Steve McCarty, Director, Demand Response, PG

1
Steve McCarty, Director, Demand Response,
PGESteve George, Principal Consultant, FSC
Enrollment Forecasting for Pacific Gas and
Electric Companys SmartRate and SmartAC
Programs
National Town Meeting on Demand Response and
Smart Grid
2
Overview

• Understanding average load impacts from dynamic
  rates is important, but so is predicting how many
  customers will sign up
• PGE began marketing a critical peak pricing
  tariff, SmartRate, in 2008 to residential and
  small commercial customers who had SmartMeters
• For resource planning purposes, we needed to
  estimate how many customers will sign up for
  dynamic tariffs between now and 2020
• PGE also has an air conditioning cycling
  program, SmartAC
• There was also a need to predict enrollment for
  this program and, importantly, to predict dual
  enrollment in the two programs

3
What is SmartRate?

• SmartRate is an overlay on existing PGE tariffs
  (a five tier, increasing block rate) that prices
  electricity during events significantly higher
  than the otherwise applicable price on up to 15
  days, known as SmartDays
• During events, residential customers pay an
  incremental 60 /kWh
• The event window is fixed from 2-7 pm for
  residential customers
• Event days can be called from May 1st through
  October 31st and notification is provided by 3 pm
  the day prior to an event
• During non-event days from June 1st to September
  30th, customers receive a rate discount
• A credit of 3 /kWh applies to all usage other
  than peak-period usage on SmartDays
• An additional credit of 1 /kWh, applies to all
  usage 200 or more above the baseline kWh
• In 2008, SmartRate was only offered to customers
  on non-TOU, so prices varied by time of day only
  on Smart Days, not on other days
• Low Income customers who are eligible to receive
  lower rates for electricity use (CARE)? can also
  sign up for SmartRate

?California Alternative Rates for Energy
4
SmartRate Marketing and Enrollment

• In 2008, SmartRate was marketed to approximately
  129,000 households in the Bakersfield region (a
  hot climate with high air conditioning
  saturation)
• Central air conditioning saturation in the area
  is roughly 70 percent
• Approximately 71 percent of SmartRate CARE
  participants had central air conditioning
• 92 percent of non-CARE SmartRate participants had
  central air conditioning
• Over 10,000 households (7.5) accepted the offer,
  most of whom received a single direct mail
  promotional piece
• A 50 Visa gift card was offered to residential
  customers who signed up early
• Customers were offered first-year bill protection
• Enrollment grew significantly over the 2008
  summer, from roughly 4,000 in early July to
  almost 10,000 in early September
• Customer enrollment significantly exceeded
  expectations and certainly would have been higher
  had PGE completed a more traditional,
  multi-contact marketing plan

5
Low Income Customers Signed Up At a Higher Rate,
but Responded Less than Other Customers

• A disproportionate number of CARE customers
  enrolled in SmartRate
• Approximately 35 percent of Bakersfield
  residential customers who were sent marketing
  materials were CARE customers
• 56 percent of customers who enrolled in SmartRate
  in 2008 were CARE customers
• The average load reduction across 9 events was
• 23 for Non-CARE
• 11 for CARE customers

6
What is SmartAC?

• PGEs SmartAC program is a reliability-driven
  air conditioning load control program
• The SmartAC program deploys both programmable
  communicating thermostats (PCTs) and direct load
  control switches
• Events can be called anytime between May 1st and
  October 31st for an event period not to exceed 6
  hours or more than 100 hours for the season
• PGE began marketing SmartAC in 2007
• Most participants are paid a one-time incentive
  of 25
• Customers are primarily motivated by the desire
  to help keep the lights on during an emergency
  situation and help the environment
• As of 6/1/2009, approximately 118,000 customers
  had enrolled in the program

7
PGE Needed to Estimate Enrollment for 10 Years

• Four relevant decisions had to be modeled
• SmartRate for customers that are not on SmartAC
• SmartRate for customers that were already
  enrolled in SmartAC
• SmartAC enrollment for customers not on SmartRate
• SmartAC enrollment for customers that were
  already on SmartRate
• The first and third decisions could be estimated
  using actual choice data
• That is, data from choices made by customers who
  had actually been offered these options and
  decided whether or not to enroll
• The other two decisions could not be modeled
  based on actual choice data because such
  offerings had not yet been made
• FSC conducted a stated preference survey that
  asked customers already enrolled in one program
  whether they would consider enrolling in the
  other if it were offered

8
Key Drivers of SmartRate Enrollment Decisions

• CARE customers are significantly more likely to
  enroll in SmartRate, and in fact this effect is
  stronger than any other single driver of
  enrollment
• The choice between a pure CPP and a CPP-TOU
  version of SmartRate appears to have no effect on
  the decision to enroll, although the actual rate
  to be offered in the future will have a
  substantially higher peak period price than the
  one presented in the survey have a TOU underlying
  rate
• Demographic factors play a relatively limited
  role in driving enrollment
• Older households are more likely to enroll in
  SmartRate
• Larger households are less likely to enroll in
  SmartRate
• The current estimation approach is based in
  grouped data. Further analysis using individual
  data may identify more relevant demographic
  factors

9
Key Drivers of SmartAC Enrollment Decisions

• Prior participation in EE programs is a strong
  driver of SmartAC enrollment
• The heavier the use of the AC, the greater the
  likelihood of participation
• CARE status is another important enrollment
  driver for SmartAC
• The percent of homes with children under 17 and
  heads of households aged 65 or older both had
  positive correlations with SmartAC enrollment

10
Dual Enrollment in SmartRate and SmartAC

• Customers who are already enrolled in SmartAC are
  roughly twice as likely to accept SmartRate as
  are customers that are not enrolled in SmartAC
  (next slide)
• Customers who are already enrolled in SmartRate
  have an even higher acceptance rate for SmartAC
  than do SmartAC customers who are offered
  SmartRate (next slide)
• That is, customers with enabling technology have
  a much higher likelihood of accepting a dynamic,
  time-varying rate than do customers without
  enabling technology, and customers on a dynamic,
  time-varying rate also have a much higher
  probability of accepting an enabling technology
  than do customers who are not on a dynamic,
  time-varying rate
• Following the stated preference survey analysis,
  PGE offered SmartAC to a sample of SmartRate
  customers and the take rate was 14 percent, which
  is close to the 16 percent estimate through the
  stated preference analysis

11
Stated Preference AnalysisEnrollment in
Multiple Programs
12
For questions, please contact

• Steve McCarty
• Director, Demand Response
• PGE
• SJM8_at_PGE.COM
• Stephen S. George, Ph.D.
• Principal Consultant, Director Energy Practice
• Freeman, Sullivan Co.
• 415 948-2328
• stephengeorge_at_fscgroup.com