20111004PAC to CAPAI Attach 2.pdf

Final Report

Washington Low-Income
Weatherization Program

Prepared for:
Pacific Power

January 19, 2007

ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 1 of 44

Quantec Offices
720 SW Washington, Suite 400
Portland, OR 97205
[phone]; [phone] fax
www.quantecllc.com
Printed on
recycled paper
Printed on
recycled paper
3445 Grant St.
Eugene, OR 97405
[phone]; [phone] fax
28 E. Main St., Suite A
Reedsburg, WI 53959
[phone]; [phone] fax
1722 14th St., Suite 210
Boulder, CO 80302
[phone]; [phone] fax
20022 Cove Circle
Huntington Beach, CA 92646(714) 287-6521
Prepared by:
M. Sami Khawaja
Sara Wist
Doug Bruchs
Eli Morris
Elizabeth Daykin
Quantec, LLC

K:\2006 Projects\2006-31 (PC) WA Low Income Weatherization\Reporting\200631 Washington Low Income Weatherization
Evaluation 010907.doc

ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 2 of 44

Table of Contents
1. Executive Summary ...................................................................................1
Program Description ............................................................................................................1
Evaluation Approach ...........................................................................................................1
Major Findings .....................................................................................................................2
Recommendations ................................................................................................................4
2. Process Evaluation ....................................................................................7
Program Services .................................................................................................................7
Data Collection ....................................................................................................................7
Process Findings ..................................................................................................................9
Energy Education ...............................................................................................................12
Overall Findings .................................................................................................................16
3. Impact Evaluation .................................................................................... 19
Deemed Savings – Measure Analysis ................................................................................19
Actual Savings – Billing Analysis .....................................................................................23
4. Non-Energy Benefits ............................................................................... 29
Participant Impacts .............................................................................................................29
Mobility..............................................................................................................................30
Arrearage Impact ...............................................................................................................32
Economic Impacts ..............................................................................................................33
Environmental Benefits .....................................................................................................34
5. Cost-Effectiveness Analysis .................................................................. 37

Quantec — Washington Low-Income Weatherization Program
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 3 of 44
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 4 of 44

1. Executive Summary
Program Description
Pacific Power’s Weatherization Program (the Program) in Washington assists low-income
households in controlling energy consumption and heating costs through comprehensive home
weatherization and energy education.
Between July 1, 2003, and June 30, 2005, the Program provided service to 419 Pacific Power
customers. As shown in Figure 1, the majority of participants were served by the Opportunities
Industrial Center (OIC) of Washington located in Yakima. The remaining participants were
either served by the Blue Mountain Action Council (BMAC) in Walla Walla or the Yakima
Valley Farm Worker’s Northwest Community Action Center (NCAC).
Figure 1. Program Participation Among Agencies
77
130
212
419
0 100 200 300 400 500
BMAC
NCAC
OIC
All agencies
Participants

Evaluation Approach
Pacific Power contracted with Quantec to conduct an impact and a process evaluation of the
Program. The process evaluation was designed to assess Program delivery and efficacy,
bottlenecks, barriers, and means of improvement. The impact evaluation assessed energy
impacts, non-energy benefits, and Program cost effectiveness. The following were the major
tasks associated with the evaluation:
Data Collection

Data that were provided by Pacific Power and the agencies included:
• Participant and non-participant billing histories
• Measure installations
• Program costs
Quantec — Washington Low-Income Weatherization Program 1
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 5 of 44

Surveys were conducted with 65 Program participants to assess multiple aspects of the Program,
including the value of the Program, Program delivery, client satisfaction levels, and customer
recall of energy education recommendations.
In-depth discussions with key staff at each agency were conducted to ensure that all facets of
Program delivery were assessed, including bottlenecks, client and agency satisfaction, methods
of improving delivery, and agency assessment of non-energy benefits.
Finally, an interview was conducted with Pacific Power’s inspector to provide insight into the
issues identified through this evaluation and by the inspector, and to discuss improvements that
have been made at the agency level.
Evaluation of Program Energy Savings
Estimated as well as actual Program energy savings were assessed in the following manners:
• Deemed Savings: A measure analysis to identify measure installation frequencies and
estimated savings was conducted.
• Actual Savings: The Princeton Scorekeeping Method algorithm was run to estimate
weather-normalized, Program-induced energy (kWh) savings based on participant and
non-participant billing data.
Assessment of Non-Energy Benefits
In addition to those that were reported by the participants, numerous non-energy benefits in the
areas of economic impact, environmental benefits, mobility, health and safety, and participant
arrearages were analyzed.
Assessment of Cost-Effectiveness
An economic analysis of the Program, in accordance with the benefit-cost tests from the
California Standard Practices Manual, was performed. Results are presented both with and
without the inclusion of non-energy benefits.
Major Findings
Cost Effectiveness
The Program did not pass the traditional cost-effectiveness test. The Total Resource Cost (TRC)
benefit cost ratios were between .60 and .65 depending on the stream of avoided costs used.
However, when non-energy benefits are included, the Program passes the TRC with a
benefit/cost ratio between 1.01 and 1.06.
We did not find that cost-effectiveness is recognized by all parties as an explicit goal of this
Program. Theoretically, only measures with a savings to investment ratio (SIR) of 1.0 or more
should be installed. However, that is not the approach followed by the agencies. We discuss this
issue further below in relation to the use of a Department of Energy (DOE) approved audit.
Quantec — Washington Low-Income Weatherization Program 2
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 6 of 44

Electricity Savings
Overall, Program net annual energy savings are estimated at 1,840 kWh (12% of pre-Program
energy consumption) per completed household. This is an improvement over the prior Program
period, which had an evaluated net annual energy savings of 1,439 kWh (8% pre-Program
energy consumption).
Estimated savings during the audit seem to greatly inflate the potential, as shown in Figure 2.
Figure 2. Actual and Expected Savings by Program Year
0
1,000
2,000
3,000
4,000
5,000
6,000
2003-2005 2001-2003
Expected
Actual

Non-Energy Benefits
The Program also provided non-energy benefits to participants, the environment, and the
economy. At the participant level these included increased comfort (reported by 94% of survey
respondents), improved health (66%), decreased work or school absences (43%), and more
money for non-energy necessities (83%). Additionally, while 68% of respondents reported that
the Program had improved their ability to stay in their current homes, an analysis of participant
billing data found that the Program may have helped to prevent approximately 68 participant
moves (16%) . Other benefits included:
• An estimated 6 net job-years of employment
• Approximately $557,605 added to the Washington economy
• Approximately $22,809 worth of air emission reductions based on relevant market values
as of August 2006
• A reduction in annual arrearages, totaling approximately $26,816
Energy Education
Great improvement was made in participant recollection of energy education materials in
comparison to the previous Program period. In fact, 75% of the participants surveyed
remembered receiving supplemental material compared to 35% in the previous Program
evaluation. Additionally, most of the participants implemented at least one of the energy
Quantec — Washington Low-Income Weatherization Program 3
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 7 of 44
education tips. Participants reported that the agency and weatherization staff were courteous, and
few problems or complaints were identified.
Although we requested from all agencies that we be allowed to participate in at least one energy
education session, no such arrangements were made for us during our site visits. As such, we are
unable to comment on the quality of the education. In general, we do not feel that energy
education delivered through the auditor without a clear curriculum, materials, clear approach,
training, etc. is considered good energy education. We feel that for the compensation received by
the agencies, they need to develop a significantly more thorough energy education program.
Agency Program Assessment
During our interviews with the Agencies, we asked for an assessment of the Program and the
relationship with Pacific Power. The answers were unanimous: all agency staff liked the Program
and thought that Pacific Power was flexible and easy to work with.
Recommendations
While the Program did not pass our cost-effectiveness tests without the inclusion of non-energy
benefits, we feel Program enhancements can greatly improve the results.
Specifications within the contract should be revisited, as it was simply extended without
alteration, to July 31, 2007. The following are issues to consider for the next contract period:
1. The requirement to use the DOE approved audit on all homes needs to be fulfilled. The
contract should state that every job must be analyzed using the DOE approved audit tool in
conjunction with the household’s pre-weatherization consumption data. Every invoice must
include the audit runs and clearly show that only measures with SIR of 1.0 or better are being
installed. Failure to follow contract requirements should have a tangible consequence.
2. The “lookup tables” that have been used in previous programs should be destroyed to prevent
continued use of this method of energy estimation.
3. Glass replacement should be moved from “Major Measures” to “Supplemental Measures,”
and should be allowed only if found to be cost-effective by a DOE approved audit, with pre
weatherization consumption incorporated. The state is currently revising their specifications
and as of Jan. 1, 2007 will just pay 25% of the cost of replacing windows. Since Pacific
Power pays up to 50%, this will likely mean that windows will not be installed unless they
are considered a repair.
4. Including rebates for dehumidifiers and air-to-air heat exchangers in the Program should be
reconsidered. They are not currently being installed.
5. Stating that showerheads are always cost effective should be reconsidered. While this is
nearly always the case, their cost effectiveness is a function of the frequency of use and water
flow rates. In order to ensure that this measure is cost effective, these rates should be
measured and replacement should be considered when frequently used showers have flows of
greater than 2.5 gallons per minute.
6. With the decrease of the cost of compact fluorescent light bulbs, increasing the maximum
number installed to 10 should be considered. Also, lowering the number of hours of use from
Quantec — Washington Low-Income Weatherization Program 4
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 8 of 44

Quantec — Washington Low-Income Weatherization Program 5
three per day to as low as one per day should be considered; this would still be cost effective
for the average home.
7. We do not believe that the energy education currently being provided by the agencies
justifies the cost of $200 per home. With the auditor performing the service at the same time
as the audit, $50 per home is more reasonable. To continue receiving $200 per home, it is
suggested that agencies have a separate employee on staff to provide energy education, and
that they should develop a model for providing energy education, attend training sessions,
and have a checklist of items to be covered. There are proven energy education approaches
that agencies need to follow.
8. Cost effectiveness acquisition of energy savings needs to be explicitly recognized by all
parties as one of the Program goals.
9. All Program spending, including multiple funding sources, needs to be reported by the
agencies. In order to improve the tracking of costs, Pacific Power should replace the “other
funding” category and record each funding source in addition to the Program rebates. This is
a common practice, and makes business sense.

ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 9 of 44
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 10 of 44

2. Process Evaluation
Process evaluations tell the story of the program. They describe program delivery, bottlenecks,
what worked and what did not, and provide overall assessment of program efficacy.
Program Services
Agencies employ energy professionals who are trained to evaluate and measure the performance
of a home. They have the knowledge to identify the important energy-saving opportunities and
measures that will result in the most savings. There is a high use of diagnostics in this Program
with an emphasis on blower door testing. The energy professional also focuses on enhancing
health and safety in the client’s home. Other Program services include conservation and energy
education. The goal of all Program services is to conserve energy, reduce clients’ energy burden,
and create a more comfortable living space for the client. All services are available in English
and Spanish.
The Program installs a variety of measures to improve the efficiency of clients’ homes, as listed
in the impact evaluation. The same criteria are used for deciding which measures to install in all
home types (site-built, manufactured home, etc.). Reasons for a “walk away” (deciding not to
install any measures in a given structure) would include that a home was built after July 1, 1991,
or that there existed physical barriers, structural damage, or unsafe/unsanitary conditions at the
home.
Data Collection
Data collection for this portion of the evaluation consisted of:
• Agency visits
• Interview with a Pacific Power inspector
• Participant surveys
• Review of relevant program documents and filings
Agency Visits
Multiple interviews were conducted on-site with staff at each agency to ensure that all facets of
Program delivery were assessed, including information regarding bottlenecks, client and agency
satisfaction, methods of improving delivery, and agencies’ assessment of non-energy benefits.
We attempted to schedule site visits to participating homes in order to observe components of
Program services. Unfortunately, for various reasons, we were only able to visit one home in
Toppenish.
Interview with Pacific Power Inspector
An interview was conducted with Pacific Power’s inspector to provide insight into the issues
identified through this evaluation and by the inspector, and to discuss improvements that have
been made at the agency level. It was found that a number of issues identified in this evaluation
Quantec — Washington Low-Income Weatherization Program 7
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 11 of 44
had already been addressed with the agencies. However, many of the resulting improvements did
not take place until after the close of the 2003-2005 Program period. For instance, the DOE
approved audit TREAT, which is now in use by the agencies, was introduced in October, 2005
which was three months after the end of the contract period. Only that information which was
found to be relevant for the evaluation period is included here, unless otherwise stated.
Participant Interviews
In addition to obtaining information on basic household characteristics, surveys were conducted
with participants in an effort to assess the value of the Program, the Program’s delivery, client
satisfaction levels, and finally, to test participant recall of energy education recommendations.
Surveys were also used to assess non-energy benefits, which are further discussed in Section 4.
Sample Selection Methodology
While the entire population of households that participated in the Program between July 2003
and June 2005 was eligible to interview as part of the evaluation, as a result of several filters, the
final sample used to conduct the surveys consisted of 211 participants. Participants were
removed from the sample of potential respondents based on the following criteria:
• Inability to match participant with Pacific Power customer information file (contains
address, home number, etc.), possibly due to relocation
• Account inactive at time of survey effort
• Invalid or missing phone number
• Repeat participant at a different location
Table 1 details the attrition associated with each filter and provides the final sample size used for
the participant survey.
Table 1. Sample Attrition Participant Survey
Metric
Number of
Households %
Number of
Unique
Participants
Removed
Percentage of
Total Unique
Participants
Removed
Total Program Participants 419 100%
Matched to Customer Information File 340 81% 79 18%
Account "Active" At Time of Survey 225 60% 115 31%
Valid Phone Number 219 59% 6 2%
Duplicate Individuals* 211 57% 8 2%
Final Sampling Frame 211 57%
*Different agreement number, but same person and phone number
Quantec — Washington Low-Income Weatherization Program 8
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 12 of 44

Sixty-five phone interviews were completed. Table 2 demonstrates the hard-to-reach nature of
the Program participants.
Table 2. Sample Attrition
Metric
Number of
Unique
Participants
Removed
Percentage of
Total Unique
Participants
Removed
Number of
Unique
Participants
Percentage of
Total Unique
Participants
Total Program Participants 211 100%
Inactive Phone Numbers 84 23% 127 60%
Participant Refusal 11 5% 116 55%
Ineligible* 9 4% 107 51%
Unresolved** 42 20% 65 31%
Final Sample 65 31%
*Client moved into home after July 2003 or did not remember receiving weatherization services
**Defined by multiple calls resulting in the following: no answer, busy signal, answering machine, “not available,” or request
for call back

Review Program Documents
In order to get a better understanding of Program delivery intent, we reviewed the individual
agency contracts with Pacific Power.
Process Findings
The following sections present our findings by major component of the Program services.
Client Eligibility
Customers are eligible to participate if they are Pacific Power customers who use electric heat
and their household incomes do not exceed 125% of the Federal Poverty Guidelines (or do not
exceed 60% of the state median income).1 Households that do not heat with electricity are
eligible to receive base load measures. Agencies identify qualified households primarily through
their Energy Assistance programs. Other identifying sources are community centers, senior
centers, schools, and government agencies. As discussed further in Section 3, it was thought that
the screening process has resulted in making it more difficult for participants to qualify for the
Program, and that this customer base is generally considered “hard to reach,” making
participation recruitment challenging.
Once an agency determines that a household is eligible, the following process is supposed to take
place:
• Referral to weatherization staff

1 According to the agencies’ contract with the Company, the Program applies to “residential customers residing in
existing dwelling built before July 1, 1991, where electricity is their primary source of hearting energy. This is
defined as an electric system that is operable and permanently installed with capacity to heat at least 51% of the
dwelling.”
Quantec — Washington Low-Income Weatherization Program 9
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 13 of 44
• Energy education (during intake process)
• Audit (which also includes in-home energy education)
• Agency crew scheduled
• Subcontractor scheduled (if necessary)
• Installation of measures
• Inspection of completed household
A number of these steps have been bypassed at the agency level due to lay-offs, lack of
communication within each agency, and a lack of accountability. In particular, it was a finding of
the Pacific Power inspector that in some cases weatherization projects had been left incomplete,
as demonstrated when follow-up visits to the weatherized homes revealed as many as 80% of the
homes in a specific region did not receive all of the services that were billed to Pacific Power. A
new system of accountability, requiring agency staff to “sign-off” any work that is completed,
was thought to be necessary to help guarantee that this issue be resolved. While this system was
not formalized, discussions between the inspector and the agencies may have resulted in some
improvements that could show up in future evaluations.
Energy Audit Specifications
This is the cornerstone of any energy saving program offering. The agency contract with the
Company states:
“To the extent that a Department of Energy approved audit determines that a major
measure is cost-effective and such a major measure qualifies for installation, it must be
installed if financial assistance will not be offered for any other measures.”
It further states that “[m]easures must be determined through audit results to be fully cost-
effective”
The intent of this language in the contract is clear:
1. A Department of Energy approved audit must be used
2. Cost-effectiveness is determined by the audit tool
3. Cost-effective measures must be installed
Energy Audit Realities
The agencies generally did not use any audit tools in estimating savings at individual sites.
Instead, they used lookup tables that were provided by Pacific Power prior to the implementation
of the current tariff and contracts, and that have been outdated for many years. These tables
drastically overstate the expected savings, which we noted in our 2001-2003 evaluation:
“We have found no reason for applying these (lookup table) numbers. The contract
between Pacific Power and the agencies does not call for their use. They are highly
inflated and should not be utilized. The agencies should estimate savings in
accordance with the contract (emphasis in original)”
Quantec — Washington Low-Income Weatherization Program 10
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 14 of 44

In addition to the use of these “deemed values,” the agencies do not make use of actual, pre-
weatherization energy consumption data obtained from Pacific Power. This information, as
valuable as it is, does not get used in estimating Program savings or in targeting Program
services. For example, overall, the agencies expected to save 3,249 kWh/home, but actually
saved about 1,840 kWh/home. In comparison, the average expected savings reported in the 2001-
2003 evaluation was 4,775 kWh per home, while the actual was approximately 1,400
kWh/home. Figure 3 displays these numbers.
Figure 3. Actual and Expected Savings by Program Year
0
1,000
2,000
3,000
4,000
5,000
6,000
2003-2005 2001-2003
Expected
Actual

Case Studies
When examining data at the individual home levels, some rather extreme cases are observed, as
displayed in Figure 4. These are not presented as being representative. Rather, they serve as
examples of what can go wrong without the use of proper tools.
Client A: This is a manufactured home with pre-consumption of 10,645 kWh annually. Based on
installation of ceiling insulation (7,514 kWh estimated savings), floor insulation (2,184 kWh),
window replacement (93 kWh), and insulated door (446 kWh), total savings were estimated at
11,170 kWh annually (105% of total pre-consumption). While this is obviously not possible, it
was still recorded as such and the measures were installed. This is a result of the failure to use
the DOE approved audit and not comparing savings estimates to pre-consumption. In this case,
the total project cost was nearly $8,000, of which Pacific Power paid nearly $4,200. When actual
energy savings are considered, estimated by Quantec to be 1,958 kWh, the dollar savings to the
owner can be estimated at about $120 annually. Therefore, this project has a simple payback of
nearly 68 years.
Client B: This is a single-family dwelling with 5,358 kWh consumed during the 12 months
preceding the weatherization. The audit produced an expected total of 4,821 kWh (90%
reduction in consumption). Measures installed included ceiling insulation, wall insulation,
window replacement, some infiltration measures, and insulated doors. Pacific Power’s
contribution was estimated at $2,217 plus a $332 administration fee. Total cost was over $4,750.
Simple savings to investment ratio analysis would have shown doors and windows would not
have been cost-effective.
Quantec — Washington Low-Income Weatherization Program 11
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 15 of 44
Client C: This is a single-family dwelling with 10,530 kWh consumed during the 12 months
preceding the weatherization. The audit produced an expected total of 6,863 kWh (65%
reduction in consumption). Measures installed included ceiling insulation, wall insulation,
window replacement, some infiltration measures, and insulated doors. Pacific Power’s
contribution was estimated at $3,189 plus a $350 administration fee. Total cost was over $6,700.
Simple savings to investment ratio analysis would have shown doors and windows would not
have been cost-effective. Actual savings was estimated by Quantec at about 4,333 kWh. Given
actual savings and the cost of installation, this project has a simple payback of over 25 years.
In all three cases, the DOE approved audit was not used. Cost effectiveness does not appear to be
a consideration. The contract clearly states that this needs to be considered and provides measure
lives to facilitate the calculation of cost effectiveness.
Figure 4. Examples of Extreme Cases
0
2000
4000
6000
8000
10000
12000
A B C
Pre Consumption
Expected Savings
Actual Savings

It was found by the Pacific Power inspector that even when an audit had been performed as
required, the final stages of reporting would revert back to the use of the inaccurate lookup tables
to input savings estimates, thereby corrupting any accurate data.
Introduction of TREAT
Starting in October 2005, after the close of this evaluation period, the State of Washington
Community, Trade, and Economic Development began requiring the use of TREAT as an audit
tool. The requirement is set at at least three homes per month. At this time it is believed that each
of the agencies does apply TREAT to all homes, though this will need to be verified during the
next evaluation. This is a positive step forward. However, use of TREAT by itself does not solve
the issue of inflated savings. TREAT, as well as all audit programs, will inflate savings if actual,
pre-audit energy consumption is not used as an input. Not using the actual consumption is a
serious shortcoming.
Energy Education
Table 3 lists the contract requirements for energy education and compares these requirements
with current agency activities. In most cases, energy education is provided by the auditor during
the same visit and does not include a review of site-specific energy consumption. Table 3
displays, as best as we could determine, the contract requirements and services actually provided.
Quantec — Washington Low-Income Weatherization Program 12
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 16 of 44

Table 3. Minimum Energy Education Required for Reimbursement
Energy Education Contract Requirements Energy Education Activities Provided
1. Conservation tips and materials provided and site-specific
energy consumption reviewed during intake session.
Each participant is provided with a Pacific Power
“Bright Ideas” handbook.
2. Auditor or weatherization crew member describes measures to
be installed and expected benefits to residents.
Participants are informed of measures installed and
expected economic benefits.
3. An in-home education session is provided to household that
includes conservation tips on a room by room basis, instruction on
reading a meter, proper use of heating system, hot water usage
and moisture control.
Client Assessment Survey and post-assessment.
Hands-on participation.
In-home education demonstration.
4. A post-weatherization session with household that addresses
how to live in a weatherized home. Follow –up home visit.
5. Follow-up contact with household is made with a discussion of
the outcome of weatherization services, and energy conservation
recommendations and actions.
Follow–up home visit.

We requested to attend energy education sessions at participating sites. However, these
arrangements were not made for our staff during our agency visits. Therefore, our assessment of
energy education is based on participant surveys and interviews with staff. From this review of
the service, it was found that for the energy education currently being provided, the cost of $200
per home is not justified. With the auditor performing the service at the same time as the audit,
$50 per home is more reasonable. To continue receiving $200 per home, it is suggested that
agencies should develop a model for providing energy education, attend training sessions, and
have a checklist of items to be covered.
As part of the energy education materials, the agencies provide the Pacific Power “Bright Ideas”
Handbook to serve as a reference for energy-saving tips. As shown in Figure 5, 75% of the
participants surveyed remember receiving supplemental material, compared to 35% in the
previous Program evaluation – a statistically significant difference.2
Figure 5. Participants Who Received Information Regarding Reducing Electricity Usage
0%
10%
20%
30%
40%
50%
60%
70%
80%
Yes No Don't know
2003-2005
2001-2003

2 z = -5.12 with a p-value of 0.001
Quantec — Washington Low-Income Weatherization Program 13
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 17 of 44
Across agencies, 85% of NCAC, 80% of BMAC, and 66% of OIC clients recalled receiving
energy education.
Of the 49 participants that recalled receiving energy education materials, 94% thought it was
easy to understand and 96% found it useful. Additionally, 80% had implemented at least one of
the recommended actions, an increase from the previous evaluation. Actions ranged from
adjusting thermostats to closing doors, and being more aware of leaving lights and appliances on
when not needed. The most common action taken was changing the heating thermostat setting, as
shown below in Figure 6. Of those who turned down their heater’s thermostat, 18% provided
before and after temperature information. The average original temperature setting of 73.2° was
lowered to 64.1°, resulting in an average net change of 9.1°.
Figure 6. Additional Actions Taken by Participants
0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50%
Bought ENERGY STAR appliances
Turned up air-conditioner thermostat
Turned refrigerator temperature setting up
Changed furnace filter
Turned down water heater temperature
Other
Installed compact flourescent light bulbs
Turned down heater thermostat
2001-2003
2003-2005

Pacific Power Involvement
This issue was examined from two perspectives. We asked the staff at the agency how they felt
about their involvement with Pacific Power. The opinion was unanimous that Pacific Power was
extremely easy to deal with.
We also examined it from the perspective of the client’s awareness of the Company’s
contribution to the weatherization of their homes. This issue was a concern in the previous
evaluation and continues to be so over the period of this study. As shown in Figure 7, only 14%
of respondents correctly identified Pacific Power as a funding source. Though this is an
improvement from the previous Program evaluation, the majority of respondents (60%) were
unable to identify any funding source for the Program. It was suggested by one agency that
having a flier or handout available to leave at the participating home would be helpful.
Quantec — Washington Low-Income Weatherization Program 14
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 18 of 44

Figure 7. Sources of Weatherization Funding Identified by Participants
0% 10% 20% 30% 40% 50%
Other
Pacific Power
Government
Agency
Don't know
2001-2003
2003-2005

Perceptions of Agencies
Participants gave high marks to the agencies that provided them with the weatherization service.
With regard to courtesy, more than 95% of the respondents rated them positively.
Participant Demographics
Survey results indicate that the average number of people per household is between three and
four (3.6) people. Only 8% of the participants said that this number had increased since receiving
the service, which is unchanged from the previous evaluation. This percentage is possibly
misleading, however, as discussions with agency staff have indicated that participants may be
hesitant to reveal household size, especially when the household has increased.
As shown in Figure 8, 78% of the participants had up to a high school education, 14% have
completed some college or trade school, and one respondent had completed graduate or
professional school.
Figure 8. Highest Education Level Attained by Participants
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
Graduate or professional school
Graduated college
Trade or technical school
Refused
Some college
High School or less
2001-2003
2003-2005

Quantec — Washington Low-Income Weatherization Program 15
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 19 of 44
While the single greatest age category was 35-to 44-years-old, shown in Figure 9, most
participants fell in the 45-and-over range. This represents a 10-year upward shift in the age
demographic from the previous Program evaluation.
Figure 9. Age Distribution of Weatherization Program Participants
0% 5% 10% 15% 20% 25% 30% 35%
Refused
65 or older
55 - 64
45 - 54
35 - 44
25 - 34
Under 25
2001-2003
2003-2005

Overall Findings
Overall, we found the agencies were highly dedicated to providing the best services to the clients
in the Program. We were alarmed, however, by the lack of understanding of the contract
requirements. This is most troubling for the estimation of savings and for determining cost
effectiveness. We feel that the agencies need to clearly understand the contract and follow it
closely. There should be a clear consequence for failure to follow contract requirements.
Recommendations
Based on the review of the contract and discussions with the agency staff and the Pacific Power
inspector, this section highlights some findings and recommendations.
The contract was simply extended to July 31, 2007. No language changes were made. The
following are issues to consider for the next contract period:
1. The requirement to use the DOE approved audit on all homes needs to be fulfilled. The
contract should state that every job must be analyzed using the DOE approved audit tool in
conjunction with the household’s pre-weatherization consumption data. Every invoice must
include the audit runs and clearly show that only measures with SIR of 1.0 or better are being
installed. Failure to follow contract requirements should have a tangible consequence.
2. The “lookup tables” that have been used in previous programs should be destroyed to prevent
continued use of this method of energy estimation.
3. Glass replacement should be moved from “Major Measures” to “Supplemental Measures,”
and should be allowed only if found to be cost-effective by a DOE approved audit, with pre-
weatherization consumption incorporated.
4. Including rebates for dehumidifiers and air-to-air heat exchangers in the Program should be
reconsidered. They are not currently being installed.
Quantec — Washington Low-Income Weatherization Program 16
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 20 of 44

Quantec — Washington Low-Income Weatherization Program 17
5. Stating that showerheads are always cost effective should be reconsidered. While this is
nearly always the case, their cost effectiveness is a function of the frequency of use and water
flow rates. In order to ensure that this measure is cost effective, these rates should be
measured and replacement should be considered when frequently used showers have flows of
greater than 2.5 gpm.
6. With the decrease of the cost of compact fluorescent light bulbs, increasing the maximum
number installed to 10 should be considered. Also, lowering the number of hours of use from
three per day to as low as one half an hour per day should be considered; this would still be
cost effective for the average home.
7. Cost effectiveness acquisition of energy savings needs to be explicitly recognized by all
parties as one of the Program goals.
8. We do not believe that the energy education currently being provided by the agencies
justifies the cost of $200 per home. With the auditor performing the service at the same time
as the audit, $50 per home is more reasonable. To continue receiving $200 per home, it is
suggested that agencies have a separate employee on staff to provide energy education, and
that they should develop a model for providing energy education, attend training sessions,
and have a checklist of items to be covered. There are proven energy education approaches
that agencies need to follow.
9. All Program spending, including multiple funding sources, needs to be reported by the
agencies. In order to improve the tracking of costs, Pacific Power should replace the “other
funding” category and record each funding source in addition to the Program rebates. This is
a common practice, and makes business sense.
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 21 of 44
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 22 of 44

3. Impact Evaluation
Impact evaluation data were obtained from a number of different sources, including:
• Program measures: Pacific Power provided information regarding the Program’s
installed measures, including measure-specific saving estimates reported by the agencies
and installation dates.
• Billing records: Pacific Power provided participant and non-participant meter records
from July 2002 through June 2006. Non-participants were defined as households that
participated in the Low Income Home Energy Assistance Program (LIHEAP), but did not
receive weatherization.
• Weather data: Quantec collected weather data for the corresponding time period for both
Walla Walla and Yakima counties from the National Weather Service (NOAA).
• Contact information: For the purpose of conducting surveys, Pacific Power provided
Quantec with all available contact information, including name, address, and phone
number for participants.
Deemed Savings – Measure Analysis
Between July 1, 2003, and June 30, 2005 (the two-year Matchmaker period), the Program
provided service to 419 Pacific Power households. This represents a 35% reduction in
participation in comparison to the previous Program period, which reported 635 participants.
Interviews with the Pacific Power inspector and with the two agencies that exhibited a reduction
in participation resulted in the following explanations:
• There was an increase in the cost per measure.
• The previous program (2001-2003) had full Matchmaker funding, while the 2003-2005
Program had half of that funding.
• There was a suspension of Program activity while waiting for funds to be redistributed.
Most of the homes were completed in 2004 (59%). Over half of weatherization participants were
serviced by OIC. The average expected household savings was estimated at 3,249 kWh annually
based on deemed savings values. This compares to an estimated average household savings of
4,775 kWh reported in the previous Program evaluation. The lower savings value may represent
an improvement in the methods used for estimation, as the previous evaluation found that the
savings estimates were inflated. However, as is discussed in this report, the expected savings
continue to be overestimated.

Figure 10 shows the participation rate and energy savings across the three agencies.
Quantec — Washington Low-Income Weatherization Program 19
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 23 of 44
Figure 10. Participant and Savings, by Agency

The most frequently installed weatherization measures were double-glass replacement and
ceiling insulation, with 84% and 83% of all households receiving these services, respectively.
The frequency of installation of window replacement and the increase of these installations from
the last evaluation is unjustifiable, as they are rarely cost effective.
77
130
212
419
BMAC
NCAC
OIC
All agencies
1,806
4,530
2,987
3,249
,000
BMAC
NCAC
OIC
All agencies
verage Predicted kWh Savings
0 2 4,000 6,000
A
0 200 400 600
Participants
This is an issue that has been identified by the Pacific Power inspector and discussed with the
agencies prior to this evaluation, with consideration given to removing window replacement as
an option. The state is currently revising their specifications and as of Jan 1, 2007 will just pay
25% of the cost of replacing windows. Since Pacific Power pays up to 50%, this will likely mean
that windows will not be installed unless they are considered a repair.
Table 4 shows the frequency of weatherization measures installed at the households in this
Program period, as well as those reported in the last evaluation. Only 13 of the 23 measures
listed had attributed kWh savings. Additionally, it was found that the agencies’ reporting errors
would occasionally result in measures that were generally associated with an estimate of 0
energy savings. For example, installation of ceiling insulation had an estimated savings value of
0 kWh in 18 households. Double glass replacement was estimated by the agencies as having 0
kWh savings in 16 households.
Quantec — Washington Low-Income Weatherization Program 20
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 24 of 44

Table 4. Weatherization Measures and Frequencies
Measure 2003-2005 2001 - 2003 Evaluated
Measures
Double Glass Replacement 84% 66%
Ceiling Insulation 83% 82%
Air Sealing/Infiltration 58% 21%
Thermal Doors3
45% 56%
Fluorescent Lights 40% 46%
Floor Insulation 36% -
Faucet Aerators 36% -
Low Flow Shower Head 32% -
Wall Insulation 21% 25%
Refrigerator Replacement 20% 2%
Pipe Insulation 15% 4%
Water Heater < 50 Gallon 12% -
Duct Sealing 9% 87%
Water Heater Blankets 8% -
Weatherstrip Windows 5% 3%
Duct Insulation/Sealing 4% 6%
Ground Cover 3% .2%
Attic Ventilation 3% .3%
Dehumidifier 1% -
Water Heater > 60 Gallon 1% -
Weatherstrip Doors 1% 3%
Clock Thermostat 0.7% -

Base-load Measures
The previous evaluation recommended that an increased emphasis should be placed upon base-
load measures, which typically account for some 30% to 40% of total energy use. This
evaluation found that an increased emphasis had been made, resulting in the installation of
previously unevaluated measures such as energy efficient shower heads and faucet aerators,
refrigerator replacements, and water-heater improvements. However, it was found that the
number of fluorescent lights that were installed actually decreased. Continued emphasis should
be placed upon these measures.
Audit-Based Savings Estimation
As discussed in the Process Evaluation, the lookup tables used during the audit process result in
extremely overstated savings estimates.
Figure 11 shows the lack of relationship between the audit-based estimates of savings and what
actually occurred at the individual homes. One would expect some positive relationship between
the expected and the actual savings; i.e., as predicted savings increase, actual savings would also
increase. In a perfect world, this would be shown as a diagonal line between the x and y axis,

3 This was titled “Door Sealing” in the previous Program.
Quantec — Washington Low-Income Weatherization Program 21
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 25 of 44
but the relationship need not (nor will ever) be exact, due to behavioral factors that are beyond
the capability of the auditor to predict. In this case, however, there is not even a positive
relationship. This is a clear indication that these deemed numbers are invalid and should not be
used. The agencies must, in compliance with the contract requirements, use audit generated
savings. Furthermore, the actual consumption data provided by the Company needs to be used by
the auditors in determining expected savings by measure.
Figure 11. Actual versus Predicted Savings (kWh) – All Agencies
-10000
-5000
0
5000
10000
15000
0 5000 10000 15000 20000
Predicted Savings
Ac
t
u
a
l

S
a
v
i
n
g
s

Figures 11-13 illustrate the lack of correlation between deemed savings estimates and actual
savings at the agency level.

Figure 12. Actual versus Predicted Savings (kWh) – NCAC
-10000
-5000
0
5000
10000
15000
0 5000 10000 15000 20000
Predicted Savings
Ac
t
u
a
l

S
a
v
i
n
g
s

Quantec — Washington Low-Income Weatherization Program 22
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 26 of 44

Figure 13. Actual versus Predicted Savings (kWh) – OIC
-10000
-5000
0
5000
10000
15000
0 2000 4000 6000 8000 10000
Predicted Savings
Ac
t
u
a
l

S
a
v
i
n
g
s

Figure 14. Actual versus Predicted Savings (kWh) – BMAC
-10000
-5000
0
5000
10000
0 2000 4000 6000 8000
Predicted Savings
Ac
t
u
a
l

S
a
v
i
n
g
s

Actual Savings – Billing Analysis
Methodology
Pacific Power provided data regarding the Program’s 2003-2005 participants from 25 cities
throughout Washington. Data were assessed, organized, and subsequently filtered to obtain
complete customer profiles for evaluation using PRISM. (Princeton Scorekeeping Method).
PRISM was used to estimate weather-adjusted annual energy consumption based on energy
usage and outdoor temperature. In order to prepare the data for PRISM, several steps were taken.
Once the billing data contained only relevant meter readings, each participant’s profile was split
into pre and post periods based on the date his or her final weatherization measure was installed.
To ensure that any consumption changes that may have occurred during the weatherization
process itself were excluded from the analysis, any meter readings collected at the time of the
installations were excluded. The participants’ average completion date was then applied to all
non-participants, creating artificial pre and post periods for them. Applying this break in periods
allowed for the comparison of changes in post-weatherization energy consumption between the
two groups over similar time periods.
Quantec — Washington Low-Income Weatherization Program 23
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 27 of 44
In order to obtain accurate PRISM results, only participants and non-participants with a
minimum of twelve eligible months of both pre- and post-consumption data were utilized for the
analysis.
Lastly, the remaining participant and non-participant profiles were separated into Yakima and
Walla Walla files in final preparation for PRISM.
Sample Data Attrition
We required the use of twelve months of pre- and post-participation data in order to achieve
more reliable results. This eliminated 3 of the participants and 10 of the non-participants. In
order to ensure that billing data for non-participants were distinct from participant data, all past
participants of the Program were removed from the non-participant data set. Observations were
eliminated from the analysis on the following grounds:
• Outliers: Unreasonable consumption levels were defined as those lying outside the 1st
and 99th percentiles (less than 108 kWh or more than 4,433 kWh per month)
• Insufficient Data Points: Customers with less than twelve months pre-and post-Program
data
• Unable to Model: PRISM is often unable to effectively model households exhibiting
significant variance in consumption
Table 5 outlines data attrition. The final sample for the analysis contained 300 participants and
583 non-participants, 68% and 30% of the total sample, respectively.
Table 5. Sample Attrition
Participants Non-Participants
Removed Remaining Removed Remaining
Original sample 371 1,9274
Unable to obtain billing data 42 330 (89%) 1,927 (100%)
Outliers 66 264 (71%) 156 1,771 (92%)
Insufficient data points 3 261 (70%) 10 1,761 (91%)
Geographic limitations - 261 (70%) 1,161 600 (31%)
Unable to model with PRISM 8 253 (68%) 17 583 (30%)
Ineligible 1 252 (68%) - 583 (30%)
Sample adjustment5
48 (added)300 (81%) - 583 (30%)
Final sample 300 (81%) 583 (30%)

4 Non-participants who had received weatherization services in previous Program periods were removed from non-
participant sample
5 During the agency visits, it was discovered that predicted savings reported for one apartment were actually
predicted savings for a forty-nine-unit apartment building. One unit’s data was provided to Quantec and used in the
PRISM analysis. The savings for that apartment was subsequently attributed to all units.
Quantec — Washington Low-Income Weatherization Program 24
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 28 of 44

Overall Actual Savings
It was found that for the 300 clients with reliable consumption data, the gross savings were
estimated at 1,452 kWh, as shown in Table 6. This gross estimate does not include any
assessment of “what would have happened in the absence of the Program.” We employed a
comparison group of 583 clients who had received some form of energy assistance, but had not
participated in the Program. During the same time span, these non-participants increased their
consumption by 409 kWh, or 2.53% of their pre-Program consumption. We assume that, had our
participants not been through the program, they also would have witnessed an increase in
consumption of about 409 kWh. This process generates the net energy savings. Overall net
Program savings are estimated at 1,840 kWh per home.
Table 6. Savings Summary
Participants
(n=300)
Comparison Group
(n=587)
Pre NAC6 (kWh) 15,343 16,161
Post NAC (kWh) 13,891 16,570
Gross Savings (kWh) 1,452 -409
Percent Change 9.46% -2.53%
Net Impacts (kWh) 1,840
Savings as % of pre 12%
The analysis was also conducted by type of home, as shown in Table 7. The largest proportion of
participating homes was single family.
Table 7. Savings Summary by Home Type
Apartments
(n=85)
Manufactured
Home (n=86)
Single Family
(n=129)
Overall
(n=300)
Pre NAC 8,093 18,404 18,079 15,343
Post NAC 7,173 16,816 16,367 13,891
Savings 921 1,588 1,712 1,452
Net Savings 1,125 2,054 2,169 1,840
Net Savings as % of pre 14% 11 % 12% 12%

Overwhelmingly, apartments were the greatest energy savers for BMAC, as shown in Table 8,
though it should be noted that 49 of the 50 apartment units that were serviced were located in a
single apartment complex. Table 9 and Table 10 show the results for NCAC and OIC.

6 Normalized annual consumption (NAC) reflects temperature adjusted consumption levels.
Quantec — Washington Low-Income Weatherization Program 25
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 29 of 44
Table 8. BMAC Savings Summary by Home Type
Apartments
(n=50)
Manufactured
Home (n=9)
Single Family
(n=9)
Overall
(n=68)
Pre NAC 5,221 17,927 16,807 8,436
Post NAC 4,395 16,349 16,029 7,517
Savings 825 1,579 778 919
Net Savings 958 2,032 1,203 1,132
Net Savings as % of pre 18% 11 % 7% 13 %

Table 9. NCAC Savings Summary by Home Type
Apartments
(n=1)
Manufactured
Home (n=27)
Single Family
(n=68)
Overall
(n=96)
Pre NAC 15,053 18,666 19,418 19,161
Post NAC 13,157 16,571 17,613 17,274
Savings 1,897 2,095 1,804 1,887
Net Savings 2,278 2,568 2,296 2,372
Net Savings as % of pre 15% 14% 12% 12%

Table 10. OIC Savings Summary by Home Type (kWh)
Apartments
(n=34)
Manufactured
Home (n=50)
Single Family
(n=52)
Overall
(n=136)
Pre NAC 12,113 18,348 16,549 16,101
Post NAC 11,081 17,032 14,796 14,689
Savings 1,032 1,316 1,752 1,412
Net Savings 1,338 1,781 2,171 1,819
Net Savings as % of pre 11 % 10% 13% 11%

When the results are broken down by home type, as shown in the following tables, the average
savings between the agencies is noticeably variable. However, the sample size between the
agencies makes direct comparisons unrealistic.
Table 11 shows the energy savings associated with apartment buildings treated under the
Program. BMAC had a 49-unit apartment building weatherized, which dominates the calculated
savings for those units. NCAC has the only apartment that was weatherized and experienced
extreme savings. That apartment received the following measures: ceiling, floor, and pipe
insulation; a low-flow showerhead; fluorescent lights; thermal doors; and double-glass window
replacements.
Quantec — Washington Low-Income Weatherization Program 26
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 30 of 44

Quantec — Washington Low-Income Weatherization Program 27
Table 11. Apartment Energy Savings (kWh)

Apartments
n Maximum Minimum Mean Median
BMAC 50 1,724 942 958 942
NCAC 1 2,278 2,278 2,278 2,278
OIC 34 5,438 -3,937 1,338 1,553

Sample size again plays a role in examining the savings of manufactured homes, shown in Table
12. The savings for BMAC are based on nine homes, one-third of those weatherized by NCAC
and less than 20% of those weatherized by OIC. It is difficult to directly compare these saving
when the sample sizes are so disparate.
Table 12. Manufactured Home Energy Savings (kWh)

Manufactured Homes
n Maximum Minimum Mean Median
BMAC 9 8,888 -5,094 2,032 2,589
NCAC 27 11,210 -2,525 2,568 1,788
OIC 50 9,493 -4,938 1,781 1,651
The savings associated with single family households is more comparable between the agencies,
as shown in Table 13.
Table 13. Single Family Home Energy Savings (kWh)

Single Family Homes
n Maximum Minimum Mean Median
BMAC 9 4,855 -1,755 1,203 1,335
NCAC 68 12,151 -6,632 2,296 2,169
OIC 52 12,716 -6,082 2,171 1,335

ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 31 of 44
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 32 of 44

4. Non-Energy Benefits
The non-energy benefits of low-income programs can be quite numerous and significant. As well
as enabling positive change within the homes of participants, these benefits may have impacts on
the environment, local economies, and society as a whole.
Participant interviews were used to assess non-energy benefits at the household level in the form
of fewer work/school absences, the ability to remain in the home, fewer illnesses, more
disposable income, and increased comfort. Additionally, billing data were used to estimate the
Program’s impact on arrearages and participant mobility. Environmental and economic impacts
were estimated using appropriate software tools, as discussed further in this chapter.
Participant Impacts
In addition to the information discussed in the process evaluation, surveys with participants were
conducted to evaluate the non-energy benefits of the Program.
Energy Burden
Program participants reported having more money to spend on necessities and fewer absences
from school or work, as shown in Figure 15. Additionally, participants reported that they were
able to avoid moving as a result of the Program. Further analysis of mobility impacts was
performed using billing data, and is discussed further in this evaluation. Each of these benefits
exceeded those reported in the last Program period.
Figure 15. Program Impacts on Energy Burden, Mobility, and Absenteeism
43%
68%
83%
36%
37%
61%
Few er school/w ork
absences
Were able to avoid
moving
Have more money to
spend on other
necessities
2001-2003
2003-2005

Improved Comfort
When asked, nearly 94% of participants said they enjoyed increased comfort as a result of their
participation in the Program, presumably because they were able to have the heating and cooling
they desired. This is a slight improvement over the last Program period, as shown in Figure 16
below.
Quantec — Washington Low-Income Weatherization Program 29
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 33 of 44
Figure 16. Program Impacts on Increased Comfort
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2003-2005 2001-2003
Improved Health
For low-income families, critical needs may compete with very finite resources, resulting in
trade-offs which may compromise the participant’s health. As shown in Figure 17, 66% of
respondents reported fewer illnesses as a direct result of receiving weatherization services, which
was an improvement over the prior Program period. This may be due to the tangible benefits of
home repairs and weatherization services, as well as the avoidance of arrearage related shut-offs.
An analysis of the impacts on arrearages is presented further in this report.
Figure 17. Participants Reporting Fewer Illnesses
0%
10%
20%
30%
40%
50%
60%
70%
2003-2005 2001-2003
In addition, 12 (18%) respondents indicated that someone in their home suffers from asthma.
Five of these respondents indicated that instances of asthma-related events had decreased since
participation in the Program.
Mobility
When energy costs are high, household funds are diverted from other uses including food,
medical care, and rent. Our research has shown that in some cases, high-energy bills may force
Quantec — Washington Low-Income Weatherization Program 30
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 34 of 44

occupants to move out of their current dwelling either to lower energy costs or to avoid paying
an energy bill. In other cases, they may be evicted for inability to pay their rent or for having
services disconnected. Not only are frequent moves expensive and inconvenient, they have other
extremely serious effects. These may include increased school dropouts and inability to hold a
job. Energy assistance and weatherization programs lower the energy burden of the participating
low-income families and their forced mobility.7 Mobility can be especially hard for the elderly
and families with children. The value of reduced mobility can be as high as $1,460 per
household.8 In another national study, the cost of moving for low-income families was found to
be between 10% and 20% of annual income.9 These costs include moving expenses, rental
deposits, bank fees, telephone connections, etc. We follow a conservative approach of assuming
only $700 per move (less than mid-point of the Oak Ridge study and in line with Skumatz
(1998)).
Methodology
Using the same sample of participants that were selected for the impact and arrearage analysis,
mobility was assessed by using billing data to determine whether participants had moved into or
out of their Program weatherized home within two years (both before and after) of participation.
Results
As apparent in Table 14, in the two years prior to participating in the Program 70% (n=238) of
the participants matched to utility site level information (n=340) moved into the home that was
weatherized. However, in the two years following the completion of weatherization work, only
50% (n=170) moved from the weatherized home. Therefore, the weatherization work conducted
by the Program, and the lower energy bills that resulted, may have helped to prevent 68
participants from moving.
Table 14. Impact on Mobility

Pre (2 Years Prior to
Participation)
Post (2 Years Following
Participation)
Moved N % n %
Yes 238 70% 170 50%
No 102 30% 170 50%
Total 340 100% 340 100%
While other factors clearly contribute to the decision to move in either of the pre or post periods,
there is a significant difference in the proportion of participants that moved prior to and after
being weatherized. Using the conservative estimate of $700 per move noted above, and the
estimated 68 prevented moves, the Program generated $ 47,600 for participants.

7 Khawaja, M. (2001). Indiana REACH Evaluation. May. Portland, OR: Quantec, LLC.
In Indiana, as a result of participating in the Residential Energy Assistance Challenge Program, the participants
received energy education that lowered their energy consumption by 12.5%, reduced their mobility by 52%, and
reduced school absences by 18%.
8 Oak Ridge National Laboratory. (2002). Nonenergy Benefits from The Weatherization Assistance Program: A
Summary of Findings from the Recent literature. April.
9 Howat, J. & Oppenheim, (1999). Analysis of Low-Income Benefits in Determining Cost-Effectiveness of
Energy Efficiency Programs. http:www.consumerlaw.or/Energy/Energy&Utility/non_energy_benefits.htm
Quantec — Washington Low-Income Weatherization Program 31
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 35 of 44
Arrearage Impact
In addition to having an immediate impact on participants’ monthly energy bills, participation in
the Program can also lead to an overall reduction in arrearages. Simply put, as a result of the
reduction in monthly energy costs, participants are better able to put additional money towards
their outstanding arrearage.
Methodology
In order to determine the net impact of the Program upon arrears, data regarding both customer
bills and payments were collected for both participants and a select group of non-participants.
Table 15 details the sample attrition associated with the arrearage analysis.
Table 15. Sample Attrition: Arrearage Analysis
Participants Non-Participants
Removed Remaining Removed Remaining
Original Sample 371 (100%) 1,927 (100%)
Unable to Match to Utility Records (Participants Only) 31 340 (92%) --- ---
Missing Installation Date (Participants Only) 33 307 (83%) --- ---
Not included in Billing Analysis Sample* (Non-Participants Only) --- --- 1,327 600 (31%)
Lacked Sufficient Billing or Payment Data** 129 178 (48%) 49 551 (12%)
Dissimilar Pre Period Arrearage*** (Non-Participants Only) --- --- 329 222 (12%)
Outliers**** 24 154 (42%) 0 222 (12%)
Final Sample 154 (42%) 222 (12%)
* Prior to being analyzed in PRISM
** Minimum of 12 months meter and billing data in both pre and post period
*** Within ± 20% of the average pre participant arrearage level (defined as the percent of total bill paid by customer in pre period)
**** 1% and 99% tails of aggregated distribution

Results
Table 14 provides the results of the analysis. As evident in the table, participants arrears levels
dropped by $35 (from $207 to $172). Conversely, once the non-participants pre-arrearage values
were calibrated to the precise level of participants, it was determined the average non-participant
arrearage amount actually increased by $29 (from $207 to $236). As a result, the net impact of
the Program is a decrease in arrears by an average of $64.
Table 16. Program Impact: Arrearage Accumulation

Average Arrearage
Accumulated
During Pre Period
Average Arrearage
Accumulated
During Post Period
Change
Participants (n=154) $207 $172 $35
Non-Participants (n=222) $207 $236 -$29
Net Impact $64

Quantec — Washington Low-Income Weatherization Program 32
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 36 of 44

Economic Impacts
This type of program has several economic impacts in addition to direct benefits such as
decreasing the energy burden and increasing participants’ disposable income. As incomes
increase, so does spending on goods and services, leading to the creation of jobs. Weatherization
work itself is also a source of job creation. Additionally, the Program affects the economy in
several ways:
• It uses money from taxes and utility ratepayers to pay salaries and buy products used in
the weatherization process
• Participants have lower energy bills and are able to use the extra money to purchase
goods and services in other economic sectors
• Utilities receive less revenue due to lower energy bills for participants; this reduces the
need for new electricity generation facilities
Input-output modeling was used to quantify the effect of each of these monetary shifts
individually, as well as the impact on the Washington economy as a whole.10 This method of
modeling allows for an in-depth look at individual economic segments, as well as the effect that
the entire economy sees. The economy is represented as a matrix that relates industries to each
other so that effects of events can be tracked. In this case, these events are Program spending,
changes in household spending, reduced utility revenue, etc. When an event is specified, the
matrix tracks all direct, indirect, and induced effects on the economy. For example, the direct
effect of participants having lower energy bills is effectively an increase in household income.
The indirect effects are the redistribution of this income across the economy, thus creating more
jobs in the industries where households are spending money. These new jobs create another
increase in household income for the new employees and the induced effects are the
redistribution of this new income across the economy. For the purpose of this evaluation, direct,
indirect, and induced benefits have all been used to determine the benefits to the Washington
economy.
Because the funding to pay for Program activities ultimately comes from tax dollars, this was
modeled as a decrease to household income. This money is then distributed to certain industries
that provide the materials and labor for weatherization. Modeling participant utility bill savings
and utility lost revenue is somewhat more complex, because they do not completely offset one
another. Although the participants’ savings are equal to their full avoided utility payments, this
amount is not all lost revenues to the utility because reduced sales to customers are offset by the
amount that the utility reduces its purchases of required fuel or energy. Because the total energy
savings are small in comparison to total energy sales in Washington, it is assumed that this will
have no effect on ratepayers’ payments towards the utilities’ fixed costs, and that the portion of
rates that are fixed is lost revenue to the utilities.
Results
In total, it was estimated that the Program created about 6 net job-years of employment and
added $550,118 to the Washington economy. Though these numbers are small compared to

10 IMPLAN Professional 2.0 was used for this analysis, utilizing state-level data for Washington from 2002.
Quantec — Washington Low-Income Weatherization Program 33
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 37 of 44
Washington’s economy and work force as a whole, this analysis shows that the Program has a
positive effect on Washington’s economy.
Environmental Benefits
Reducing participants’ energy consumption also reduces the amount of pollution created by
electricity generation and fuel use. In order to determine the total amount of avoided pollution
and assign a dollar value to this environmental benefit, four steps were necessary:
1. Calculate the total Program kWh energy savings
2. Apply fuel mix specific to Pacific Power to determine the amount of fuel that was saved
because of avoided electricity demand
3. Use Clean Air and Climate Protection Software11 to calculate the avoided emissions
attributable to the Program
4. Obtain dollar values by pollutant to determine societal benefit
Results
Table 17 shows fuel saved through avoided electricity generation, based on the fuel mix specific
to Pacific Power. These are fuels that would have been necessary for the purpose of electricity
generation had the participant’s homes not been weatherized. The first column specifies the type
of fuel. The second column is the avoided annual electricity generated, while the third column
quantifies the annual fuel savings in the units commonly used for the respective fuel types. The
last column is the total fuel savings over the 30-year life of weatherization.
Table 17. Total Energy Savings by Fuel Type
Fuel Type12 Annual Avoided Electricity
Generated (kWh) Annual Fuel Savings Lifetime
Savings
Natural Gas 98,297 9,693 therms 193,852 therms
Fuel Oil 100 7 gal. 145 gal.
Coal 332,338 116 tons 2,327 tons

11 Developed and provided by the International Council for Local Environmental Initiatives (ICLEI), the State and
Territorial Air Pollution Program Administrators (STAPPA) and the Association of Local Air Pollution Control
Officials (ALAPCO)

12 Pacific Power’s fuel mix also includes nuclear, biomass, hydro, and other fuel types; however they do not
generate significant emissions.
Quantec — Washington Low-Income Weatherization Program 34
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 38 of 44

Quantec — Washington Low-Income Weatherization Program 35
Dollar values were assigned to the three most substantial air emission reductions based on
relevant market values as of August 2006, and are summarized in Table 18. In total, an
Environmental Benefit of $125,529 was estimated.13
Table 18. Avoided Emissions and Societal Benefits
Pollutant Lifetime Avoided
Emission (tons) Value Per Ton ($) Societal Benefit
(2003 $)
NOx 1,321 $1800 $35,936
SOx 7 $670 $70,880
Carbon Dioxide 302 $4.10 $18,713
Total $125,529
a Value from Seattle NOx price curve: August 18, 2006 b Value from Seattle SOx price curve: August 18, 2006
c Value from the Chicago Climate Exchange: August 22, 2006

13 CO2: Chicago Climate Exchange. “CCX is the world’s first, and North America’s only, voluntary, legally binding
rules-based greenhouse gas emission reduction and trading system.” (www.chicagoclimatex.com)
SOx and NOx: Evolution Markets Weekly Market Update. “Evolution Markets publishes a weekly report
covering U.S. SO2 and NOx emissions trading markets and global greenhouse gas emissions markets.”
(www.evomarkets.com)

ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 39 of 44
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 40 of 44

5. Cost-Effectiveness Analysis
As part of this evaluation, we conducted an economic analysis of the Program in accordance with
the benefit-cost tests from the California Standard Practices Manual. Program costs and benefits
were analyzed from the following perspectives:
• Total Resource Cost Test (TRC) – This test examines Program benefits and costs from
the perspectives of Pacific Power and Pacific Power’s customers. Benefits include
generation cost reduction, and costs include those incurred by Pacific Power as well as
additional funding from Matchmaker. A 10% conservation adder is applied to generation
cost savings in Washington.
• Utility Cost Test (UCT) – From Pacific Power’s perspective, benefits are in the form of
reduced generation and line loss costs. Costs include any incurred administrative or
measure costs.
• Ratepayer Impact Test (RIM) – All ratepayers (participants and non-participants) may
experience an increase in rates to recover lost revenue. This test includes all Pacific
Power Program costs as well as lost revenues. On the benefits side, this test includes all
avoided energy and capacity costs.
• Participant Cost Test (PCT) – This test examines benefits from a Program participant
perspective, including participant utility bill reductions. Costs include any measure costs
incurred by participants and the net of any utility-generated rebates. For this Program,
participants did not incur measure-related costs and did not receive any direct rebates.
They did, however, realize energy savings from the measures and their own energy-
saving behaviors.
The analysis results are presented in multiple ways, including:
• Levelized Cost per kWh – Cost of achieving each kWh of savings levelized over time.
The levelized cost per kWh can be compared to the cost of alternate resources to assess
the cost effectiveness of an efficiency investment.
• Net Present Value (NPV) – The difference between the discounted Program benefits and
costs. A net present value greater than zero would indicate that Program benefits exceed
costs.
• Benefit to Cost (B/C) Ratio – The ratio of Program benefits to Program costs. The
benefits and costs are determined over the life of the Program impact and discounted to
reflect the time value of money. A B/C ratio greater than 1.0 indicates that Program
benefits exceed costs.
With the exception of Matchmaker funding, which was incorporated into the TRC as described
above, cost data used for this analysis were limited to those provided by Pacific Power. Although
multiple attempts were made to secure total Program cost data from each of the agencies, it was
discovered that this information was not readily available.
Finally, the value of savings is determined by using various avoided cost scenarios. We used the
following Pacific Power forecasts of avoided costs in our analysis: Pacific Power’s official
market price forecast for Mid-Columbia, the base case of June 30, 2006, and Pacific Power’s
Quantec — Washington Low-Income Weatherization Program 37
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 41 of 44
updated IRP 67% load factor decrement. The IRP decrement represents the avoided cost as
determined by Pacific Power’s long-term resource plan. Table 19 shows the discount rates, line
loss, and residential rates used in our analysis.
Table 19. Pacific Power Discount Rates
Parameter Value
Discount Rate for TRC test 5.15%
Discount Rate for UCT, RIM, PART tests 8.74%
Line Loss 11.03%
Net Residential Energy Rate ($ per kWh) $0.0516
The cost-effectiveness analysis results are shown in Table 20 and Table 21. Under the Base Case
and IRP Decrement scenarios, the Program is not cost effective from any of the perspectives.
Table 20. Cost-Effectiveness Results: Base Case Forward Curves (Excluding Non-Energy
Benefits)
Perspective
Total
Discounted
Benefits
Total Costs Net Present
Value
Benefit to
Cost
Ratio
Levelized
Cost
($ per kWh)
Utility (UCT) $633,647 $1,526,475 -$892,828 0.42 $0.170
Participant (PCT) $575,505 $0 $575,505 $0.000
Ratepayer Impact (RIM) $633,647 $2,101,980 -$1,468,332 0.30 $0.234
Total Resource Cost (TRC) $990,189 $1,526,475 -$536,286 0.65 $0.170

Table 21. Cost-Effectiveness Results: IRP Decrement-Load Factor = 67% (Excluding Non-
Energy Benefits)
Perspective
Total
Discounted
Benefits
Total Costs Net Present
Value
Benefit to
Cost
Ratio
Levelized Cost
($ per kWh)
Utility (UCT) $557,514 $1,526,475 -$968,961 0.37 $0.170
Participant (PCT) $575,505 $0 $575,505 $0.000
Ratepayer Impact (RIM) $557,514 $2,101,980 -$1,544,466 0.27 $0.234
Total Resource Cost (TRC) $917,319 $1,526,475 -$609,156 0.60 $0.170
However, these results do not incorporate the non-energy benefits that were analyzed in this
evaluation, including the Program’s impact on forced mobility, arrearages, economic, and
societal impacts. These benefits are presented in Table 22.
Quantec — Washington Low-Income Weatherization Program 38
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 42 of 44

Table 22. Total Program Non-Energy Benefits
Non-Energy Benefit Program Impact Perspective Adjusted
Mobility $47,600 TRC
Arrearage $26,816 UCT, RIM, TRC
Economic $550,118 TRC
Environmental $125,529 TRC
Total $750,063
As with the previous evaluation, when these benefits are included in the analysis the Program
becomes more cost effective. As presented in Table 23 and Table 24, the Program passes TRC
with a benefit cost ratio of 1.06 and 1.01 respectively.
Table 23. Cost-Effectiveness Results: Base Case Forward Curves (Including Non-Energy
Benefits)
Perspective
Total
Discounted
Benefits
Total Costs Net Present
Value
Benefit to
Cost
Ratio
Levelized
Cost
($ per kWh)
Utility (UCT) $660,463 $1,526,475 -$866,012 0.43 $0.170
Participant (PCT) $575,505 $0 $575,505 $0.000
Ratepayer Impact (RIM) $660,463 $2,101,980 -$1,441,517 0.31 $0.234
Total Resource Cost (TRC) $1,716,674 $1,526,475 $88,271 1.12 $0.170
Table 24. Cost-Effectiveness Results: IRP Decrement-Load Factor = 67% (Including Non-
Energy Benefits)
Perspective
Total
Discounted
Benefits
Total Costs Net Present
Value
Benefit
to Cost
Ratio
Levelized Cost
($ per kWh)
Utility (UCT) $584,330 $1,526,475 -$942,145 0.38 $0.170
Participant (PCT) $575,505 $0 $575,505 $0.000
Ratepayer Impact (RIM) $584,330 $2,101,980 -$1,517,650 0.28 $0.234
Total Resource Cost (TRC) $1,643,804 $1,526,475 $15,401 1.08 $0.170

When normalized to the household level, the benefit-cost ratio is higher before the addition of
non-energy benefits when compared to the results of the previous Program evaluation, as shown
in Table 25.
Table 25. Normalized Cost Effectiveness Between Program Periods
Perspective (Base Case)
Normalized - Without Non-Energy
Benefits
Normalized - With Non-Energy
Benefits
Benefits Costs Benefit to
Cost Ratio
Benefits Costs Benefit to
Cost Ratio
TRC 2003 Program $2,502 $4,606 .54 $5,652 $4,606 1.23
TRC 2005 Program $2,363 $3,643 .65 $3,854 $3,643 1.06

While the addition of non-energy benefits raises each benefit-cost ratio substantially, the 2005
Program has a slightly lower ratio than what was reported for the 2003 Program. This may be a
result of the more detailed assignment of non-energy benefits in the current evaluation, which
Quantec — Washington Low-Income Weatherization Program 39
ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 43 of 44
Quantec — Washington Low-Income Weatherization Program 40
utilized billing data to analyze Program impacts. This was not within the scope of work for the
previous evaluation.

ID PAC-E-11-13
CAPAI 2 Attachment CAPAI 2
Attach CAPAI 2.pdf Page 44 of 44