Paying Upfront: A Review of Salt River Project’s MPower Prepaid Program
1020260
Paying Upfront: A Review of Salt River Project’s M-Power
Prepaid Program
1020260
Technical Update, October 2010
EPRI Project Manager
B. Neenan
ELECTRIC POWER RESEARCH INSTITUTE
3420 Hillview Avenue, Palo Alto, California 94304-1338 ▪ PO Box 10412, Palo Alto, California 94303-0813 ▪ USA
800.313.3774 ▪ 650.855.2121 ▪ askepri@epri.com ▪ www.epri.com
DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN ACCOUNT OF
WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH INSTITUTE, INC. (EPRI).
NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE ORGANIZATION(S) BELOW, NOR ANY
PERSON ACTING ON BEHALF OF ANY OF THEM:
(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I) WITH
RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM
DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS FOR A PARTICULAR
PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR INTERFERE WITH PRIVATELY OWNED
RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL PROPERTY, OR (III) THAT THIS DOCUMENT IS
SUITABLE TO ANY PARTICULAR USER'S CIRCUMSTANCE; OR
(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER (INCLUDING
ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE HAS BEEN ADVISED
OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THIS
DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN
THIS DOCUMENT.
THE FOLLOWING ORGANIZATION(S), UNDER CONTRACT TO EPRI, PREPARED THIS REPORT:
Electric Power Research Institute (EPRI)
This is an EPRI Technical Update report. A Technical Update report is intended as an informal report of
continuing research, a meeting, or a topical study. It is not a final EPRI technical report.
NOTE
For further information about EPRI, call the EPRI Customer Assistance Center at 800.313.3774 or
e-mail askepri@epri.com.
Electric Power Research Institute, EPRI, and TOGETHER…SHAPING THE FUTURE OF ELECTRICITY
are registered service marks of the Electric Power Research Institute, Inc.
Copyright © 2010 Electric Power Research Institute, Inc. All rights reserved.
ACKNOWLEDGMENTS
The following organization(s), under contract to the Electric Power Research Institute (EPRI),
prepared this report:
EPRI
3420 Hillview Avenue
Palo Alto, CA 94304
Principal Investigators
B. Neenan
J. Robinson
This report describes research sponsored by EPRI.
EPRI would like to thank the many SRP employees who contributed time and resources to the
development of this report, including:
Marsha Caplan
Loren Kirkeide
Jen Collins
Mike Lowe
Lisa Day
Mike McGinnis
Kelly Drow
Betty Pruitt
Cathy den Dulk
Susan Pyle
Dena Emary
Karen Smith
Erin Erben
Susan Smith
Mike Fish
Bill Twardy
Debbie Kimberly
Carrie Young
Jennie King
This publication is a corporate document that should be cited in the literature in the following
manner:
Paying Upfront: A Review of Salt River Project’s M-Power Prepaid Program. EPRI, Palo Alto,
CA: 2010. 1020260.
iii
ABSTRACT
Arizona’s Salt River Project (SRP) has operated M-Power, the largest electricity prepayment
program in the United States, since 1993. The customer population has grown to about 100,000
(approximately 12% of all residences served by SRP), and it has expanded from the initial target
population—consumers with arrears facing service terminations and low-income customers—to
include consumers with different expectations from M-Power service. The in-home portion of
the SRP prepay configuration consists of a user display terminal (UDT) that communicates with
the customer’s meter. The purchasing component of the M-Power program is the self-service
kiosk, known as a PayCenter, accessed via a Smart Card, which is also the conduit through
which electricity consumption information is transferred back to SRP.
The constant aspects of the M-Power experience have been a high level of customer satisfaction
and an overall conservation effect reported by SRP of approximately 12%. SRP attributes the
conservation effect to a variety of factors, noting that M-Power requires consumers to pay
attention to when and how they use electricity, allowing them to make immediate adjustments in
usage to lower their bills.
This report provides an overview of how the M-Power program works along with an
examination of the technology, systems, and costs associated with the program. The overview is
followed by an analysis of customer perceptions of the program as well as a discussion of the
program’s potential conservation effect. The report concludes with a discussion of impact studies
needed to answer several outstanding research questions, including the effect of various types of
payment options on conservation as well as whether SRP’s experience is transferrable to other
markets, climates, customer circumstances, and supply conditions.
Keywords
Salt River Project (SRP)
M-Power Prepaid Metering Program
Energy Efficiency
Energy Consumption Information
Consumer Behavior
Residential
Consumers
v
CONTENTS
1 INTRODUCTION ....................................................................................................................1-1
The M-Power Story ..............................................................................................................1-1
Lessons Learned—30,000 Feet.....................................................................................1-3
2 THE M-POWER PROGRAM—AN OVERVIEW .....................................................................2-1
Initial Set-Up.........................................................................................................................2-1
Purchasing Electricity...........................................................................................................2-2
Other M-Power Features......................................................................................................2-3
Back-Office Processes.........................................................................................................2-4
3 M-POWER TECHNOLOGY, SYSTEMS, AND COSTS..........................................................3-1
Technology...........................................................................................................................3-1
User Display Terminal (UDT) or In-home Display (IHD) ................................................3-1
Meter ..............................................................................................................................3-2
PayCenters (Self-service Kiosks)...................................................................................3-3
M-Power Business Systems ................................................................................................3-3
Customer Information and Back Office Systems............................................................3-3
Revenue Reporting ........................................................................................................3-5
M-Power Costs.....................................................................................................................3-5
Customer Costs .............................................................................................................3-5
SRP Cost-Benefit of M-Power........................................................................................3-8
4 THE CUSTOMER EXPERIENCE ...........................................................................................4-1
Past Study Overviews and Findings ....................................................................................4-1
Overall Satisfaction Levels .............................................................................................4-2
Perceived Customer Benefits of M-Power .....................................................................4-3
Perceived Disadvantages ..............................................................................................4-5
M-Power Customer Characteristics ...............................................................................4-6
Customer Purchase Behavior ........................................................................................4-7
Customer Retention .......................................................................................................4-7
5 INFLUENCE OF M-POWER ON ELECTRICITY USAGE ......................................................5-1
Many Reasons for a Conservation Effect.............................................................................5-1
Managing What You can Measure.................................................................................5-2
Deposit Avoidance .........................................................................................................5-3
Accommodating Particular Circumstances ....................................................................5-3
Pay-as-You Go Household Accounting..........................................................................5-3
Arrears Financing Through a Rate Differential...............................................................5-3
Conservation Ethic .........................................................................................................5-5
Summary of Potential Influences ...................................................................................5-5
vii
SRP Impact Assessment .....................................................................................................5-6
Characterizing the Conservation Effect..........................................................................5-6
Research Design............................................................................................................5-8
Measuring the Potential of M-Power ............................................................................5-10
6 CONCLUSIONS AND RECOMMENDATIONS ......................................................................6-1
A REFERENCES...................................................................................................................... A-1
B M-POWER BACK OFFICE SOFTWARE SCREEN SHOTS ................................................ B-1
C MARKET RESEARCH STUDY DETAILS ............................................................................ C-1
viii
LIST OF FIGURES
Figure 1-1 M-Power Program Chronology .................................................................................1-4
Figure 2-1 M-Power Overview (Source: Pyle, 2009) .................................................................2-1
Figure 2-2 Example of Customer Credit Purchase Receipts (Left: no arrears, Right: arrears)..2-3
Figure 3-1 M-Power Meter, UDT, and Two Smart Cards...........................................................3-1
Figure 3-2 PayCenter Interfaces (Source: SRP, 2009)..............................................................3-4
Figure 5-1 Example Price Elasticity ...........................................................................................5-5
Figure 5-2 M-Power Program Size ............................................................................................5-8
Figure 5-3 2005/2006 M-Power Impact Analysis Timeline ......................................................5-10
Figure B-1 Account Details Screens ......................................................................................... B-1
Figure B-2 Arrears Details Screens .......................................................................................... B-2
Figure B-3 Transaction History Details ..................................................................................... B-3
Figure B-4 Meter Read Data..................................................................................................... B-3
Figure B-5 Meter Credit Status ................................................................................................. B-4
Figure B-6 Self-Disconnection History ...................................................................................... B-4
ix
LIST OF TABLES
Table 3-1 UDT Features ............................................................................................................3-2
Table 3-2 Meter Features ..........................................................................................................3-2
Table 3-3 PayCenter Features...................................................................................................3-3
Table 3-4 M-Power per-kWh Charges (Effective November 2009)............................................3-7
Table 3-5 M-Power per-kWh Charges (Effective November 2009)............................................3-8
Table 4-1 Past M-Power Market Research Studies ...................................................................4-1
Table 4-2 SRP Ratings: Comparing M-Power and Non-M-Power Customers (FY2010)...........4-3
Table 4-3 M-Power Customer Demographic Trends .................................................................4-6
Table 4-4 Electricity Purchase Amount and Frequency Information ..........................................4-7
Table 5-1 Potential Influences of the M-Power Program on Consumption ................................5-6
Table C-1 M-Power Market Research-related Studies Performed............................................ C-1
xi
1
INTRODUCTION
As the largest electricity prepayment program in the United States with over 100,000 customers
at the time of publication, Salt River Project’s award-winning1 M-Power program provides a
potential wealth of experience for other utilities who are considering their own prepayment
programs.
SRP personnel respond to frequent inquiries from other utilities about their program, and given
that it combines a form of electricity-use feedback with a unique payment approach, the newly
reinvigorated interest in behavior change-inducing feedback programs are also spurring interest
in the program. Indeed, it was in response to this widespread industry interest that the idea for
this report emerged. Its purpose is to capture in one report the details of the M-Power program
that would be of use to other utilities considering similar endeavors. This includes an overview
of how the M-Power program works (Section 2), the technology, systems, and costs associated
with the program (Section 3), an examination of customer perceptions of the program (Section
4), and finally, an examination of impact analysis studies which have reported a significant
conservation effect attributable to the program, and a discussion of additional impact studies that
would be of use going forward to answer some outstanding research questions (Section 5).
To begin, some historical context will now be provided as a foundation for the report, including
the impetus for the M-Power program and the chronological details of its inception, as outlined
in Figure 1-1.
The M-Power Story2
The M-Power program began in 1993 when the Arizona state legislature proposed the
development of programs aimed to assist low income consumers with bill payment. At the time,
SRP sought input from various community organizations to learn what programs, in addition to
federally funded weatherization programs, could be implemented to aid low income customers in
saving energy—prepayment was one component of the pilot that was developed.
The prepay component began with a 100-home pilot, and one of the key findings was resounding
customer support for the concept, which came as somewhat of a surprise. SRP found that
customers felt that, for the first time, they were in control of their electricity bill, not SRP. After
the pilot completion, the program expanded to other regions of SRP’s service territory, and was
no longer limited to low income customers.
1
The M-Power program won the National Energy Resources Organization (NERO) first place award for energy
efficiency. NERO is a non-profit organization that recognizes organizations active in the promotion of energy
efficiency.
2
The content for this section is based on Personal communication, Mike Lowe, Customer Services Manager, SRP,
September 17, 2009; Traasdahl (2009); and Personal communication, Karen Smith, Measurement and Evaluation
Manager, May 29, 2010 (forwarded email from Bonnie Temme).
1-1
From the beginning, the in-home portion of the SRP prepay configuration consisted of an inhome display, referred to at SRP as an User Display Terminal (UDT), which communicates with
the customer’s meter; the initial vendor of the meter and UDT was CIC Global. As with any new
program there were bugs, and in the early days, the technical difficulties made it necessary to
hard wire the meter to the in-home UDT, which was an expensive configuration. Nonetheless, it
was in this configuration that enrollment in the program grew to 1,600.
In 1999, Motorola became the provider of both the UDT and meter, and the new configuration
employed power line carrier (PLC) communication between the two. Although this configuration
was an improvement, there were still communication issues in some instances when other
devices in the home prevented communication between the UDT and the meter (e.g., lamps, pool
pumps).
Motorola exited the business in 2002, although SRP had purchased the rights to manufacture
both the meters and the UDTs. AMPY Metering Limited, a UK-based company that is now part
of Landys+Gyr, was the new vendor that was eventually selected.3 In 2005 the AMPY
technology was rolled out to M-Power customers, and a dual frequency approach was employed
that solved the PLC performance issues. AMPY (now Landis+Gyr UK, Ltd) remains the
provider of both the meter and UDT today.
Another technology component of the M-Power program is the self-service kiosk, known as a
PayCenter. These are similar in nature to automated teller machines (ATMs) used by banks, and
are located throughout the Phoenix area. The idea for the PayCenters was proposed early on as a
cost-effective way of dealing with high volumes of power purchases—M-Power customers
generally buy small quantities of power, but relatively frequently (see Section 4 for more
details). Customers purchase electricity at the PayCenters through the use of a card, known as a
Smart Card. The card is then inserted into the UDT at the home. As will be explained in more
detail below, the Smart Card is also the conduit through which electricity consumption
information is transferred back to SRP.
In the late-1990s, the first PayCenters were purchased from Diebold Incorporated, which at the
time was working with Duke Power on payment machine applications for banked customers. The
PayCenters were initially purchased for bill-paying customers, but SRP worked with Diebold to
retrofit them to accept the M-Power Smart Cards. This configuration worked well, and enabled
SRP to reduce costs by eliminating over-the-counter customer transactions altogether.
This arrangement continued until Diebold exited the business in 2000, at which time SRP
contracted with AllKiosk, a division of GECO Incorporated, an Arizona-based engineering
house, to develop their own PayCenters. They also used this opportunity to address the issues
they had had with the Diebold design, and the result was a less costly and more reliable system.
AlKiosk is now licensed to sell the PayCenter design to others as well.
As of April 2010, there were over 100,000 customers enrolled in the M-Power program, with 95
PayCenters in the SRP customer territory.
3
Prepayment systems are prevalent in the UK, and much of their early technology was coin operated. However,
rising electricity prices in the 1980s led to increased instances of home break-ins to steal the coins, and this
precipitated the invention of an electronic version of prepay.
1-2
Lessons Learned—30,000 Feet
As with any new program, mistakes were made. Likely the biggest one relates to fact that SRP
tried for years to treat the M-Power program like any other rate program, by forcing the back
office systems to “make” monthly bills for revenue recognition purposes. Utilities and their
employees are used to thinking in terms of a billing cycle, and so it was initially difficult to think
of the M-Power program using a different mindset. It took quite a bit to change this mindset, and
it has just been since 2007 that the system has changed to cash-based revenue recognition, rather
than trying to force M-Power program accounts into the standard billing cycle.
Another lesson learned early on was that customers on the M-Power program buy power
frequently. Indeed, the utility industry has forced customers to pay on a monthly basis when in
fact most customers are paid weekly or bi-weekly. On average, M-Power customers purchase
power in the $20 range approximately four times a month in the winter, and seven times a month
in the summer. Vending also peaks on Friday nights, likely coinciding with payday for many
customers.
From a customer experience perspective, another lesson learned is to work in partnership with
potential program critics to get to the root of any concerns. SRP worked directly with customer
advocacy groups who were initially opposed to the idea of a prepay program due to concerns that
it might unfairly force low income customers into power outage situations. SRP partnered with
one such group, the Arizona Community Action Association, to design market research to assess
their concerns. The ACAA is now an advocate of the M-Power program.
As technology continues to evolve, so likely will the M-Power program. Having considered the
historical context of the program, the following section will now provide an overview of how MPower works in its present-day configuration.
1-3
Initial meters and displays
(CIC Global); needed to
hardwire display to meter,
which was expensive.
Partnered with
Motorola in 1999
using PLC
communication.
Partnered with
Diebold for
PayCenter
production.
1993
100
1994
…
Motorola
exited
business.
Contracted
AllKiosk to
design and
manufacture
PayCenters.
AMPY meters
and UDTs
introduced.
First AllKiosk
PayCenters
installed.
Diebold
existed
business.
1999
2000
2001
2002
2005
2006
2007
2008
2009
2010
1996
1997
1998
48
589
1,123
1,742 1,918
1,613 4,706 17,122 29,267 31,252 30,282 40,899 52,345 41,481 73,440 100,000+
M-Power Subscription Rates (pre-2001, known as “Pay As You Go”)
First prepay program, 100
meters. High levels of
customer satisfaction
prompted pilot
continuation and growth.
2003
2004
1995
AMPY back office
system and customer
information system
interface complete in FY
08.
Figure 1-1
M-Power Program Chronology
1-4
2
THE M-POWER PROGRAM—AN OVERVIEW
This section builds on the previous inception story by providing an overview of the basic system
functionality in its current configuration. Figure 2-1 provides a high-level illustration of system
components.
Figure 2-1
M-Power Overview (Source: Pyle, 2009)
A customer can learn about the M-Power program through a number of channels—SRP markets
M-Power through direct mail, bill inserts, via email and through their web page, and through
Customer Service Representative scripting upon customer call in. The general process by which
a customer becomes an M-Power account, and the subsequent business process at SRP, is
outlined as follows:
Initial Set-Up
•
•
Customer calls SRP and asks to be on M-Power
The customer pays a $99 deposit for the UDT, $87.50 of which is refundable upon the return
of the device (this is in comparison to the $275 deposit that is required to set up a standard
SRP residential account). The customer also pays a $28 (plus tax) service establishment fee.
2-1
•
•
•
•
•
•
SRP back office will write two Smart Cards for the customer, one of which will be loaded
with a $30 credit.
The credit, equipment deposit, and service establishment fee, as well as any other arrears a
customer may have from their previous account, is transferred to the prepay account, so that
they can be automatically paid off over time at a rate of 40% each time the customer makes
an electricity purchase.
A field order will be created to deliver the meter and UDT to the customer; the meter and the
UDT need to be married.
The meter is installed by an SRP technician, but the UDT and a Start-Up Packet (containing
a UDT User’s Manua,l4 a Quick Start Guide,5 and a map6 listing all the PayCenter locations
in Phoenix) are left with the customer—the technician does not enter the home. No
conservation tips are provided with the Start-Up Packet, although tips are available on the
SRP website, as well as the M-Power microsite (www.mysrpmpower.com) and various other
communications pieces.
The UDT is plugged in, the card loaded with the $30 credit is inserted into it, and the UDT
and meter are “married” (a one-time event); the system is then ready to go.
When the card is inserted into the UDT, the full amount of credit that was on the card is
transferred through the UDT to the meter, and the card balance goes to zero. At the same
time, the meter consumption level at that time is transferred to the card, so that it may be
transferred to SRP via the PayCenter the next time electricity is purchased.
Purchasing Electricity
•
•
•
•
•
•
Electricity is purchased at any of the 95 automated PayCenters located throughout Phoenix
(these PayCenters can also be used to pay non-M-Power customers’ bills)
PayCenters are located in grocery and convenience stores, some of which are accessible 24
hours a day, seven days a week. These PayCenters can be good for store business as well, as
SRP’s market research findings indicate that many customers report also purchasing store
goods at the time of their electricity purchase.
PayCenters take cash and “e-Chex”, an electronic checking service.
Phone purchases can be made using credit cards and e-Chex, although credit cards can take
up to three days to post to an account, and the customer must still have the credits transferred
to their card by visiting a PayCenter and inserting $1 (the minimum amount).
Although rare, some customers pay by mailed check. When these checks are received by
SRP they are input into the back office system as credit.
The customer inserts one of their Smart Cards, and makes a payment (as low as $1, as high as
$2,001). The customer is then issued a receipt that outlines the payment made, and any
arrears deducted, and the total electricity purchased. See Figure 2-2 for receipt examples.
4
http://www.mysrpmpower.com/pdfx/MPowerUserManual.pdf
5
http://www.mysrpmpower.com/pdfx/quickstartguide.pdf
6
http://myaccount.srpnet.com/paymentlocations/
2-2
•
SRP occasionally uses the PayCenters to broadcast information to customers (e.g.,
notification of a planned outage)
Figure 2-2
Example of Customer Credit Purchase Receipts (Left: no arrears, Right: arrears)
Other M-Power Features
•
•
•
•
Friendly Credit: The system can be programmed such that a customer can have power even if
their credit goes negative during certain periods of time—in SRP’s case this is between the
hours of 6 p.m. and 6 a.m., seven days a week. The customer must purchase more electricity
by 6:00 a.m. the following day to avoid a disconnect. The friendly credit used will be added
to the arrears balance, to be paid off gradually with each subsequent purchase transaction.
Disconnects: When a customer’s account goes negative outside the Friendly Credit hours, the
meter will disconnect the power source to the home. To reconnect the power, the customer
must purchase credit at a PayCenter (or call to invoke Emergency Credit – see below). Once
credits are purchased and the card is inserted into the UDT, the customer’s power
automatically reconnects. There is no disconnection/reconnection fee involved (for a
standard residential account, the disconnection/reconnection fee is in the $60-$100 range,
and requires a utility truck-roll).
Emergency Credit: SRP may grant emergency credit over the phone by inputting a code to
reconnect the customer’s meter. This feature was added to address concerns regarding the
loss of electricity in situations where a shutdown of the overall SRP communications
network prevented customers from purchasing electricity.
M-Power customers receive no monthly bill and pay no late fees. They do, however, receive
annual letters summarizing their monthly electricity use for the previous year for information
2-3
purposes only (standard rate customers receive these reports as well). These reports are also
available online.
Back-Office Processes
M-Power customer meters are not read like standard program meters (unless a customer requests
this in specific situations). As such, the Smart Card is the conduit through which SRP has access
to M-Power customers’ consumption information.
•
•
•
•
•
•
As previously mentioned, when the card is inserted into the UDT, the full amount of credit
that was on the card is transferred to the meter through the UDT, and the card balance goes to
zero. At the same time, the meter consumption level at that time is transferred to the card, so
that it may be transferred to SRP via the PayCenter the next time electricity credits are
purchased.
When a customer purchases electricity credits at a PayCenter, the following information is
transferred from the SmartCard to SRP’s back office system:
• Meter read (cumulative kWh and max kW)
• Reading date/time: that is, when the card was last read by the UDT
• Cash purchase amount
• Transaction date/time: that is, time stamp at PayCenter when the customer made a
purchase, and therefore different than the “reading date/time”
• Status date: that is, when the card is inserted into and read by the UDT (some time after
the transaction date/time)
• A code indicating whether a disconnect occurred (i.e., the customer ran out of power)
• Also the remaining credit on the UDT is recorded; if this is negative, this is another
method of knowing if a disconnect occurred
• Any Friendly Credit used (and corresponding date/time stamps)
During the same transaction, data are pulled from the back office to the PayCenter in realtime, and recorded on the customer’s transaction receipt:
• Any Emergency Credit that was required (and corresponding date/time stamp): this will
be added on to the arrears balance.
• Arrears owing, arrears payment, updated arrears balance: the system accesses how much
the customer owes in arrears and takes 40% of the cash purchase value to apply to that
amount; the remaining 60% is used to purchase electricity credits.
• Confirmation number if paid by e-Chex
PayFirst: SRP can use this function to input any additional debts the customer may have,
which must be paid off first at a rate of 40% before the funds can be used to purchase
electricity (e.g., covering any fees and amount owing from previous NSF checks)
GiveFirst: SRP can use this field to input funds they may owe the customer (e.g., if they have
paid by mailed check the value is input in this field, if they are owed any customer service
rebates)
Other information that is transferred to the UDT from the back office via the Smart Card:
• Rates
2-4
• Holiday dates
• Configuration changes (e.g., display changes, credit limit changes)
• Messages to customers
• Given each account has at least two Smart Cards, there may be instances when a card is used
that has not been used in a long time, and therefore contains “old” meter read information;
these are called “stale reads”, and they are tracked and taken into account using both the
“reading date/time” stamp and the “transaction date/time” stamp. Once the card is finally
used again, the consumption information is automatically re-sequenced in the back office
system.
• In cases where the card or UDT becomes corrupt, the meter can be physically probed to
obtain the latest consumption reading. Corruption does not affect overall revenue reporting to
any extent given the infrequency of such occurrences.
This section has provided a high-level of overview of the systems and general processes by
which the M-Power program functions in its current configuration. The following section will
examine the technology and business processes that have been developed in more detail.
2-5
3
M-POWER TECHNOLOGY, SYSTEMS, AND COSTS
Technology
As described in Section 1, the technologies involved in the M-Power program have evolved over
the years. Each component of the technology is now examined in more detail to highlight its
functionality and interconnection with the overall M-Power system. Information regarding the
costs to the customers is also included, although SRP’s technology cost information has not been
included as it is protected under non-discloser agreements with the vendors. The meter, UDT,
and Smart Cards are illustrated in Figure 3-1.
Figure 3-1
M-Power Meter, UDT, and Two Smart Cards
User Display Terminal (UDT) or In-home Display (IHD)
The UDT currently in use at SRP has not substantially changed since 2005. Its trade name is the
ecoMeter, and it is manufactured by Landis+Gyr UK, Ltd. (formerly known as AMPY Metering
Limited). Table 3-1 provides an overview of its functionality.
3-1
Table 3-1
UDT Features
Information Displayed
Current rate per hour is $0.XX
(Customer toggles through
screens using the “Display
Cycle” button)
Rate Y charge is $0.XXXX/kWh
Cost today is $X.XX
Cost yesterday was $X.XX
Cost this month is $X.XX
Cost last month was $X.XX
Enough credit for XX days
Remaining credit is $XX.XX
Languages
English and Spanish
Communications with Meter
Powerline carrier (PLC)
Real-time Display Update Rate
~3 seconds
Installation
By customer (once meter has been installed by SRP)
Power Source
UDT plugs into a standard 110V outlet.
Also a battery back-up for power outage situations.
Manufacturer
Landis+Gyr UK, Ltd.
Model
A version of the ecoMeter (industry name)
Cost to Customer
$99 deposit
(87.50 of which is refundable upon return of the UDT to SRP)
Customer Support Options
Customer service telephone support
Trouble shooting tips on mysrpmpower.com, and in the Quick Start Card
and User’s Manual
Meter
The meter is very similar in functionality to a “smart meter”, and also has disconnection
capability. One main distinction from today’s smart meters which are a part of advanced
metering infrastructure is that there is no real-time communication capability between the meter
and the utility; as previously described, this is accomplished via the Smart Card instead.
Table 3-2
Meter Features
Manufacturer
Landis+Gyr UK, Ltd.
Model
5252 (latest model)
Communications with UDT
Powerline carrier (PLC)
Installation
By SRP. Must marry with specific UDT intended for the residence.
Cost to Customer
None
3-2
PayCenters (Self-service Kiosks)
As of September 2009, SRP had 95 PayCenters located throughout the Phoenix area. PayCenter
attributes are listed in Table 3-3.
Table 3-3
PayCenter Features
Manufacturer
AllKiosk
Languages
English and Spanish
Payment Methods
Cash and e-Chex
Non-M-Power customers can pay bills via the PayCenter as well
Communications with M-Power
Network
Real-time connection to the SRP back office system (the AMPY server),
PayCenter maintenance, iNovah interface (point-of-sale software), and the
e-Chex server (see Figure 3-2)
Installation
By SRP; must marry with specific UDT intended for the residence
PayCenter Support
Dedicated maintenance team
PayCenters can be swapped out on the spot for faster recovery time
M-Power Business Systems
Customer Information and Back Office Systems
Multiple systems are used to support M-Power, including the back office system, known as the
AMPY Management System or AMS, the Customer Information System (CIS), called Phoenix,
and iNovah, which is the PayCenter server. The interrelation of these systems as well as the eChex server (the interface with the electronic checking system) and the PayCenter support and
maintenance system are illustrated in Figure 3-2.
3-3
Figure 3-2
PayCenter Interfaces (Source: SRP, 2009)
Phoenix, the CIS, is now the system of record for M-Power customers, as it is for all other
residential and commercial customers at SRP. All M-Power customer accounts are set up
through Phoenix, and all customer information is housed in the system, including addresses,
locations within the service territory, programs the customers are involved in, financial
information including customer credit history, as well as records of customer interactions with
SRP’s customer service center. The Phoenix system was developed in-house by SRP with
contractor support.
The back office system (AMS) was developed by AMPY and configured for the SRP M-Power
program. It allows for the tracking of customer accounting details including customer service
area and billing information, overall transaction history, arrears history, meter read details, meter
credit status, and self-disconnection history. See Appendix B for various screen shots of the back
office software.
3-4
The CIS (Phoenix) and the back office system (AMS) were originally completely separate
systems in the early days of the M-Power program. Using this configuration, SRP would use the
back office system to enroll M-Power customers, but would
Lesson learned: use one
then need to enroll the customers in the CIS as well. This
integrated system from the
configuration entailed a lot of manual labor to ensure the
beginning. The integration
two systems were reconciled and accounts were not out of
of the customer information
balance. As M-Power customer numbers grew, so did the
and back office system took
motivation to merge the systems. The process was
several years to develop and
complicated, taking approximately five years to complete.
debug.
However, as of March 2008, the systems were
synchronized, and Phoenix is now used to enroll customers,
and any arrears owed by new M-Power customers on prior accounts flow into the back office
system. The systems are also synchronized to reconcile when payments are made, etc. A report is
run every night to reconcile the two systems, and although there are still some manual
adjustments required to keep the systems synchronized, they have decreased dramatically since
the initial synchronization in March 2008, and continue to decrease.
Revenue Reporting
Although M-Power customers do not receive bills, the systems were originally used to
essentially “make” a monthly bill for M-Power accounts, so they could be “forced” into a billing
cycle such that they would be consistent with the other residential accounts for revenue and tax
reporting purposes, energy efficiency reporting, etc. While such a set-up was manageable when
M-Power electricity purchase volumes were relatively low, recent enrollment rates have
necessitated the need for an alternative approach. In the 2008 fiscal year (May 2007 through
April 2008), the accounting methodology has switched to cash-based revenue recognition (i.e.,
revenue is recognized based on cash purchases) rather than the conventional billing cycle
approach. For load forecasting purposes, the cash value is transferred back to an energy quantity.
It is felt that this method of revenue recognition is much more accurate than the previous “forced
billing cycle” approach. This new approach is also tied more closely to the synchronized
business systems previously described as well.
M-Power Costs
Customer Costs
M-Power Start-Up Costs
As previously described, for initial M-Power account set-up, customers pay a $99 deposit for the
equipment ($87.50 of which is refundable upon return of the UDT), as well as a $28 (plus tax)
service establishment fee. In addition, a $30 electricity credit is placed on one card. This total
amount, known as a “pay-down balance”, is uploaded as an arrears balance that the customer
will pay off gradually with each subsequent purchase transaction.
M-Power Electricity Rates
M-Power customers pay a per-kWh flat rate, which varies seasonally, and is comprised of
various adjustment charges. In addition, there is a monthly service charge of $15, which is
collected either through hourly or periodic deductions from the account balance. The charges
3-5
effective November 2009 are listed in Table 3-4, and are compared against the standard
residential price plan (also referred to as E-23). On average, in the summer months the M-Power
rates generally result in slightly lower electricity bills compared to the Standard rate beyond a
threshold consumption level. In the winter months, the M-Power rate is always more expensive
than the Standard rate. Using average seasonal consumption levels for M-Power customers, MPower customers may actually pay $38 more on an annual basis. It is, however, difficult to
compare the two rates using the same consumption levels, as it is possible that a conservation
effect induced through the M-Power program may result in lower overall consumption levels.
3-6
Table 3-4
M-Power per-kWh Charges (Effective November 2009)
Summer
Charge
Component
Peak Summer
Winter
(May-June and SeptemberOctober)
(July-August)
(November-April)
M-Power*
Total per-kWh
$0.0984 per kWh
Standard (E-23)**
$0.1019 per kWh
(≤2,000 kWh)
M-Power
$0.1030 per
kWh
$0.1061 per kWh
(>2,000 kWh)
$12 per month
Standard (E-23)
$0.1073 per kWh
(≤2,000 kWh)
M-Power
$0.0872 per kWh
$0.1119 per kWh
(>2,000 kWh)
$12 per month
$0.0879 per kWh
(≤400 kWh)
$0.0688 per kWh
(>400 kWh)
Monthly Service Charge
$15 per month
Estimated Monthly
Rates
Assuming 1,069 kWh/month***
Assuming 1,539 kWh/month***
Assuming 697 kWh/month***
$120 per month
$174 per month
$76 per month
$121 per month
$15 per month
Standard (E-23)
$177 per month
$15 per month
$12 per month
$68 per month
(excl. taxes)
M-Power cheaper at:
> 857 kWh/month
> 697 kWh/month
(excl. taxes)
Sources:
* http://www.srpnet.com/payment/mpower/pdfx/MPowerE24Nov2009.pdf
** http://www.srpnet.com/prices/pdfx/BasicPlan1009.pdf
*** Based on average M-Power customer usage for FY10
3-7
Never (Standard always less expensive)
SRP Cost-Benefit of M-Power
Table 3-5 provides an overview of the 2010 benefit-cost calculations for the common energy
efficiency screening tests.
Table 3-5
M-Power per-kWh Charges (Effective November 2009)
Benefits
(NPV)
Costs (NPV)
Net Benefits
Benefit/Cost
Ratio
Cost of
Conserved
Total
Resource
(TRC)
$10,960,736
$5,574,298
$5,386,438
1.97
$0.037
Utility (UCT) /
Program
Administrator
Cost Test
(PACT)
$10,960,736
$5,574,298
$5,386,438
1.97
$0.037
Participant
(PCT)
$15,844,237
--
$15,844,237
--
Ratepayer
Impact (RIM)
$10,960,736
$21,418,535
-$10,457,799
0.51
$0.144
Societal Cost
(SCT)
$10,960,736
$5,574,298
$5,386,438
1.97
$0.037
The benefits in terms of electricity savings are derived using a 12% value for the conservation
effect of the M-Power program, which SRP assessed through by three different conservation
impact studies (see Section 5 for more details). To attribute kWh savings, SRP applies the
conservation effect savings percentage to the load of current year subscribers. In a departure
from conventional energy efficiency program evaluation methods, SRP does not attribute any
impact beyond the current year. This is partly to ensure a conservative estimate of savings, but
also because the impact studies were not constructed to resolve whether the behaviors
undertaken by M-Power subscribers would persist into the future if the customer remained on MPower, or if that customer migrated to another SRP rate. All tests pass the benefit/cost test except
the Ratepayer Impact Model, which includes customer bill savings as a cost, given the savings
represents a revenue shortfall. It is rare that an energy efficiency program that elicits a bill
savings will pass the Rate Impact Test. The estimated program Societal Test benefit/cost ratio for
2010 is 1.97, which places it above many conventional energy efficiency programs.
This section has provided a high-level overview of the basics of the M-Power program in terms
of the technologies and systems that support it and the resulting costs. Building on this
information, as well as how the systems work together to provide the M-Power service to
customers as outlined in Section 2, the following section outlines the results of several SRPcommissioned studies aimed at understanding customer perceptions of the M-Power program.
3-8
4
THE CUSTOMER EXPERIENCE
Past Study Overviews and Findings
SRP has dedicated significant resources to understanding what customers think about the MPower program. This is partly in response to critics who are concerned that prepay programs may
be unfairly forcing low income customers into power outage situations. Some of the initial
market research studies were in partnership with the Arizona Community Action Association
(ACAA), a low income advocacy group, to jointly investigate such concerns. As a result of some
of the early findings that suggested high levels of customer satisfaction, and even “increased
self-esteem”7 due to perceived financial benefits and feelings of self-efficacy in relation to bill
payment, the ACAA now publicly supports the M-Power program.8
Many of the various studies that were performed are outlined in Table 4-1, and a more detailed
review of the results can be found in Appendix C. Considering the results of these studies, some
themes emerge.
Table 4-1
Past M-Power Market Research Studies
Reference
Study Title
Study
Period
Subject
SRP, 2009b
CCTS M-Power
Quarterly Scorecard
May-Jul 2009
(performed
every quarter)
Summary presentation of telephone survey results
from M-Power customer who visit the PayCenter
kiosk to purchase power
PRIZM, 2009
PRIZM Analysis
Jul 2009
PRISM analysis/marketing
2009
Overview presentation of program
Traasdahl,
2009
WestGroup
Research,
Inc., 2007
SRP M-Power Materials
Survey: Topline Report,
November 2007
Oct 2007
Telephone interviews to assess salience of updated
communications materials; 201 M-Power
customers
WestGroup
Research,
Inc., 2006a
SRP M-Power
Communications
Focus Group Research
Nov 2006
Focus group report, 3FGs, opinions on M- Power,
the Starter Kit and different M-Power
communications materials; customers randomly
selected from M-Power population who meet
desired criteria (e.g., on M-Power for at least two
months)
7
Personal communication, Betty Pruitt, M-Power Marketing, September 16, 2009
8
Anecdotally, Ms. Pruitt is a former ACAA employee who was initially a critic of the M-Power program. She was
convinced through customer testimonials of its benefits, and is now an SRP employee working on M-Power
Marketing.
4-1
Table 4-1 (continued)
Past M-Power Market Research Studies
Reference
Study Title
Study
Period
Subject
SRP, 2006
SRP M-Power Shadow
Project September 2006
– Executive Summary
Sep 2006
Executive summary of the SRP M-Power Shadow
Project, which involved 8 in-depth, in-home
interviews from customers from a range of credit
codes, length of time on program, language,
income and housing type
WestGroup
Research,
Inc., 2006b
SRP
M-Power/AMPY
Benchmark Study
2006
Mar 2006
Telephone survey results, 402 M-Power
customers; purpose was to obtain baseline data on
attitudes towards M-Power, with emphasis on
PayCenter machines, as new machines were to be
installed starting in April 2006.
Reiley &
Johnson,
2006
What Determines MPower Customer
Satisfaction: How SRP
Can Attract and Retain
M-Power Customers
2006
University of Arizona Assessment, telephone
survey conducted by WestGroup of 401 M-Power
customers that had been on the program for 18
months or more. Goal was to assess customer
satisfaction with the aim of understanding what
would make customers stay on M-Power so as to
reduce turnover rates; another goal was to
understand traits of long-term M-Power customers.
WestGroup
Research,
Inc., 2002
SRP M-Power
PayCenter Research:
Summary of Findings
Nov 2002
Telephone interviews regarding convenience of
PayCenters in Circle Ks (convenience store); 214
M-Power customers; interviewees had been on
program for at least 2 months and had a telephone
number on file.
WestGroup
Research,
Inc., 2001
SRP M-Power Focus
Group Report—June
2001
May 2001
Focus group report, 4 focus groups, opinions on
areas of process and operational improvement.
One focus group was comprised of people who
had left the M-Power program.
WestGroup
Research,
Inc., 1999a
Pay As You Go Focus
Group Report
Aug 1999
Focus group report, 2 focus groups, joint between
SRP and Arizona Community Action Association,
opinions on M-Power program
WestGroup
Research,
Inc., 1999b
SRP/ACAA Pay As You
Go Study
Sep 1999
Telephone survey report, 179 Pay as You Go
customer (precursor to M-Power program); joint
study between SRP and Arizona Community
Action Association (ACAA), opinions on MPower program.
Overall Satisfaction Levels
One obvious finding, which is consistent with every customer study, is that M-Power customers
are very fond of the program. This finding was first found with the 100-home pilot discussed in
Section 1, with the oldest studies reviewed for this report (circa 1999), and with the most recent
studies reviewed, in the form of the Customer Contact Tracking Study (CCTS). This is an
ongoing quarterly customer service market research report, which as of 2007 began to include
M-Power-specific questions. For the M-Power section of the CCTS, the respondents are chosen
randomly from M-Power customers who use PayCenters to purchase power, i.e., the vast
4-2
majority of M-Power customers. From the range of data available, the percent of customers who
are satisfied or very satisfied with M-Power ranges from 83% to 96% (SRP, 2009b).
Overall satisfaction of M-Power customers with SRP is generally favorable as well. In the three
surveys performed prior to the start of the CCTS, those who rated SRP’s performance as very
good or excellent ranged from 70 to 73%. When the CCTS began, the question changed
somewhat, but CCTS data from FY07 to FY10 suggest M-Power customers who were satisfied
or very satisfied with SRP ranged between 85 and 89%.9
Given the high importance SRP places on the customer satisfaction ratings it receives from its
M-Power customers, it is instructive to consider how this satisfaction with the utility compares
between M-Power and non-M-Power customers. As illustrated in Table 4-2, “overall
performance” ratings are comparable, but statistically more M-Power customers perceive high
overall value in the service SRP provides compared to non-M-Power customers. Statistically
fewer M-Power customers gave SRP top ratings for “overall experience”. Therefore, by some
metrics M-Power customers are generally happier with SRP, but not by all metrics.
Table 4-2
SRP Ratings: Comparing M-Power and Non-M-Power Customers (FY2010)10
Percent 9/10 Rating
(Out of 10)
M-Power
N=337
Non-M-Power
(all other rates)
N=3574
Overall Performance as an Electric Company: A
service quality metric, and SRP’s core business
measure
66%
63%
Overall Value: Value received from SRP considering
amount you pay for services
46%*
37%
Overall Experience: Broad measure of service
received, impressions of the company, customer
experience
44%
50%*
* Statistical difference between groups
Perceived Customer Benefits of M-Power
Most of the survey and focus group reports have similar findings in terms of what M-Power
customers perceive to be the advantages of the program.
One of the main advantages reported is the educational and awareness aspect of the program: the
UDT provides each customer with real-time consumption information about their home (update
rate of 3 seconds) allowing householders to see the electricity consumption effect of their
household behaviors. Some reports included mention of extending the education to other
household members, including children.
9
Personal communication, Dena Emary, Senior Analyst, SRP, August 16, 2010.
10
Data are from SRP’s FY2010 “Customer Perspectives” survey, a marketing study that includes questions intended
to track overall opinions of SRP, and include representation of all residential sectors.
4-3
Indeed, there have been several studies that report conservation effects by providing this sort of
electricity use feedback in non-prepaid conditions.11 In the case of the M-Power program, one
can hypothesize that customers are more likely to use the UDT to educate themselves about their
unique usage patterns, as the consequences are more severe if they do not do so: they are more
likely to have their electricity shut off unexpectedly.
“I tell all my friends that it was the best teacher for me to teach me how to use electricity
efficiently.” WestGroup Research, Inc., 2001, p. 4.
“Because I live paycheck to paycheck, it makes me more conscious of how much I’m
using. I don’t let my daughters open the refrigerator because I know it’s using more
electricity. It makes me more aware of what I’m using and where it’s being wasted.”
WestGroup Research, Inc., 2001, p. 3.
“I found out how much my dishwasher runs, if I have to use it in the summer time, it's
cheaper to just hand wash my dishes and dry them in the sink, rather than have my air
conditioning catch up for that dishwasher.” WestGroup Research, Inc., 1999a, p. 4.
Perhaps a consequence of the knowledge gained regarding a household’s electricity usage
patterns, another advantage that is often reported is the sense of control the M-Power program
provides customers. This includes the ability for customers to monitor their power usage, to pay
for power at their own pace (e.g., daily or weekly instead of monthly), and even the ability know
and be ready for a disconnection if necessary.
“It allows me to budget. Being able to see what I spent last month allows me to break it
down into pay periods so I can put money aside that I know I’m going to put on the card.
You don’t have that bill coming. I wish there was a way that we could refill the card over
the phone because sometimes I’m going “Oh man, I’ve got to put money in, it’s late, and
I’ve got a four year old so I am pulling him out of bed to go fill the card because of
forgetting.” WestGroup Research, Inc., 2006a, p. 17.
“Actually for me it’s a lot better than getting a bill at the end of the month especially in
summer. It’s a lot easier to pay $40 a week because the end of the month is when all the
rest of my bills are due so I’m flat broke.” WestGroup Research, Inc., 2006a, p. 17.
“You have control over how much power you are using. There might be reasons for the
spikes like leaving on the air conditioning while you were gone that day and things like
that. You have more control over it.” WestGroup Research, Inc., 2001, p. 3.
“You can plan your usage better.” WestGroup Research, Inc., 1999a, p. 5.
The other main advantage often reported by M-Power customers is the perception that the
program saves them money. As illustrated in Section 3, kilowatt for kilowatt, the M-Power
program can actually be more expensive than the standard program. However, many customers
reported that their increased knowledge of consumption patterns makes them more likely to be
able to use less electricity, thus resulting in an overall monetary savings as well. In addition, as
reviewed in Section 3, start-up costs are less for the customers. Finally, given there are no late
11
Studies involving an in-home display (IHD) in non-prepaid conditions have reported conservation savings in the 0
to 7% range.
4-4
fees or disconnection charges, this can be another source of monetary savings for customers who
previously had to pay these charges through the standard program.
“An advantage is that it saves you money and you can put like $2 if you have to, if you're
low on cash or whatever. You don't have to worry about your lights going out.”
WestGroup Research, Inc., 1999a, p. 4.
Other benefits that were reported included:
•
•
The provision of an alternative to the hassle and sometimes embarrassment of going through
the conventional program’s disconnection/reconnection process
The elimination of the fear of not knowing how much a monthly bill will be
Perceived Disadvantages
Most of the M-Power disadvantages reported relate to the PayCenters. A qualitative study from
2006 found that more payment options at PayCenters would be preferred by customers, as well
as the ability to pay by phone or online; also, there were complaints about the working order of
the PayCenters (WestGroup Research, Inc., 2006a). Quantitative research from this era found
that 71% reported encountering a PayCenter machine that was not working in the previous year
(WestGroup Research, Inc., 2006b). In addition, satisfaction with PayCenters decreases the
longer customers are on the program (choices ranged from
