirnmental Ecnmics, Inc Page 5 5

7 While the PLS definitin is brad, there are many elements that this reprt has fund t be utside the scpe f PLS. First, PLS is nt slely event-based demand respnse. Secnd, PLS is nt behavir-based energy efficiency. PLS is prvided and quantified by discrete equipment r cntrls, nt slely by general custmer behavir mdificatin, and it des nt reduce the level f custmer service. Third, the lad reductin and shifting that can be achieved by best practices cmmissining, retr-cmmissining r adjustment f cntrls is nt cnsidered PLS, unless such practices are being applied directly t existing legacy PLS technlgies (such as unused thermal strage tanks) and are nt currently being implemented thrugh energy efficiency prgrams. We als exclude, by stakehlder cnsensus, the inclusin f electric vehicles in PLS. Finally, PLS is nt achieved thrugh fuel switching PLS Cst-Effectiveness The prject team emphasized the imprtance f cst-effectiveness f PLS thrughut the develpment f the study. E3 fcused n the verall scietal and ratepayer cst-effectiveness f PLS, given current Califrnia electricity market cnditins. StrateGen Cnsulting fcused n the value prpsitin t the enduser and whether a given PLS prgram design was likely t result in significant adptins. This apprach was designed t prvide mre infrmatin t the Jint Utilities as they decide the scpe and scale f their prpsed PLS prgrams and t prvide mre infrmatin fr establishing incentive levels that balance the csts t ratepayers and expected prgram adptin rates. T value the benefits f PLS t ratepayers, and t Califrnia as a whle, E3 develped a PLS cst-effectiveness framewrk that is similar t the framewrk used t evaluate the benefits f utility distributed generatin prgrams such as the Califrnia Slar Initiative (CSI) and the Self-Generatin Incentive Prgram (SGIP) [Decisin , August 20, 2009] 1. A similar framewrk is als currently being cnsidered fr use in evaluating the cst-effectiveness f demand 1 Energy and Envirnmental Ecnmics, Inc Page 6 6

8 respnse [R ]. The precursr t each f these was the develpment f avided csts fr energy efficiency adpted by the CPUC in 2004 and 2005 [R ] 2. The avided cst benefits prvided by PLS include electrical energy, lsses, ancillary services, system (generatin) capacity, transmissin and distributin capacity, envirnmental csts, and avided renewable energy purchases. We als investigated the renewable integratin benefits f lad fllwing and vergeneratin that culd be prvided by PLS. As shwn in Figure 1, using this new PLS cst-effectiveness framewrk, the lifecycle value f the avided cst benefits f PLS technlgies (assuming 15 year prject life estimates) is in the range f $500/peak kw t $2500/peak kw, depending n the number f hurs the PLS system can shift lad, and what hur the lad shifting starts. These figures are calculated based n the kw value f the lad shift and are technlgy neutral, and d nt include benefits frm ther value streams. They assume the best case peratinal prfile in that they assume the maximum lad shift every day f the year, and ff-peak usage at the least cst perid during the night. Fr example, a 6-hur lad reductin beginning at 12pm ver an assumed 15-year life is valued at ~ $2200/kW (r $365/kWh stred capacity). 2 Energy and Envirnmental Ecnmics, Inc Page 7 7

9 Figure 1: Brad Scenari Analysis Avided Csts Avided Cst Benefits $/kw Peak Capacity Reductin $3,000 $2,500 $2,000 $1,500 $1,000 $500 1 Hur 2 Hurs 4 Hurs 6 Hurs 8 Hurs 10 Hurs $ Shift Start Hur While the prject team believes these figures are apprpriate fr currently available PLS technlgies, we nte that the benefits in Figure 1 d nt include the prvisin f ancillary services such as regulatin that sme PLS technlgies plan t prvide 3. In additin, sme stakehlders have suggested that a 15-year life is t shrt and lnger lived installatins will have greater lifecycle value. T address these issues, the reprt presents these sensitivities and many thers. Fr example, an assumed 30-year prject life cycle is estimated t increase lifecycle avided cst benefits by apprximately 30%. The main results are als shwn in terms f lifecycle $/kwh-stred, which is a cmmn capacity metric fr batteries. The in-situ cst-effectiveness f bth simulated and real installatins (such as frm the utility PLS pilts) are als prvided. Using the Califrnia Standard Practice Manual (SPM) framewrk fr evaluating the cst effectiveness f ratepayer funded prgrams that the CPUC relies n fr 3 A number f battery technlgies prviders have indicated their ability and interest t prvide ancillary services as well as lad shifting. Energy and Envirnmental Ecnmics, Inc Page 8 8

10 ther distributed resurces, the installed PLS system csts must be less than the lifecycle benefits in rder t pass the Ttal Resurce Cst (TRC) test. While the installed system csts specific t PLS are ften difficult t ascertain (fr example, due t custmer cnfidentiality), r the csts were btained under a nndisclsure agreement (NDA) and cannt be shared in this reprt, certain classes f thermal strage are likely t pass the TRC (e.g., warehuse precling achieved by cntrls mdificatins, imprvement f existing thermal strage systems, medium-sized ice-based strage, chilled water fr new cnstructin and expansin applicatins); these technlgies are mre mature and their lifecycle values tend t be within the range f the avided cst benefits. Emerging grid cnnected battery technlgies and smaller scale 4 thermal strage systems with higher csts are less likely t pass the TRC cst-effectiveness test at their current system csts. One f the bjectives f this study was t determine what level f incentive payment wuld be apprpriate. Frm a ratepayer perspective, an incentive can be prvided t reduce the incremental csts f PLS systems ver standard nn- PLS technlgy withut any crss-subsidy at a level equal t the lifecycle benefits presented in Figure 1 less the bill savings the end-user receives by perating the PLS system. One can think f the bill savings as paying the enduser fr the scietal benefits they prvide with their PLS system. This study finds that even when the PLS peratins are designed t maximize bill savings, there are sme situatins when an incentive payment can be prvided withut any crss-subsidy. The actual value f this ratepayer neutral incentive level depends n the PLS system peratin and the specific retail tariff. Using a generic rate that is representative f medium and large cmmercial custmers rate structures, we find an incentive payment f ~ $100/peak kw t $800/peak kw fr PLS is pssible withut any crss-subsidy. When mdeling specific IOU rates, the rate payer neutral incentive levels range frm rughly 4 Smaller scale thermal strage is defined as units < 10kW, such as thse installed n small cmmercial buildings that d nt have central cling plants. Energy and Envirnmental Ecnmics, Inc Page 9 9

11 -$800/peak kw t $1600/peak kw. Ratepayer neutral incentive levels fr specific installatins are als prvided in the main bdy f the reprt Value Prpsitin t End-User While the ecnmic analysis f demand side prgrams fcuses n the csts and benefits t sciety and the funding levels needed t develp cst effective prgrams, custmers will ultimately need t see the direct benefits f PLS technlgy adptin t their cre business t justify their investment f capital and time, and their assumptin f varius prject risks. The StrateGen Cnsulting team evaluated the end-user value prpsitin t determine incentive levels that wuld be needed t prmte the likely adptin f specific PLS technlgies, based n stakehlder feedback n custmer-specific required payback perids. The analysis als prvides insights int ther elements f the prgram design that are imprtant t encurage PLS technlgy adptin, such as the investment business mdel, financeability, and mitigatin f tariff risk related t changes in bill savings ver time. Numerus stakehlders prvided cnsistent input that the end-user s financial hurdle fr adptin is a minimum 3 t 5 year payback. This is a significant financial hurdle that typically requires greater than 15% internal rates f return. Stakehlders als unifrmly expressed cncern n hw tariff structure changes can undermine the ecnmic return f PLS prjects, and that t date, such tariff risk has been largely uncntrllable. StrateGen tested these required payback hurdles by cnducting a prject-specific value prpsitin analysis f simulated PLS systems and IOU pilt prject data. Fr simplicity, a $/max kw incentive level fr shifted ff peak lad was calculated t achieve three and five year paybacks. Hwever, it is imprtant t nte that such incentives can be structured in a variety f ways, which is further described in the prgram recmmendatins sectin. The fllwing graph verlays several simulated PLS system payback scenaris fr varius building types in different Califrnia climate znes fr thermal strage, alng with simulatins fr battery strage simulatins fr manufacturing building Energy and Envirnmental Ecnmics, Inc Page 10 10

12 lad prfiles. The simulatins cmpare the amunt f the required incentive levels ($/kw) fr encuraging PLS custmer adptin fr 3 and 5 year paybacks. Als included are tw f the SPM cst effectiveness evaluatin tests 5 fr cmparisn: Figure 2: Required Incentives. Lifecycle Benefit & Ratepayer Neutral Incentive Levels Upfrnt Incentive in $/kw Maximum Peak Reductin $6,000 $5,000 $4,000 $3,000 $2,000 $1,000 $0 $1,000 Office, Chilled Water, Suthern CA Hspitality, Chilled Water, Central Valley Lifecycle Benefit 3 year Payback Incentive 5 year Payback Incentive Ratepayer Neutral Incentive Levels Hspitality, Med. Ice System, Central Valley Refrigerated Warehuse, Suthern CA Theater, Small Ice System, Suthern CA Sim Office CZ 3 Actual Incentive Sim Retail CZ 3 Sim Office CZ 4 Sim Retail CZ 4 Sim Office CZ 12 Sim Retail CZ 12 Sim Office CZ 13 Sim Retail CZ 13 Manf Facility, Lead Acid Battery, Suthern CA Manf Facility, Lead Acid Battery Lad Leveling, Suthern CA Manf Facility, Flw Battery, Suthern CA Manf Facility, Flw Battery Lad Leveling, Suthern CA Thermal Strage Installatins Thermal Strage Simulatin Battery Simulatin The chart abve indicates that the required incentive levels fr the thermal strage simulatins range frm abut $100 t $1,000/kW t achieve a 5 year payback fr the end user and apprximately $860 t $1,800/kW t achieve a 3 year payback. The battery simulatins required incentive levels range frm $1,100 (5 year payback) t ver $5,000 (3 year payback) t achieve required custmer investment payback levels. It is imprtant t nte that the battery simulatins were perfrmed fr nly tw different battery technlgies amng a wide range f pssible battery technlgies. The results will vary tremendusly depending n the specific type f battery technlgy used. Fr many simulated examples, the 3 and 5 year payback incentive levels are less than the ttal lifecycle benefits, but are still greater than the ratepayer neutral incentive levels. 5 The Prgram Administratr Cst (PAC) and Ratepayer Impact Measure (RIM) tests Energy and Envirnmental Ecnmics, Inc Page 11 11

13 1.4. PLS Market Assessment The assessment f the PLS market pprtunity is based n an verview f PLS incentive prgrams in the U.S. and stakehlder feedback gathered frm Califrnia IOU prgram persnnel, third party vendrs, engineers, PLS technlgy suppliers, and ther individuals and cmpanies 6. The majrity f the prgrams arund the cuntry are utility-spnsred thermal energy strage standard ffers. Other prgram types include special TOU rate structures r technlgy-neutral lad shifting prgrams. The fllwing cnclusins are based n a review f fifteen utility prgrams in the U.S.: Funding feasibility studies imprves utcmes and custmer cmmitment, and is a cre part f many prgrams' incentive structure. A number f prgrams ffer special TES/PLS rates that accmpany incentives, which nt nly reduce tariff risk and prvide greater certainty fr ecnmic return, but als imprve payback and encurage efficient system peratin. Prgrams that d nt prvide an adequate up frnt incentive will struggle t attract custmers, particularly in tday s challenging ecnmic climate. Utility-wnership reduces csts thrugh increased purchase vlume and mre efficient custmer targeting, but this mdel may nt be f interest t many utilities due t the cmplexity f utility wnership fr behind-themeter, custmer sited assets (particularly very small PLS systems). While this study is explring a variety f PLS technlgies, due t PLS prgram eligibility requirements, it is imprtant t nte that mst f the prgram design feedback reflects experience with thermal energy strage systems frm the PLS pilts. Table 2 summarizes the stakehlder feedback int cnsensus feedback, r feedback that was expressed and agreed upn by mst stakehlders, and nn- 6 Over 30 stakehlder interviews were cnducted Energy and Envirnmental Ecnmics, Inc Page 12 12

14 cnsensus feedback, representing areas f disagreement regarding the ideal apprach t encuraging PLS. Table 2: Cnsensus and nn-cnsensus feedback Cnsensus Lack f cnsistent and transparent rate structures that prmte PLS are an impediment A standard ffer is preferable t an RFP, as it mre easily encurages technlgy neutrality, and participatin by smaller stakehlders Incentive levels need t take int accunt all prject and market entry csts, deliver 3-5 year payback, and nt exclude any technlgies frm participatin Cnsistency in prgrams acrss IOU service territries is imprtant Prgram cmplexity adds csts and discurages market participatin Lack f educatin/training abut PLS technlgies their design, implementatin and peratin is a severe challenge Nn-Cnsensus Desired incentive levels and structure f incentive (e.g., Tariff based nly r tied t capacity/ hurs shifted) Tailring f incentives t technlgy class and size. Required metering/mnitring, specifics as t what needs t be mnitred and at what level f detail Allcatin f PLS budget (e.g. marketing vs. implementatin funding) Ptential fr market expansin 1.5. PLS Prgram Recmmendatins There are a number f dimensins by which the CPUC can cnsider standard ffers fr PLS prgram design. The mst fundamental dimensins are the prgram structure and the mnetary value f the incentive itself. The fllwing chart illustrates these dimensins, each with its wn respective cntinuum. Shwn left t right, a PLS prgram at ne end f the spectrum can have n impact t ratepayers. In this case, the incentive wuld be ratepayer neutral. This level f incentive culd have a lenient prgram limit since there is n crsssubsidy fr ratepayers, nr an explicit gal f encuraging large amunts f well-perated PLS systems in the field. At the ther end f the spectrum wuld be incentives whse levels are set based n the technlgy cst t encurage mre rbust cmmercial adptin at the technlgy specific level, perhaps based n achieving certain payback r internal rate f return requirements by targeted end-users. This level f incentive wuld be useful t encurage market Energy and Envirnmental Ecnmics, Inc Page 13 13

15 transfrmatin f the PLS technlgy, wuld have tighter prgram caps t prtect ratepayers, and a gal t reduce csts (and incentives) ver time. Frm tp t bttm, the prgram can be geared tward incentivizing energy shifting n peak ver time, r, at the ther end f the spectrum, be mre fcused n pure capacity. Figure 3: Standard Offer Prgram Design Framewrk Given the currently higher csts f grid cnnected battery applicatins and smaller scale thermal systems, the Jint Utilities and the CPUC may cnsider develping prgrams t encurage these technlgies fr market transfrmatin reasns, as they can play a rle in prviding a high value use f super ff-peak renewable energy generatin ( ver-generatin ) in the future. As described in the prgram design findings, shuld the Jint Utilities and CPUC seek t develp an incentive prgram fr PLS, we recmmend segmenting the PLS prgram ffering int at least tw general technlgy categries; a mature PLS technlgy categry that is available t any PLS technlgy with nearly ratepayer neutral incentive levels; and an emerging PLS technlgy categry that prvides higher incentive payments (thugh limited in quantity) t specific PLS technlgies such as small thermal strage and electrical battery strage that have the ptential t prvide mre dynamic system respnse in the future Energy and Envirnmental Ecnmics, Inc Page 14 14

16 suitable t supprt renewable integratin. We recmmend that prcess shifting be further evaluated t determine apprpriate industries and lads t target fr prgram develpment. 7 In additin, a number f best practices were bserved frm the pilts and ther PLS prgrams natinwide that are wrth cnsidering fr Califrnia. The fllwing summary f the PLS prgram design recmmendatins shuld be cnsidered: Divide PLS Prgram int at least tw categries based n technlgy; ne fr mature large scale PLS, and ne fr emerging PLS technlgies, with different prgram designs and gals. Mature: Large scale PLS deplyment that minimizes ratepayer incentives and prvides thermal-based slutins Emerging: Market transfrmatin fr strage with fcus n integratin with renewable resurces and energy efficiency Prgram design shuld address each f the three stages f the PLS system deplyment thrugh incentives, reprts, r EM&V, t increase the quality f the deplyed PLS systems. These include; (1) feasibility and design f PLS systems, (2) quality cntrl f cnstructin and pst-cnstructin functinal perfrmance testing, and (3) persistence f PLS peratins. Prvide cnsistent and predictable bill savings t encurage lng term custmer investment in PLS technlgy, that 7 These prgram recmmendatins are based n ur survey f best practices, utility pilt data analysis, stakehlder interviews, cst-effectiveness results, and wrkshp discussin. Energy and Envirnmental Ecnmics, Inc Page 15 15

17 Prvides a financeable level f lng term rate stability t encurage the initial capital utlay in a PLS system. This can be dne with a separate PLS rate, r by a guarantee f minimum n- t ff-peak rate differentials r grandfathering existing TOU rates Offers a super ff-peak rate t encurage charging after midnight r 2am when the vergeneratin prblem is expected t be the wrst and energy has the lwest cst, and Encurage sustained PLS perfrmance using perfrmance-based incentives and regular EM&V; Perfrmance-based incentives culd be achieved thrugh ne f tw appraches depending n technlgies; A PLS tariff with TOU rate differentials prvides sme incentive t perate the PLS system well, and des nt require a specific baseline develpment. This apprach is mre suitable fr thermal strage. A standard ffer mdel based n an energy payment ($/kwh shifted) prvides a direct perfrmance-based incentive, but wuld require strict guidelines fr calculating baselines fr thermal r prcess shifting PLS technlgies. Therefre, this apprach is easier t prvide t electrical battery systems. This apprach als reduces ptential fr gaming with battery systems (where batteries are used fr nn-pls purpses such as fr prviding uninterruptible pwer supply). Bth incentive appraches shuld be cupled with an EM&V requirement t prvide an peratins reprt and peratinal data f the system and the whle custmer lad. Energy and Envirnmental Ecnmics, Inc Page 16 16

18 Incentives and incentive structure directly influence PLS design and peratins, s it is imprtant t prvide incentives cnsistent with prgram gals Simplicity and transparency f the perfrmance metrics are critical t minimizing prgram cst and encuraging custmer adptin As per the CPUC rder that initiated this reprt, the Prject Team has included a detailed discussin f a PLS standard ffer prpsal that culd apply generally t any permanent lad shifting technlgies including, but nt limited t, thermal energy strage. The specifics f the Standard Offer are cvered in detail in Sectin 6 f this reprt. Energy and Envirnmental Ecnmics, Inc Page 17 17