Energy Efficiency Workshop: Power Markets, System Benefits & Key Design Issues

Transcription

1 Energy Efficiency Workshop: Power Markets, System Benefits & Key Design Issues Rich Cowart, Regulatory Assistance Project Chris Neme, Energy Futures Group May 9, 2011 The Regulatory Assistance Project 48 Rue de Stassart Building C, BE-1050 Brussels, Belgium Phone: web:

2 Workshop Agenda Part 1: Wholesale power markets introduction Energy markets Day-ahead, Real-time and Locational pricing markets Capacity markets Part 2: System Benefits of EE Participant bill savings Lower market clearing prices Lower system peak demands Easier to meet RES % goals Lower Reserve Margins Avoids T&D Upgrade costs Reduces Line Losses Part 3: Design Issues Performance-based obligations Balancing competing objectives Designing for long-term goals Low income Special treatment of industrials? Decoupling Markets Approach Quality control EM&V 2

3 Part 1: Introduction to Power Markets 3

4 When the laws of man and the laws of physics conflict, physics will win 4

5 Wholesale Markets: A (partial) issues list (1) Bid stacks and clearing prices Single price auction (2) 3 types of energy markets: Bi-lateral, day-ahead and real-time ( spot ) (3) LMP -- Locational marginal prices (4) Reserve margins and other reliability rules (5) Long-term capacity and resource adequacy

6 On the Demand Side: Typical load curve across 24 hours But load and generators are not really so smooth so additional services are needed to keep load and supply in balance 6

7 On the Supply Side: Generation -- Marginal Cost Curve 7

8 How They Combine: Bid Stack and the Market Clearing Price 8

9 Market Clearing Price: Simple example Portfolio of generators ready. What is the clearing price? Generator Units: MW MW MW MW MW MW Load

10 Three Power Markets: New York example NYISO Day-Ahead Market 45 50% Bilateral (forward) Contracts 50% Real Time <5% Bilateral Contracts outside the NYISO 50% NYISO Day-Ahead Market 45-50% NYISO Real-Time Market <5% 100%

11 29,000 27,000 25,000 23,000 Hourly New England Load (2005) Load Duration Format For less than 100 hours per year, New England needs an additional 2,500 MW of capacity to maintain reliability 21,000 MW 19,000 17,000 15,000 13,000 11,000 9, Hours per Year 2005 ISO New England Inc. 11

12 Some weeks, hours are much more expensive than others Hourly Day-ahead Market Prices MISO and PJM August afternoons H ours M onths H o u r

13

14 Location matters too: congestion and locational marginal pricing Load Densities - Southern New England

15 Example: Locational Marginal Prices (LMP) in New York State Note: Western NY at $48/MWH, Long Island at $88/MWH

16 Capacity Markets exist in a few places too: Forward Capacity Markets (FCMs) New England and PJM Power Markets FCMs cover regions with installed WA generation capacity on the order of: MT OR ID 22% of the EU-27 member states WY combined NV UT The combined installed generation CO CA capacity of Poland, Germany and Belgium AZ NM ND SD NE KS MN WI IA IL MO AR MI OH IN KY TN VT NY PA WV VA NC SC ME NH MA RI CT NJ DE MD DC MS AL GA TX LA FL 16

17 Coal--Existing Cumulative PJM Capacity Revenues ($42 billion over 6 annual auctions) Gas--Existing Hydro-Existing Nuclear-Existing Oil-Existing Renewables-Existing Renewables-Planned Oil (Planned) 0.00% Other Renewables (Existing) 0.68% Nuclear (Planned) 0.00% Oil (Existing) 8.14% Other Renewables (Planned) 0.05% Demand Resources 2.43% Energy Efficiency Resources 0.07% Demand Response Energy Efficiency Nuclear (Existing) 21.06% Coal (Existing) 30.01% 70% to existing fossilfueled plants; 20% to legacy nuclear; Where are the new gas, flexible peakers? Hydro (Planned) 0.00% Gas (Planned) 0.66% Hydro (Existing) 4.91% Gas (Existing) 31.83% Coal (Planned) 0.16% Source: Market Monitoring Analytics (PJM) 17

18 Part 2: The Many Benefits of Efficiency 18

19 1. Energy Efficiency Has Many Production Energy Production Capacity Avoided Emissions Transmission Capacity Distribution Capacity Line Loss Reduction Avoided Reserves Plus Non-Energy Benefits including: Add l resource benefits (water), building durability, health & safety Power System Benefits Cents per kwh Avoided Reserves Line Losses Distribution Capacity Transmission Capacity Avoided Emissions Production Energy Production Capacity * Note: numbers presented in graph are Illustrative

20 1. Participant Bill Savings: High Returns from Efficiency Investments Thru ESOs Cost of Conserved Energy (Euro/kWh) Retail Price (Euro/kWh) Benefit Cost Ratio United Kingdom Electric Gas France* Electric Gas Italy Electric Gas *Values from France are kwh cumac cumulative savings discounted over the measure life.uk and Italy are discounted lifetime values a well. Values are not directly comparable due to differences in metrics used to derive them. Note: these numbers compare customer and utility costs and retail prices. Data from: Energy supplier obligations and white certificate schemes: Comparative analysis of experiences in the European Union; Paolo Bertoldi et al. Energy Policy V38:3 March 2010 p1465

21 2. Efficiency Lowers Clearing Prices benefits all consumers Lower clearing price benefits all consumers Demand reduction from EE 21

22 The Public Value. of Efficiency. Demand Curve Price Assumed Demand Supply Curve P1 P2 Shaded area is savings to wholesale market Q2 Q1 Quantity 22

23 Efficiency Multiplier: New England Examples EE program costs are almost entirely offset by power market price reductions obtained by all consumers (in addition, program participants will have lower consumption and lower bills) Connecticut: Cost of Energy Saved = 2.8 cents/kwh Bill reductions via DRIPE = 2.72 cents/kwh Massachusetts: Cost of Energy Saved = 3.36 cents/kwh Bill reductions via DRIPE = 3.31 cents/kwh Source data: Avoided Energy Supply Costs in New England, July 2011 Report, Synapse Energy Economics; Saving Energy Cost Effectively: A National Review of the cost of Energy Saved Through Utility Sector Energy Efficiency Programs, ACEEE report U092, Friedrich et al September

24 3. System Peak Capacity Savings Efficiency measures save at time of system peak Virtually all measures Some measures much more than others Need for new generating facilities deferred/reduced E.g., 1 million CFLs saved ~5 MW at U.S. peaks Capital investment saved Market clearing price for peak capacity can also fall 24

25 New England Load Duration Curve System Load (MW) Hours per Year 25

26 New England Load Duration Curve w/demand Response System Load (MW) Hours per Year 26

27 New England Load Duration Curve w/energy Efficiency System Load (MW) Hours per Year 27

28 Residential Lighting Load Shape 30.0% 25.0% % of Lights That Are "On" 20.0% 15.0% 10.0% 5.0% 0.0% System Peak: Res & Com cooling-driven Hour of the Day Winter Summer 28

29 4. Demand-side participation reduces cost of capacity/reliability to ALL customers Calculation for New England (6 states) Demand-side resources reduced total capacity costs by ~$290 million per year Savings >15% of total cost Calculation for PJM for the latest auction : Total consumer savings $1.2 Billion due to D-S resources Savings 10 to 20% regionwide, 30% in the constrained power zone These are systemwide savings -- Participating EE and DR customers also receive capacity payments and pay lower bills * 29

30 New England Capacity Market: 1 st Auction Result Final Results of ISO-NE FCM Auction #1 Supply available (MW) at end of round 40,000 39,000 38,000 37,000 36,000 35,000 34,000 33,000 32,000 31,000 30,000 Start at $15.00 $9.00 $8.00 $7.00 $6.00 $5.625 Final Capacity Needed (ICR) Yellow boxes give end-of-round price $5.25 $4.875 $4.50 Start Close Round 30

31 5. Easier to meet Renewables Obligation Hypothetical: System need in 2020 without EE: 100 GW 20% RPS = 20 GW 25% EE reduces system need to 75 GW 20% RPS now = 15 GW 5 GW reduction in renewables need Result is either: Financial savings Increase RPS requirement to 27% (20/75 GW) at no extra cost 31

32 6. Reserve Requirements lower loads mean lower reserves 10 MIN SYNC 600 MW min 30 MIN 600 MW 10 MIN NON-SYNC What is needed in addition to 10 min SYNC (600) to total 1200 MW 10 MINUTE TOTAL 1,200 MW SINGLE LARGEST CONTINGENCY 1,200 MW RESERVE REQUIREMENT: 1,800MW

33 7. T&D System Savings from EE Passive deferrals System-wide savings impact every local area peak Magnitude varies a lot: depends on customer & program mixes Defers need for all load-driven T&D investment Active deferrals Intentional geographic targeting of EE to defer T&D Additional EE over and above system-wide obligations; or Targeting system obligation to areas of greatest system benefit Requires integrated planning to assess Opportunities for deferral Cost-effectiveness of deferral 33

34 Importance of T&D U.S. Power Sector Investment Needs, (billions of $) AMI, EE & DR $90 Distribution $619 Generation $537 Transmission $317 34

35 Residential Lighting Load Shape (2) 30.0% 25.0% % of Lights That Are "On" 20.0% 15.0% 10.0% 5.0% 0.0% Local peak: Res heating-driven System Peak: Res & Com cooling-driven Hour of the Day Local peak: res cooling-driven Winter Summer 35

36 Passive T&D Deferrals Routinely estimated benefit in leading US states Typically valued at ~$50 to $80/kW-year Worth about $2 per CFL But not usually very carefully analyzed Con Edison New York City utility is exception Separately forecasts impacts of system-wide EE on each of its 91 network and load areas for 10 years Forecast over $1 billion in reduced capital expenditures 36

37 Example: New York City Substation Peak MW Need if no EE Need with EE demand without EE demand with EE current capacity Source: Data from Figure 1 in Gazze, Chris & Madlen Massarlian, Planning for Efficiency: Forecasting the Geographic Distribution of Demand Reductions, in Public Utilities Fortnightly, August

38 Active T&D Deferral Keys Institutionalization of long-term T&D planning Synchronize T&D forecasts with EE forecasts Requirement to pursue least cost path Equal financial treatment of T&D & EE Acquire good data on EE savings load shapes smart grid tools may help Relationships with system planners Integrate with demand response, other resources Expect the unexpected both good and bad 38

39 Active T&D Deferral History in U.S. Successful pilots in US in early 1990s But rarely pursued until last few years Several interesting pilot efforts VT, NV More on the way RI, ME, Northwestern states New York City is major exception Active use of targeted EE since

40 Active Deferral New York City Results ( Benefits & Costs in millions of $) $500 $450 $400 Other Savings $350 $300 $250 T&D Savings $200 $150 $100 $50 Energy Savings Other Costs Vendor Payments $ Benefits Costs 40

41 8. EE Lowers Line Losses Losses should be applied for both energy and capacity savings Average losses are typically used in ratemaking for recovery of losses Marginal losses measure the change in losses due to change in load Approximately 2x average losses Average Marginal losses are typical the average of the marginal loss savings over a period of time Losses (MVA Losses rise rapidly as a function of load Average Losses = 8% Marginal Losses = 15% Load Level (MVA)

42 Part 2: EE Initiative Design Issues 42

43 North American Experience Overview >30 states/provinces have savings obligations >10 have had them for 20 years or more 6 have annual electric obligations of 2.0% or more Most aggressive spend 3-5% of system revenues Total U.S. utility spending of $5.5 billion in 2010 ~30% annual increase since 2006 increases expected to continue due to growing goals 43

44 Performance-Based Obligations Monetary performance incentives lead to success Rewards seem more effective than penalties Rewards seen more positively, can drive utilities or energy suppliers to significantly exceed goals Penalties viewed negatively, promote very conservative behavior, only get you to minimum requirements Top 5 U.S. states all offer monetary incentives Several exceeding 2% electric savings obligations Performance metrics must align with policy goals Seems obvious, but often not entirely the case 44

45 Balancing Competing Objectives Total Energy Savings Savings Depth per Participant Low Income Participation 45

46 Balancing Competing Objectives (2) Eminently doable Need to put constraints on how savings obligations will be met Minimum % of savings from low income (UK) Minimum % of savings from each county (VT) Minimum % of savings from long-lived measures (draft ESD) Must also transform specific markets (various U.S.) Minimum % of savings from deep retrofits (none yet) Performance incentives should be tied to all goals 46

47 Designing for Long-Term Goals Most obligations expressed as 1 st year savings E.g. savings must be at least 1.5% of annual sales Encourages cheaper short-lived measures Option 1: express obligation as lifetime savings E.g., Ontario gas EE guidelines (2011) Option 2: cap % savings from short-lived measures E.g., draft EU ESD? Eliminates worst, but doesn t encourage best solutions Option 3: annual goals are cumulative over time E.g., 1.5% annual = 15% cumulative after 10 years Discourages worst, but not enough to incent 20 or 30 year measures Option 4: goal as 1 st year savings, but with min. avg. life 47

48 Low Income/Fuel Poverty Understandable concerns about EE rate impacts Quantification of all system benefits may alleviate concerns Targeting EE to low income also helps Very common in US (typically 15% to 25% of EE) Much larger portion of UK s residential program Some jurisdictions use lower cost-effectiveness threshold for low income EE Reduced credit and collections costs (system benefit) Other substantial non-energy benefits 48

49 Large Industrials Concerns Many argue that they ve already done all costeffective EE to avoid paying for more Studies routinely prove otherwise but argument often has political sway Opt out agreements in some U.S. jurisdictions System charges set aside for those who opt out but they must demonstrate that they are doing EE Puget Sound Energy example: have two years to spend system charge unspent funds become available for other large industrials to access. 49

50 Decoupling Profits from Sales for Wires and Pipes Companies Distribution Utility revenues and profits are linked to unit sales (kw, kwh, therms, etc.) But, in the short run, a utility s marginal costs are only vaguely related to demand for gas or electricity (more on this in a moment) Loss of sales due to successful acquisition of customersited resources energy efficiency and distributed generation / combined heat and power will lower utility profitability This is true regardless of the means of delivering the energy efficiency and other programs The effect may be quite powerful 5% loss of sales could yield a 60% loss in profits There are several ways to address the problem 50

51 Assumptions for a Sample Distribution Utility Assumptions Operating Expenses $160,000,000 Rate Base $200,000,000 Tax Rate 35.00% Weighted Cost Rate Dollar Amount Cost of Capital % of Total Cost Rate Pre-Tax After-Tax Pre-Tax After-Tax Debt 55.00% 8.00% 4.40% 2.86% $8,800,000 $5,720,000 Equity 45.00% 11.00% 4.95% 7.62% $9,900,000 $15,230,769 Total % 10.48% Revenue Requirement Operating Expenses $160,000,000 Debt $5,720,000 Equity $15,230,769 Total $180,950,769 Allowed Return on Equity $9,900,000 51

52 How Changes in Throughput Affect Earnings % Change in Sales 5.00% 4.00% 3.00% 2.00% 1.00% 0.00% -1.00% -2.00% -3.00% -4.00% -5.00% Revenue Change Pre-tax After-tax $9,047,538 $5,880,900 $7,238,031 $4,704,720 $5,428,523 $3,528,540 $3,619,015 $2,352,360 $1,809,508 $1,176,180 $0 $0 -$1,809,508 -$1,176,180 -$3,619,015 -$2,352,360 -$5,428,523 -$3,528,540 -$7,238,031 -$4,704,720 -$9,047,538 -$5,880,900 Impact on Earnings Net Earnings % Change Actual ROE $15,780, % 17.53% $14,604, % 16.23% $13,428, % 14.92% $12,252, % 13.61% $11,076, % 12.31% $9,900, % 11.00% $8,723, % 9.69% $7,547, % 8.39% $6,371, % 7.08% $5,195, % 5.77% $4,019, % 4.47% 52

53 The Decoupling Calculation Utility Target Revenue Requirement determined with traditional rate case By class & by month (or other period coinciding with how often decoupling adjustment is made) Each future period will have different actual unit sales than Test Year The difference (positive or negative) is flowed through to customers by adjusting Price for that period (see Post Rate Case Calculation) Periodic Decoupling Calculation From the Rate Case Target Revenues $10,000,000 Test Year Unit Sales 100,000,000 Price Post Rate Case Calculation $0.10/Unit Actual Unit Sales 99,000,000 Target Revenues (from above) $10,000,000 Required Total Price Decoupling Price Adjustment $ /Unit $ /Unit 53

54 Markets Approach Success requires understanding of markets Who are the key market actors? What barriers do they face? What are their business models and needs? Portfolio of strategies mapped to barriers Cash, financing, info, tech support, training, etc. Mix will/should vary by market Relationships with trade allies is critical Articulate value in human and business terms Yes, it is about selling 54

55 Traditional Model of Markets: Everything fits into neat administrative boxes New Buildings Commercial & Industrial C&I NC Residential RNC Existing Buildings C&I Products C&I Retrofit Products Low Income HVAC Res Retrofit

56 Reality: Markets are Messy Commercial & Industrial Residential New Buildings Existing Buildings C&I Retrofit C&I NC C&I HVAC Products RNC Res Retrofit

57 Flexibility Answer to Messy Markets Obligated parties must be able to adapt to customers needs Make it easy for the customer Single point of contact for all their needs Common branding Don t let reporting rules or other admin needs dictate how obliged parties interact with customers Capture potential synergies across programs E.g., between residential and commercial E.g., between efficiency and customer-sited renewables 57

58 Quality Control is Essential Key to good EE program management Sampling inspections of EE work quality Processes for review of savings estimates Even more important to support for EE policy Bad new stories of poor quality work, non-existent savings, safety issues, etc. are difficult to overcome Even if political support is unchanged, participation can be hurt Solution 1: add QC requirements to savings obligations Solution 2: develop quality specs & certifications E.g., Building Performance Institute certifications required by most U.S. whole house retrofit programs 58

59 Evaluation, Measurement & Verification (1) EM&V is a very well-established science and art Evolved over three decades of analyzing EE programs Very large field of experts Grid operators in Northeastern US rely on its answers It can/should address both: Technical questions: how much was saved? And Attribution questions: how much was due to program? Note: North American programs spend about 2-4% of EE budgets on EM&V 59

60 Evaluation, Measurement & Verification (2) It needs to be central part of any savings obligation scheme To keep obligated parties honest To build political credibility To provide feedback on how to improve Need to balance precision and cost Deemed savings (or algorithms) for measures with small savings, predictability over many customers Simple M&V (e.g. short-term metering) for measures whose savings vary primarily on one key variable for large projects Full M&V for appropriate sample of large, complex projects to develop realization rates applied to full portfolio of such projects It should be managed independently 60

61 Evaluation, Measurement & Verification (3) Need annual savings verification process Ensure deemed savings used correctly Sample review of custom project savings calculations Sample on-site assessments for custom projects On-going evaluation studies To refine deemed savings values To assess free ridership (and spillover, if any) Both should be managed independently 61

62 The Regulatory Assistance Project (RAP) RAP is a global, non-profit team of experts providing technical and policy assistance to government officials on energy and environmental issues. Richard Cowart is the Director of European Programmes for RAP, based in Brussels. A member of the IEA DSM Executive Committee, he served 12 years as Chair of the Vermont PSB (utilities regulator), Chaired the National Council on Competition and the Electric Industry. Now also Chair of the US Department of Energy s Electricity Advisory Committee. Chris Neme is a Principal of Energy Futures Group. Since 1990 he has advised on efficiency policies and programs in more than 25 states and provinces across North America. He is an advisor to RAP in both the U.S. and Europe. Contacts : rcowart@raponline.org;