Estimating Retail Market Potential by hdv10836


Estimating Retail Market Potential document sample

More Info
									Estimating Large Customer Demand
    Response Market Potential:
              A Scoping Study

             Charles Goldman
E. O. Lawrence Berkeley National Laboratory

       Co-authors: N. Hopper, R. Bharvirkar
       B. Neenan and P. Cappers (Utilipoint)

               DRCC Webinar
               March 16, 2007

• Review of market potential concepts
  and methods
• Proposed methodology for
  estimating DR market potential of
  large customers for certain types of
  DR programs
• Data inputs: participation rates and
  elasticity values
• Market potential simulation results
• Recommendations
• Why estimate DR market potential?
   - To determine how much DR is available? from which
     market segments and DR options?
   - To develop the demand-side assessment section in a
     resource plan
   - To assist with planning or screening of potential
     demand response programs
   - To determine if DR goals are feasible
• Goals of this study:
   - Review methods and present conceptual framework for
     estimating DR market potential for large customers
   - Compile and publish available data inputs—
     disaggregated participation rates and elasticity values
   - Identify gaps in data and methodologies
     Technical Potential Less Meaningful for DR
                than Market Potential
• Analytic framework used in EE
  potential studies:
   - Technical potential—complete                                   potential
     penetration of all energy efficiency
     measures that were technically                                     potential
   - Economic potential—the subset of the
     technical potential that is cost-
     effective to implement
   - Market potential—‖achievable‖
     potential taking into account customer
     cost-effectiveness criteria,
     awareness, assumed levels of pgm
     incentives & activity
• Conceptual and practical issues in applying framework to DR:
   - EE assumes constant service/amenity while DR depends on customer
     willingness and behaviors to curtail/shift load and accept reduced
     service/amenity level
   - Policymakers are interested in market—or achievable—potential directly
Approaches Used to Estimate DR Market
• Customer surveys— ask customers about expected
  actions if offered hypothetical DR options; derive
  participation rates and expected load curtailments
• Benchmarking—use participation rates and load
  reductions observed among customers in other
• Engineering approach—bottom-up engineering
  techniques (similar to EE market potential studies);
  apply assumed participation and response rates to data
  on local customers, loads or equipment stock
• Elasticity approach—estimate price elasticities from
  customers exposed to DR programs or dynamic pricing
  tariffs and apply them (after determining expected
  participation levels) to estimate load impacts
 What Makes DR Different from EE?
• The nature of participation:
    - For DR, participation involves two steps:
           enrolling in a program or tariff
           providing load reductions during specific events (e.g., system
            emergencies or periods of high prices)
    - For EE, ―participation‖ is a one-time decision to invest in
      energy-efficiency measures or equipment.
• The drivers of benefits:
    - DR benefits hinge on customer behavior in response to hourly
      prices, financial incentives, and/or system emergencies
    - EE cost savings largely depend on technical characteristics
      and performance of installed equipment or measures
• The time horizon and valuation of benefits:
    - From a customer perspective, DR benefit streams may be
      highly variable, and depend on short-term price fluctuations or
      emergency curtailment incentives
    - EE investments provide a multi-year stream of cost savings
      that the customer can value at expected retail energy rates
      Five Steps to Estimating DR Market
• 1) Establish study scope—identify target population and
  types of DR options considered
• 2) Customer segmentation—identify customer market
• 3) Estimate net program penetration rates—use available
  data to estimate customer enrollment in voluntary
  programs and exposure to default pricing programs
• 4) Estimate price response—develop elasticity estimates
  for various DR options, customer market segments, and
  factors found to influence price response
• 5) Estimate load impacts —use info from steps 2 to 4 to
  estimate the amount of DR that can be expected from the
  target customer population at utility at reference price (or
  incentive level).
   Selecting a Measure of Price Elasticity

• Most studies of large
  customer price
  response estimate
  substitution or arc
• Substitution elasticity:
    - change in peak: off-
      peak usage in
      response to a 1%
      change in peak: off-
      peak prices
    - requires several
      observations per
• Arc elasticity:
    - % change in usage / percent change in peak price
    - Very easy to estimate, but limited explanatory power
                     DR Options Evaluated
DR Option        Description
Optional         Dynamic pricing tariff with bundled charges for delivery and commodity
Hourly Pricing   Offered on an optional basis by vertically-integrated utilities (usually)
                 Typical rate design: two-part tariff with customer baseline load (CBL)
Default Hourly   Dynamic pricing tariff with unbundled distribution and commodity charges
Pricing          Offered as default service tariff in states with retail electric competition
                 Commodity component: Hourly price indexed to a wholesale energy
                 market (day-ahead or real-time)
Short-Notice     DR program that offers financial incentives for curtailing load when called
Emergency        on short notice (i.e., 1-2 hours) in response to system emergencies
Program          Customer response is voluntary; no penalties for non-performance
Price-           DR program that pays for measured load reductions when day-ahead
Response         wholesale market prices exceed a floor
Event            Some programs may include bid requirements and/or penalties
Critical-Peak    Dynamic-pricing tariff similar to a time-of-use rate, except that on ―critical-peak‖ days
Pricing          a pre-specified higher price is effective for a specific time period
                                 Data Sources

DR Option            Data Source(s)                                         Eligible Customers
                                                                            (peak demand)
Optional Hourly      Central and Southwest (CSW) Utilities’ (now            > 1500 kW
Pricing              American Electric Power) two-part RTP rate

Default Hourly       Niagara Mohawk Power Corporation (NMPC),               > 2000 kW
Pricing              a National Grid Company, SC-3A tariff

Short-Notice         NYISO Emergency Demand Response                        > 100 kW
Emergency            Program (EDRP)
Program              ISO-NE Real-Time Demand Response (RTDR)
                     Program                                                > 100 kW

Price-Response       ISO-NE Real-Time Price Response (RTPR)                 > 100 kW
Event Program        Program

Critical-Peak        California Utilities1 Critical Peak Pricing            > 200 kW;
Pricing              Program
                                                                            > 100 kW (SDG&E)
1Pacific Gas & Electric (PG&E), Southern California Edison (SCE) and San Diego Gas & Electric
        Participation Rates: Selected Values

DR Option        Market Segment      Customer Size (peak demand)
                                                                   • DR participation =
                                     0.5–1 MW          >2 MW
                                                                       - enrollment in
Optional         Commercial/retail       0%              2%
hourly pricing
                 Gov’t/education         4%             25%              programs/tariffs
                 Manufacturing           5%             25%
                                                                       - Or remained on
Default hourly   Commercial/retail      11%             43%              default RTP tariff
                 Gov’t/education        10%             42%
                                                                   • Participation rates
                 Manufacturing           8%             33%
                                                                     collected for 5
Short-notice     Commercial/retail      23%             20%
                                                                     market segments
                 Gov’t/education         5%              9%          and 4 customer
                 Manufacturing          15%             23%          size groups
                 Commercial/retail       1%              6%
                                                                   • Some data were
                 Gov’t/education         3%             10%
                                                                     not available—
program          Manufacturing          10%             30%          used ―expert
Critical-peak    Commercial/retail       3%              4%          judgment‖ (red-
                 Gov’t/education         4%              2%          italicized values)
                 Manufacturing           4%              7%
                 Average Elasticity Values

Customer                                        Demand Response Option
Market          Optional         Default            Short-Notice      Price-          Critical-
Segment         Hourly           Hourly             Emergency         Response        Peak
                Pricing          Pricing            Program           Event Program   Pricing

Commercial/     0.01             0.06               -0.03             -0.09           -0.10
Government/     0.01             0.10               -0.02             -0.16           -0.06
Healthcare      0.01             0.04               -0.04             -0.05           -0.01
Manufacturing   0.26             0.16               -0.04             -0.16           -0.05
Public works    0.07             0.02               -0.08             -0.22           -0.08

                   elasticity of substitution                      arc elasticity
 Putting it all together: Estimating Market
      Potential at utility in Northeast
• Simulation: Demonstrate approach by estimating DR
  market potential at a Northeast utility
• Utility service territory characteristics:
   - Relatively small, urban utility
   - Peak demand of large non-residential customers is ~1700 MW
     (40% of utility’s peak demand)
   - Mostly commercial/retail, gov’t/education and healthcare facilities
   - Fewer manufacturing facilities than most suburban or rural areas
• Load impacts estimated assuming that RTP peak and
  DR ―event‖ prices were $500/MWh
• Estimated DR market potential under several scenarios
  to illustrate effects of various factors (e.g., high
  participation rates, customer response at high prices)
Large-customer DR Market Potential at a NE
        Utility: Base Case Results
                                                                     Note: Program results are not additive.

                                                                           350-500 kW
 Demand Reduction

                    40                                      2%             500-1000 kW
                                                                           1-2000 kW

                                                                           > 2000 kW

                                                                      0% of class peak demand

                         Optional   Default   Short-      Price      Critical
                         Hourly     Hourly    notice      Response   -peak
                         Pricing    Pricing   Emergency   Event      Pricing
                                              Program     Program
• Base case results indicate market potential of up to 3% of class-peak demand
  for each DR option individually
• Largest customers (> 2 MW) provide bulk of load response
• Implicit Implication: necessary to target smaller customers too
                           Impact of Program Participation Rates
                                  on DR Market Potential
(% of non-coincident class demand)

                                                     Note: Program results are not additive.

                                                           base participation rates
        Demand Reduction

                                                           doubled participation
                                                           tripled participation



                                          Optional      Default         Short-             Price      Critical
                                          Hourly        Hourly          notice             Response   -peak
                                          Pricing       Pricing         Emergency          Event      Pricing
                                                                        Program            Program

     • With very aggressive marketing (i.e., 3x current participation rates),
       market potential for 3 DR options increases to ~3 to 7% of eligible
       large customers’ peak demand at this utility (or ~75-165 MW)
            Sensitivity Analysis: Accounting for
                 Response at High Prices
                            80                  4%                     Note: Program results are not additive.
                                 Base                     High-Price
                                                                               350-500 kW
                                 Case                     Case
         Demand Reduction

                                                                               500-1000 kW
                            60          3%                                     1-2000 kW
                                                                               > 2000 kW

                            40                                    2%

                                                                          0.4% of class peak demand
                                         Default Hourly            Price Response
                                         Pricing                   Event Program

•   Base-case elasticity estimates based on response over a range of prices
•   ―High-price‖ case refines elasticity estimates to reflect response at higher prices
     -       Default hourly pricing—higher substitution elasticities at higher prices
     -       Price response event program—removing arc elasticity observations based on lower
             prices produces lower average estimates because customer response (% change in
             load) is similar but the price differential (the denominator) is greater  illustrates
             limitation of arc elasticities
                Summary of Findings:
• Simulation exercise provides ―reasonable range‖ of DR
  market potential values of large non-residential customers
  for DR options
   - Note: load impact results are not additive
• Other analysts can use elasticity values (and participation
  rates) as starting point for market assessments for these
  five DR options
• Customer participation rates have largest impact on DR
  market potential estimates; but represent the largest data
   - Need for more info on eligible target population
   - Drivers for participation
• Important to refine and disaggregate elasticity estimates
  for different groups of customers
    A Market Assessment Research Agenda
•        Need to develop broader information base on customer participation
         and price responsiveness for more robust DR Market Assessments
•        1) Link Program Evaluation to Market Potential Studies:
     -      Evaluations of DR programs should systematically collect data on:
               customer characteristics,
               hourly loads and prices,
               drivers of customer participation and response, and
               size and characteristics of the target/eligible population

•        2) Program Participation:
     -      Develop predictive methods for estimating participation rates in DR
            programs and dynamic pricing tariffs that incorporate customer
            characteristics and other factors that drive participation.
•        3) Price Response:
     -      Estimate price elasticity values for different market segments,
            accounting for the relative impact of driving factors. Where possible,
            estimate demand or substitution elasticities rather than arc elasticities.
            Recommendations (cont):

•   4) Assess the Impacts of Demand Response Enabling
    -   Document the impacts of specific DR enabling technologies
        on customer participation and load response (given limited
        evidence and mixed results from existing evaluations)
    -   At a minimum, gather information on onsite generation, peak
        load controls, EMCS and EIS.
•   5) Publicize Results:
    -   Current situation: Data reporting and methods are not
        standardized; disaggregated results are often not transparent
    -   Need to explore ways to pool customer-level data, while
        protecting customer confidentiality.
LBNL Reports on DR Market Potential

• ―Estimating Demand Response Market Potential among
Large Commercial and Industrial Customers: A Scoping
   - Charles Goldman, Nicole Hopper, Ranjit Bharvirkar (LBNL),
   Bernie Neenan, and Peter Cappers (Utilipoint), LBNL-61498,
   January 2007.

                    Reports available at:

To top