20190701Ehrbar Exhibit 1.pdf

DAVID J. MEYER
VICE PRESIDENT AND CHIEF COUNSEL FOR
REGULATORY & GOVERNMENTAL AFFAIRS
AVISTA CORPORATION
P.O. BOX 3727
14I1 EAST MISSION AVENUE
SPOKANE, WASHINGTON 99220 -37 27
TELEPHONE: (s09) 49s-43t6
FACSIMILE: (509) 495-885 I
DAVID.MEYER@AVIS TAC ORP. COM
iilIg JUL : I
JT !T
BEFORE THE IDAHO PUBLIC UTILITIES COMMISSION
IN THE MATTER OF THE APPLICATION
OF AVISTA CORPORATION FOR THE
EXTENSION OF AVISTA'S ELECTRIC
AND NATURAL GAS FXED COST
ADJUSTMENT MECHANISMS IN THE
STATE OF IDAHO
)
)
)
)
)
)
)
CASE NO. AW-E-19-0_k
CASE NO. AVU-G-19-03
EXHIBIT NO. 1
FOR AVISTA CORPORATION
(ELECTRIC AND NATURAL GAS)
RECEIVED
l0:28
L it,rssloid
I I
H. Gil Peach& Associates LLC
AVISTA DECOUPLING EVALUATION
Final Report
6
I
W":4'." -*-,, - i
-a
Hugh Peach - H. Gil Peach & Associates LLC
Mark Thompson - Forefront Economics Inc.
John Joseph - Joseph Associates, Inc.
10/0 U20r8
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
H
1 o1224
6
Vision Statement
To be a world leader in developing truthful measurement and useful results; to support
development of efficient, ethical, and effective practices, sustained economically; to advance
human development. To improve the quality of life during the era of climate change.
a
Goals Statement
To build inclusion, diversity and social justice in support of all technical goals.
Inclusion, diversity and social justice is the top technical goal.
Excellence in the integration of knowledge, method, and practice
Improvement and leaming at all levels
Contextually sound measurement, analysis, and reporting
Anticipate and meet the needs of our clients
Awareness of human relevance and of the ethical core of research
To go further, to find better ways
a
a
a
a
a
a
Mission Statement
With extensive experience in North America we can provide the fuIl range of evaluation,
verification, policy, management, planning, regulatory and adaptation services - wherever and
whenever there is a need.
Environmental Policy Statement
Collectively, we are at a Darwin moment. Either we move to a better model for production;
work intensely to mitigate climate change; anticipate and actualize inclusive climate
adaptation - or we face being edited out of history.
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
6
Table of Contents
Table of Contents........................I
I
1-1
Decoupling Mechanism -2015 Electric (Schedule 75) and Gas (Schedule 175)t-2
Electric Group I (Residential) and Group 2 Q,{on-Residential)...........
Natural Gas Group I (Residential) and Group 2 Qtron-Residential)........... t-11
l-19
.1-19
I -20
t-20
I -20
1-23
t-26
Electric Group I (Residential) and Group 2 (Non-Residential)........
Natural Gas Group I (Residential) and Group 2 (Non-Residential).
....I -26
....1-34
Schedule 75E - Electric 30% Rate Increase Test.......
t-43
t-43
1-45
t-47
1-47
t-48
t-49
t-49
1-57
r-66
t-66
t-68
t-70
I -70
1-7 I
1-72
l-74
Schedule l75E - Natural Gas 3% Rate Increase Test....
Decoupling Mechanism - 2017 Electric (Schedule 75) and Natural Gas (Schedule 175)........
Electric Group 1 (Residential) and Group 2 (Non-Residential)
Natural Gas Group I (Residential) and Group 2 Qtlon-Residential).......................
.......... I -3
Schedule 75D - Electric Earnings Test...............
Schedule l75D - Natural Gas Earnings Test......................
2015 Three-Percent Annual Rate Increase Limitation
Schedule 75E - Electric 30% Rate Increase Test................
Decoupling Mechanism - 2016 Electric (Schedule 75) and Natural Gas (Schedule 175)........
Schedule 75D - Electric Earnings Test...............
Schedule l75D - Natural Gas Earnings Test.................
2016 Three-Percent Annual Rate Increase Limitation
2017 Eamings Test
Schedule 7 5D - Electric Earnings Test .... ... ........
Schedule l75D - Natural Gas Earnings Test......
2017 Three-Percent Annual Rate Increase Limitation
Schedule 75E - Electric 30% Rate Increase Test
Schedule l75E - Natural Gas 3% Rate Increase Test......................
Audit Statements: Is the Source Data Credible?
Summary - Task I
Section 2. Billing Impacts and Recovery of Cost of Service Analysis....2-l
aaL-LSummary of Decoupling Mechanics and Results
Factors In/luencing Use per Customer.
Weather Compared to Normal ............,
Task 2 Part l: Impact of Decoupling Tracker Adjustment by Customer Class...........
....2-4
....2-9
.2-tt
2-1 I
2-16
Task 2 Part2: Are Allowed Revenues Recovering Cost of Service by Rate Group?a1a
2-24
2-24
Summary - Task 2 ..2-28
Section 3. Low-Income Analysis and Contrasts............. ......3-1
Low-Income Billing Impacts (includes Parts A and D)
Impact on Electric Low-Income Customers
Impact on Natural Gas Low-Income Customers ..
3-Z
.3 -3
)-/
Summary - Task 3, Parts A and D,
Low-Income Savings, Expenditures and Customers Served.............................................. .Exhihit.No.l......3-10
Page i
Page 3 of 224
............3- I 0
C ons erv ation Pr o gram S avings..
Low-Income Bill Assistance
Low-Income Rate Assistance Program (LIRAP) .....
Rate Discount Pilot Programfor Seniors
Low-Income Home Energt Assistance Program (LIHEAP)
Proj ect 9hare...............
Miscellaneous B ill As s is tance ..............
Bill Assistance Funding Trends...........
Number of Bill Assistance Grants
Average Bill Assistance Grant.....,
Low-Income Ll/eatherization Seryices
Avista Low-Income Weatherization Funding .......
Number of Low-Income W'eatherization Grants .....
Average Weatherization Job Costs...
Inflation Adjusted Funding Levels
Summary - Task 3, Part 8............
Modifications to Low-Income Programs
Effect on Low-Income vs. Average Residential..
Other Factors
Overview of Approach..
3-10
3-1 I
3- 12
3- 16
3-t6
3- 16
3-17
3-18
3-19
3-20
3-2 I
3-22
3-2 5
3-2s
3-26
3-27
Energt Usage ...........
3-28
3-31
3-31
.3-3 t
.3-33
.3-34
.3-35
.3-36
H ous ing C haracterist ics.........
Housing Type and HVAC Equipment.
Summary - Task ie..........
Section 4. Analysis of Revenue Effects 4-l
.4-l
.4-3
.4-8
.4-9
Has Decoupling Stabilized Revenue
Revenue Deviations from Planning Assumptions and Causes
Review of Rate Cap and Earnings Test
Review of Defenals
Summary - Task 4..
Section 5. Fixed Costs and Charges, Non-Decoupled Customers .........5-1
Electric Customers..5-l
5-2
5-3
Section 6. Analysis of Conservation Achievement...... ..........6-1
What is the Impact of Fuel Conversion on Decoupling Revenue?
Summary - Impact of Fuel Conversion on Decoupling Revenue
Have the Mechanisms had an Impact on Natural Gas or Electric Conservation?...............
Decoupling and Conservation Achievement (Totals) : P erspective
Residential Electric Group......
Low-Income Electric Group....
Nonresidential Electric Group
Residential Natural Gas Group
Low-Income Natural Gas Group...............
Nonresidential Natural Gas Group ...........
..6-10
..6- 12
..6-13
..6-14
..6-15
..6-t6
6-2
6-6
6-7
6-7
7-l
7-)
7-8
Summary - Impact on Conservation Achievement............ ............6-16
Section 7. Analysis of Possible Adverse Impacts 7-l
Customer Service and Service Quality Indices (5Q0.......
Cost Control and Operational Efficiency
Summary - Task 7 (Adverse Impacts) ..
..........7- I 2
o
Page ii case Nos. AVU-E-19-0_ and AVU-G-I9-0_
P. Ehrbar, Avista
Page 4 ot 224
6
Section 8. Low-Income Appendix 8-1
Attachment G - Estimate of the Number of Households in Povertv 8-1
8-2
8-5
8-9
Attachment H - The Self-Sufficiency Standard for Washington State 2014
Making Sense of Federal Poverty Level vs. Income Insufficiency
Understanding Low-Income within the Overall Allocation of Income ... ..........8- I I
Seetion 9. Weather Appendix. ............9-1
Section 10. Recommendations................ ..............10-1
Exhibit No. 1
uase Nos. AVU-E-l9-U_ and AVU-U-I9-U_
P. Ehrbar, Avista
Page 5 of 224
Page iii
6
List of Tables
Table 1-1. GeneralRateCaseandTestYearDefinitionsbyDeferralYear........... ....................1-2
Table l-2. 2015 Development of Electric Decoupled Revenue per Customer.................. .........1-5
Table l-3. 2015 Electric Decoupled Revenue per Customer.................. ................1-6
Table l-4. 2015 Development of Monthly Electric Decoupled Revenue per Customer t-7
Table l-5. 2015 Electric Deferral Calculations..l-8
Table l-7. 2015 Development of Natural Gas Decoupled Revenue per Customer l-l 3
t-14
1- 15
1-16
1- 18
1-19
t-19
Table l-8. 2015 Natural Gas Decoupled Revenue per Customer
Table 1-9. 2015 Development of Monthly Natural Gas Decoupled Revenue per Customer..................
Table l-10. 2015 Natural Gas Deferral Calculations
Table l-11. 2015 Development of Natural Gas Deferral
Table l-12. 2015 Electric and Natural Gas Eamings Tests
Table 1-13. 2015 Electric Earnings Test Sharing Adjustment
Table l-14. 2015 Electric 302Incremental Surcharge Test -2t
Table l-15. 2015 Residential Electric Carryover Deferred Revenue......aa
Table l-16. 2015 Natural Gas 3% Incremental Surcharge Test..-23
Table l-17. 2015 Residential Natural Gas Carryover Deferred Revenue -24
Table I - I 8. 201 5 Non-Residential Natural Gas Carryover Deferred Revenue t-25
t-28
r-29
l-30
t-32
r-33
l-36
t-37
l-38
t-41
r-42
t-43
r-44
t-44
t-45
r-46
t-46
t-47
t-48
l-51
Table l-19. 2016 Development of Electric Decoupled Revenue per Customer
Table l-20. 2016 Development of Electric Decoupled Revenue per Customer
Table l-21. 2016 Development of Monthly Electric Decoupled Revenue per Customer
Table l-23. 2016 Development of Electric Deferral..
Table l-24. 2016 Development of Natural Gas Decoupled Revenue per Customer
Table l-25. 2016 Natural Gas Decoupled Revenue per Customer................
Table l-26. 2016 Development of Monthly Natural Gas Decoupled Revenue per Customer..................
Table l-27. 2016 Natural Gas Deferral Calculations
Table l-28. 2016 Development of Natural Gas Deferral
Table l-30. 2016 Electric Earnings Test Sharing Adjustment.
Table I -3 1 . Derivation of 2016 Electric Gross Up Factor and Revenue Conversion Factor
Table l-32. 2016 Natural Gas Eamings Test
Table l-33. Derivation of 2016 Natural Gas Gross Up Factor and Revenue Conversion Factor.........
Table l-34. 2016 Natural Gas Earnings Test Sharing Adjustment
Table l-35. 2016 Electric 302Incremental Surcharge Test
Table l-36. 2016 Natural Gas 3% Incremental Surcharge Test..
Table l-37. 2017 Development of Electric Decoupled Revenue per Customer..................
Table l-38. 2017 Electric Decoupled Revenue per Customer..................-52
Table l-39. 2017 Development of Monthly Electric Decoupled Revenue per Customer -53
-55
Table l-41. 2017 Development of Electric Deferral..-56
Table l-42. 2017 Development of Natural Gas Decoupled Revenue per Customer -59
Table 1-43. 2017 Natural Gas Decoupled Revenue per Customer -60
Table l-44. 2017 Development of Monthly Natural Gas Decoupled Revenue per Customer..................-61
Table 1-45. 2017 Natural Gas Deferral Calculations -64
Table 1-46. 2017 Development of Natural Gas Deferral..................-65
-66
Table 1-48. 2017 Electric Earnings Test Sharing Adjustment -67
Table l-49. 20ll Derivation of Electric Gross Up Factor and Revenue Conversion Factor.........-68
Table l-50. 2017 Natural Gas Earnings Test -68
Table l-51. 2017 Derivation of Gross Up Factor and Revenue Conversion Factor (Natural Gas)............Exhibii.No1.T...
P. Ehrbar, Avista
Page6 ol 224
Page iv
-69
uase t\os. t\vu-tr- tY-u_ ana r\vu-\)- tY-u_
6
Table l-52. 2017 Natural Gas Earnings Test Sharing Adjustment.
Table l-53. 2017 Electric3Yolncremental Surcharge Test
Table 1-54. 2017 Natural Gas 3Yo Incremental Surcharge Test.............
Table 2-1. Electric and Natural Gas Rate Groups and Customer Classes (Rate Categories)
Table2-Z. Summary of Deferral Balances and Decoupling Recovery Rates
Table2-3. Electric Use per Customer Variance from Test Year
Table2-4. Natural Gas Use per Customer Variance from Test Year
Table 2-5. Comparison of Actual and Normal Annual Heating Degree Days
Table 2-6. Annual Electric Customer Counts by Customer Class ...........
Table 2-7. Annual Electric Revenue by Customer Class
Table2-8. Annual Decoupling Tariff Revenue by Electric Customer Class
T able 2-9. 20 1 6 Electric Monthly Billing Data ............
Table 2-10. 2017 Electric Monthly Billing Data
Table 2-11. Annual Natural Gas Customer Counts by Customer Class
Table 2-12. Annual Natural Gas Revenue by Customer Class
r-69
r-70
t-71
..2-l
..2-3
2-5
2-7
2-10
2-tt
.2-12
.2-12
.2-14
.2-15
.2-17
Table 2-13. Annual Decoupling Tariff Revenue by Natural Gas Customer Class
Table2-14. 2016 Natural Gas Monthly Billing Data.
Table 2-15. 2017 Natural Gas Monthly Billing Data.
Table 2-16. Electric Revenues and Cost of Service by Rate Group (thousands of dollars)
Table 2-17 . Natural Gas Revenues and Cost of Service by Rate Group (thousands of dollars) ...
Table 3-16. Low-Income 100% Approved Measures (2015)
Table 3-17. Low-Income Partial Rebate Measures (2015)
Table 3-18. Low-Income 100% Approved Measures (2016)
Table 3-19. Low-Income Partial Rebate Measules (2016)
Table 3-20. Low-Income 100% Approved Measures (2017)
Table 3-21. Low-Income Partial Rebate Measures (2017)
Table 3-22. Avista Customer Counts by Residential Group and Service Type
T able 3 -23. Comparison of Housing Characterir;tics ...............
Table 3-24. Distribution of Housing Types
Table 3-25. Distribution of Heating Equipment
Table 3-26. Distribution of Cooling Equipment
Table 4-1. Authorized and Actual Electric Decoupled Revenue per Customer
Table 4-2. Test Year and Actual Electric Usage, Customers and Use per Customer
Table 4-3. Authorized and Actual Natural Gas Decoupled Revenue per Customer..................
..2-17
..2-18
..2-19
..2-20
..2-26
..2-27
Table 3-1. All Residential and Low-Income Electric and Natural Gas Customer Counts..J-J
Table 3-2. Comparison of Average Annual Electric Revenue per Customer.................. ...........3-4
Table 3-3. Monthly Electric Usage, Meters and Revenue, Low-Income and All Residential 3-6
Table 3-4. Comparison of Average Annual Natural Gas Revenue per Customer 3-8
Table 3-5. Monthly Natural Gas, Meters and Revenue, Low-Income and All Residential 3-9
Table 3-6. Total Electric Energy Savings - Coru;ervation and Conversions (kwh).......... ........3-10
Table 3-8. Electric Conversion to Natural Gas S,avings (kwh) ........3-l I
Table 3-9. Percentage Electric Savings Due to Conversions from Electric to Natural Gas.............. ............3-11
Table 3-10. Total Natural Gas Conservation Sa.rings (therms) ........3-11
Table 3-14. Electric Service DSM 3-22
3-23
3-29
3-29
3-29
3-30
3-30
3-30
3-55
3-34
3-35
3-36
3-36
..4-3
..4-3
Table 4-4. Test Year and Actual Natural Gas Usage, Customers and Use per Customer
Table 4-5. Earning Test Shared Revenue.
Table 4-6. History of Rate Cap Results - Was Rate Cap Reached?.................. ......4-9
Table 4-7. Electric Revenue from Decoupled Rate Groups........ ....Exhibit.No..:t........4-9
P. Ehrbar, Avista
PageT ot224
.4-5
.4-6
.4-8
Page v
6
Table 4-8. Natural Gas Revenue from Decoupled Rate Groups .......4-10
Table 4-9. Summary of Deferral Balances and Decoupling Recovery Rates..........
Table 4-10. Deferred Revenue at Normal Weather
Table 5-1. Electric Revenue from Fixed Charges and Fixed Cost (thousands of dollars)
Table 5-2. Fixed Cost and Fixed Charges, Non-Decoupled Natural Gas Customer Classes....
Table 6-1. Electric Residential Conversions as Percentage of Conservation Achievement (kwh).........
Table 6-2. Electric Non-Residential Conversion as Percentage of Conservation Achievement (kwh)
Table 6-3. Residential Fuel Conversion Program Savings
Table 6-4. Allocation of Nonresidential Revenue based on Gross Verified Savings (kV[h).........
Table 6-5. Residential Gas Decoupling Revenue Based on Gross Verified Savings (therms)
Table 6-6. Nonresidential Gas Decoupling Revenue (Gross Verified Savings - therms)
Table 7-1. 2015 Indicators of Customer Service Quality - DR 52
Table 7 -2. 201 6 Indicators of Customer Service Quality - DR 52
Table 7-3. 2017 Indicators of Customer Service Quality - DR 52
Table 7-4. Customer Service Indicators for Before and After Decoupling - DR 52
Table 7-5. Indicators of Electric Service Reliability - DR 52
TableT-6. 2016 Customer Service Guarantee - DR 52
Table 7 -7 . 201 7 Customer Service Guarantee - DR 52
Table 7-8. Electric Decoupling Signal as Percentage of Average Bill for Calendar 2016
Table 7-9. Natural Gas Decoupling Signal as Percentage of Average Bill for Calendar 2016.
Table 7-10. Electric Decoupling Signal as Percentage of Average Bill for Calendar 2017
..4-10
..4-tt
5-1
5-2
6-3
6-3
6-4
6-5
6-s
6-6
7-3
7-4
7-4
7-5
7-6
7-7
7-9
7-9
Table 7-l L Natural Gas Decoupling Signal as Percentage of Average Bill for Calendar 20ll............
TableT-12. Rate of Retum vs. Cost of Capital - DR 066, Revised, Attachment A
Table 8-1. 150% of Poverty or Less - Receiving Bill Assistance or Avista Weatherization
Table 8-6. Self-Sufficiency Standard Spokane County (2001 vs.20l7)
Table 8-7. 150% ofFPL vs. Self-Sufficiency Standard, Spokane County,2001
Table 8-8. 150% ofFPL vs. Self-Sufficiency Standard, Spokane County,20l7
Table 9-1. Model Summary and Parameter Estimates
Exhibit No. 1
7 -tt
1-tl
7 -17
Table 8-2. Self-Sufficiency Standard Expressed as a Percentage ofPoverty
Table 8-3. Results at200o/o Poverty based on American Community Survey
Table 8-4. Monthly Costs included in the Self-Sufficiency Standard - Spokane 2014............ .......................8-7
Table 8-5. 150% Poverty Guidelines (2001 vs. 2017)8-8
8-2
8-3
8-5
..8-8
..8-8
8-8
9-4
Cese Nos.TVUI-F1IFO'ead AVU-G-1 9-0
P. Ehrbar, Avista
Page 8 of 224
Page vi
6
Table of Figures
Figure 1-1. Timing of Deferral Balance Accumulation and Decoupling Rate l-l
Figure l-2. 2016 Financial Audit Opinion for Calendar 2015
Figure l-3. 2017 Financial Audit Opinion for Calendar 2016
Figure l-4. 2018 Financial Audit Opinion for Calendar 201'7
Figure 2-1. Timing of Deferral Balance Accumulation and Decoupling Rate 2-3
Figure 2-2. Percentage Change in Use per Customer, Electric Residential 2-5
Figure 2-3. Percentage Change in Use per Customer, Electric Non-Residential 2-6
Figure 2-4. Percentage Change in Use per Customer, Natural Gas Residential.............. ...........2-7
Figure 2-5. Percentage Change in Use per Customer, Natural Gas Non-Residential 2-8
t-72
t-73
t-73
Figure 2-6. Monthly Heating Degree Days (difference from normal)
Figure 2-7. Monthly Cooling Degree Days (difference from normal)
2-9
Figure 2-8. Schedule 75 as a Percent of Monthly Customer Class Revenues
Figure 2-9. Schedule 175 as a Percent of Monthly Customer Class Revenues
Figure 3-1. The Parts of Task 3
Figure 3-2. Annual Electric Use per Customer, Low-Income and All Residential..
Figure 3-3. Comparison of Average Monthly Electric Schedule 75 Revenue per Customer
Figure 3-4. Annual Natural Gas Use per Customer, Low-Income and Average Residential
Figure 3-5. Comparison of Average Monthly Natural Gas Schedule 175 Revenue per Customer....
Figure 3-6. Electric Service LIRAP Tariff (Weighted Average)................ ..........3-13
Figure 3-7. Natural Gas Service LIRAP Tariff (Weighted Average) ................ ...3-15
Figure 3-8. Value of All Bill Assistance Grants
Figure 3-9. Value of Bill Assistance by Funding Source.
Figure 3-10. Number of Bill Assistance Grants Provided
Figure 3-11. Number of Bill Assistance Grants by Funding Source
Figure 3-12. Average Bill Assistance Grant by Funding Source
Figure 3-13. Electric Service DSM Tariff (Weighted Average).
Figure 3-14. Natural Gas Service DSM Tariff (Weighted Average)
Figure 3-15. Avista Low-Income Weatherization Funding Trends..
Figure 3-16. Number of Low-Income Weatherization Grants
Figure 3-17. Average Cost of Weatherization Jobs
Figure 3-18. Inflation Adjusted Bill Assistance (All Sources)
Figure 3-19. Inflation Adjusted Avista Weatherization Funding
Figure 3-20. Annual20lT Unadjusted Billed Energy Usage per Premise.
Figure 3-21. Annual2DlT Unadjusted Billed Energy per Square Foot............ ....3-34
Figure 4-1. Electric Revenue Variability (2015-2017)
...2-10
...2-16
...2-21
3- l8
3- l8
3-r9
3-20
3-21
3-22
3-23
3-24
3-25
3-2s
3-26
3-26
J-JJ
Figure 4-2. Natural Gas Revenue Variability (2015-2017)
Figure 4-3. Percentage Change in Use per Customer, Electric Residential
Figure 4-4. Percentage Change in Use per Customer, Electric Non-Residential
Figure 4-5. Percentage Change in Use per Customer, Natural Gas Residential.
Figure 4-6. Percentage Change in Use per Customer, Natural Gas Non-Residential
Figure 6-1. Task 6 - Conservation Achievement...........
Figure 6-3. Conservation Achievement - Residential E1ectric....... ...6-11
Figure 6-4. Conservation Achievement - Low-Income Electric .......6-12
Figure 6-5. Nonresidential Electric Conservation Achievement (MWh) .............6-13
Figure 6-6. Residential Natural Gas Conservation Achievement...
Figure 6-7. Low-Income Natural Gas Conservation Achievement.
Figure 6-8. Nonresidential Natural Gas Conservation Achievement.
Figure 7-2. Increasing Earnings in a Decoupled Utility (RAP)" " " " Exhiblt'N - ; 1" " "7 -12
3-l
J-J
3-5
3-7
3-8
4-2
4-2
4-4
4-5
4-7
4-7
6-t
6-8
Figure 6-9. Regulatory Assistance Project on Decoupling
6-14
6- l5
6-16
6-17
..7-l
uase Nos. AVU-E-I9-U_ ano AVU-G-I9-U_
P. Ehrbar, Avista
Page I of 224
Page vii
0
Figure 8-1.
Figure 8-2.
Figure 8-3.
Figure 8-4.
Figure 9-1.
Figure 9-2.
Figure 9-3.
Figure 9-4.
Figure 9-5.
Figure 9-6.
Figure 9-7.
Figure 9-8.
Variation of Self-Sufficiency Standard across Washington Counties 8-4
Historical Divergence of BLS CPI (Courtesy of ShadowStats.com) .......................8-9
Income Donut for Washington State (Census 1990)
Income Donut for Washington State (Census 2000)
8-1 1
8-t2
..9-1
..9-2
..9-2
..9-3
Drought Conditions
Wildland Fire Potential Outlook
Pattem of Heating Degree Days (Spokane)
Pattern of Cooling Degree Days (Spokane)
Regression of HDD on Year ........................9-5
Year in which HDD : Zero is Reached, Using different Numbers of Analysis Years.......... ......9-5
Years from 2018 until Zero HDD, using different Numbers of Analysis Years.......... ................9-6
Thirfy Year Average vs. Other Averages for HDD 9-7
Exhibit No. 1
P. Ehrbar, Avista
Page 10 of 224
Page viii
Introduction
This evaluation of Avista's Decoupling Mechanisms is partly a compliance evaluation and partly
a policy evaluation of Avista's decoupling as a specific rate reform (alternative form of rate
making) within a specific window of time.
The skucture of the evaluation is in section. Each section from Section I through Section 7
corresponds to a specific task (Task 1 through Task 7).
o Section 1 is a compliance evaluation: Did Avista comply with the specifics of the
decoupling order?
o Section 2 is concerned with billing impacts and recovery of cost of service.
o Section 3 is focused on low-income customers and contrast between low-income and
residential customers generally.
o Section 4 analyzes overall revenue effect.
o Section 5 examines fixed costs and charges for non-decoupled customers.
o Section 6 is an analysis of conservation achievement.
o Section 7 examines possible adverse impacts of decoupling.
o Section 8 is an appendix on a more extensive analysis of low-income customers.
o Section 9 is an appendix on the effects of weather.
o Section 10 covers evaluation recommendations.
We find that Avista's decoupling is working well within the specific window of time examined.
Exhibit No. 1
6
Page I Case Nos. AVU-E-19-0_ ancl AVU-G-19-0_
P. Ehrbar, Avista
Page11 o1224
6
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 12 of 224
Page 2
6
For this analysis, the evaluation objective is to complete a review of whether the deferrals and
rates were calculated in accordance with the Commission order approving the mechanisms. Or,
in other words, were the mechanisms administered and calculated correctly, per the Amended
Petition? This first task is an assessment of compliance. Operationally, we compare the
Decoupling Mechanism Development of Deferrals as submitted by Avista in20l6 for the 2015
deferral yearr, as submittedin2}lT for the 2016 deferralyear, and as submitted in 2018 for the
2017 defenal year to the specification of method in Schedule 75 (75,75A,758,75C,75D,758)
for electric service and in Schedule 175 (175,175A,7758,175C,175D,175E) for natural gas
service. This includes the Earnings Test and the3Yo Annual Increase Test.
In order to facilitate and order discussion, it will be useful to define decoupling deferral years
and rate years as shown in Figure 1-1.
Time JFMAMJJASOND JFMAMJJASO ND JFMAMJJASO ND JFMAMJJASO ND
Deferra I
Year
Deferral Year 1 Deferral Year 2 Deferral Year 3 Deferral Year 4
Rate
Year
Rate Year 1 Rate Year 2
/-
\-
Figure I - I . Timing of Deferral Balance Accuntulalion and Decoupling Rote
The timing of deferral balance accumulation and decoupling rate adjustments is shown in Figure
1-1. Avista's decoupling mechanism allows for the recovery of the difference between actual
revenue and allowed revenue.2 This difference is referred to as the decoupling deferral balance
and is tracked for the two electric and two natural gas customer groups subject to decoupling;
residential and non-residential.
Beginning in 2015, monthly deferrals are accumulated over a calendar year and used with other
determinants to calculate the decoupling rate required to collect or refund the under or over
collected revenue. Decoupling rates become effective in Schedule 75 (electric) and Schedule
175 (natural gas) November 1 of the year following the year in which deferral balances were
calculated.
lPat Ehrbar, Sr. Manager, Rates and Tariffs to Mr. Steven King, Executive Director and Secretary, Washington
Utilities and Transportation Commission, August 31,2016, with attachments.
2 The details of Avista's decoupling mechanism are included in Final Order ("Order 5") for Docket Numbers UE-
140188 and UG-140189.
Exhibit No. 1
P. Ehrbar, Avista
Page '13 ot 224
Page l-l
Section 1. Fidelitv Analysis
o
The first deferral year resulted in a deferral balance at the end of 201 5 that was used, along with
other determinants, to calculate the decoupling rate in effect during the first rate year (November
1,2016 through October 31,2017). The same process is followed in the second deferral year
and rate year. Any deferral balance carried over from the first rate year due to the application of
the 3oh cap is included in the calculations of decoupling rates in effect during the second rate
year (November 2017 through October 2018). Each year, electric and natural gas results are
separately developed. Also, within each year and energy source, Residential and Non-
Residential Rate Groups are separately analyzed.
It is also useful to understand the test year in effect during each deferral year. Table l-l shows
test year definitions used in each general rate case (GRC).
Table l-1. General Rate Case and Test Year Definitions by Deferral Year
Item Electric Natural Gas
Defenal Year 20t5 2016 &2017 201 s 2016 &20t7
General Rate Case uE-140188 uE-150204 uG-140189 uG-150205
Test Year Proforma 2015 Oct 2013-Sep 2014 Proforma 2015 Oct 2013-Sep 2014
In the first decoupling deferral year (201 5) the decoupling mechanism used a forecast of 201 5
customers, usage and revenue as the test year. During the20l6 and20l7 defenal years a new
GRC was in effect for electric and for natural gas, both of which used a l2-month period ending
with September 2014 as the test year. This means that GRC rates and cost of service changed
between the GRCs in effect for the 2015 deferral year and the GRCs in effect for the 2016 and
2017 defenal years. This has implications for some of the calculations and relationships
reported in this study. For example, the determination of decoupled revenue is the same for 2016
as it for 2017 since both years use the same GRC and test year. When in our opinion the change
in test year or other GRC assumptions have a meaningful influence in observed patterns or
relationships being considered we will point this out to the reader.
We next examine the working of the electric decoupling mechanism and of the natural gas
decoupling mechanisms in detail for the 2015 deferralyear. The same detailed review is
repeated forthe 2016 and2017 defercal years.
Decoupling Mechanism - 2015 Electric (Schedule 75) and Gas (Schedule 175)
Exhibit No. 1
Page l-2 CaSe IIos.AVUt-E-1 9-0_ and AVU-G-1 9-0_
P. Ehrbar, Avista
Page 14 ot 224
Essentially, the decoupling mechanism is designed to capture all fixed cost that is to be collected
from the volumetric portion of rates. With decoupling, the total amount remaining for recovery
is allocated to customer bills according to a model, and recovered in a structure manner on an
ongoing basis. The decoupling deferrals applied beginning in November 2016 are based on
comparison of the value of actual sales in 2015 to the value of projected sales that would have
met the revenue requirement from January through December 2015.
As specified in Schedule 75 and Schedule 175, calculations were carried out separately and in
parallel, for Residential and Non-Residential accounts. For each of these groups of accounts, the
sum of monthly deferral amounts over 2015 is the cumulative deferral (rebate or surcharge) for
2015. The cumulative deferral for 2015 is then applied over the twelve months beginning
6
November 2016. Amortization of the cumulative deferral balance developed over calendar 2015
was implemented over the twelve-month time window from November l, 2076 to October 3 1 ,
20t7.3
o For Schedule 75, Group 1 is Residential customers (Schedules 1 and 2).
o For Schedule 75, Group 2 is Non-Residential customers (Schedules 71,12,21,22,30, 31
and32).
o For Schedule 75, two rate schedules were not decoupled (Schedule 25 - Extra Large
General Service and Schedule 41-48 - Street and Area Lighting). The non-decoupled
schedules are not included in this analysis.
o For Schedule 175, Group 1 is Residential customers (Schedules 101 and 102).
o For Schedule 175, Group 2 is Non-Residential customers (Schedules 7ll,l2l and 131.
Electric Group I (Residential) and Group 2 (Non-Residential)
Schedule 75A is used to develop the Decoupled Revenue per Customer. Schedule 75B uses the
results from Schedule 75A to develop the Monthly Decoupling Deferual.
Schedule 75A - Decoulled Revenue per Cus
For electric service, following steps in Schedule 75A, Decoupled Revenue per Customer (by
Rate Group) is developed. Calculation of Decoupled Revenue per Customer (by Rate Group) is
specified in seven steps in Schedule 75A. These steps are implemented in Table 7-2 and Table
14.4
Step 1: Step 1 is to enter the Total Normalized Revenue, which is equal to the final approved
base rate revenue approved in the Company's last general rate case, individually for each Rate
Schedule. Table l-2,Line 1 shows initial Total NormalizedNet Revenue. In addition, Line 2
shows Settlement Revenue Increase. The sum of Line 1 and Line 2 is shown on Line 3 as the
Total Rate Revenue (January 1,2015). This corresponds to the full value specified in Step 1.
Step 2: Step 2 is to determine the Variable Power Supply Revenue. This value is shown on
Line 6 and is the product of Normalized kWh (2015 Rate Year) from Line 4 and Retail Revenue
Credit from Line 5.
Step 3: Step 3 is to enter Delivery and Power Plant Revenue. This is constructed by subtraction
of Variable Power Supply Revenue (Line 6) from the Total Normalized Revenue (Line 3) and is
entered on Line 7.
Step 4: Step 4 is to Remove Basic Charge Revenue. Because the decoupling mechanism only
tracks revenue that varies with customer energy usage, revenue from Fixed Charges is removed.
3 While calculation of defenal amounts begins in January 2015, customers first encountered the decoupling
mechanism in customer bills November 1,2016.
4 Table l-2,Table l-3, Table l-4, and Table l-5 are attachments or parts of attachments to the Electric Decoupling
Rate Adjustment filing of August 31,2016.
Exhibit No. 1
P. Ehrbar, Avista
Page 15 ol 224
Page l-3
6
Basic Charge Revenue is shown on Line 10. It is the product of the number of Customer Bills
(2015 Rate Year) on Line 8 times the Proposed Basic Charge (Line 9).s
Step 5: In Step 5, the Decoupled Revenue is equal to the Delivery and Power Plant Revenue
(Step 7; Line 7) minus the Basic Charge Revenue (Step 4; Line 10). Decoupled Revenue is
shown on Line I l.
Step 6: [n Step 6, (see Table 1-3) Decoupled Revenue is put on a per customer basis. The
Decoupled Revenue (by Rate Group) is divided by the approved Rate Year number of customers
(by Rate Group). This determines the annual Allowed Decoupled Revenue per Customer (by
Rate Group).
Step 7: Step 7 is different from the other steps because it converts the annual Allowed
Decoupled Revenue per Customer (by Rate Group) into monthly values. The assignment of
monthly values is carried out by modeling monthly kWh use (by Rate Group) in relationship to
the annual kWh use for the rate year. This modeling is shown in Table l-4. Kilowatt hours
(kwh) for Group 1 (Residential) for 2015 is shown in Line 3 and for Group 2 (Non-Residential)
in Line 6. Both monthly values and the annual kWh value are shown. Below the monthly values
(Lines 4 andT) monthly percentages are shown. Lines 1l and 14 shows the use of these
percentages, applied to annual Allowed Decoupled Revenue per Customer (by Rate Group) to
generate monthly values.
The monthly values developed following the steps in Schedule 75Aare then taken forward to be
used in the implementation of Schedule 75B.
5 Basic charge includes minimum charge revenue for non-residential
Exhibit No. 1
Page l-4 P. Ehrbar, Avista
Page 16 of 224
llos
BENi;<bO .: r"-Ll (ErP€ o?-EaElfcO-a,
o
br
EsIoo2
o)aG[)
t_)
I
\)hn
a.
c;z
=€ExIU
oFJ€at' T
s.t +&<Ed9Fah
9-6Nz3EEqH
tH
aa
\o@oloo+\o - l'-.dA+N-€.-i o €-\ONrN
--i
oe
r-+-6d€\on-
\o \o
6Aq
ooo*61 6F-F-$€*o\o F-
+@6
odA -iNYOc1 O n
c.l
€Yoo
osli:€BOA
_9 Iv)=€E
N;o!
EOoX.< u.lo(r)
0!.eooo>!ooza
-E -E
g6ob;tgd!(!id,*R
ato\o\oo-
O\*oolm
[e-mail]{o\
od
t'-Oi
Fj9ldNY6o:o€-rNN
eg
6+66l\o@nNr
a ag
o\ *oor-OO
€ *OF-+ olnoF-OONo
@ 4@
-@*€o@j-€6C! FjoY@€^ooI
*+
-aQoo6€6lofo-olo
€o+€o€\d1O\t\O\onm .o"
\o
N€t'-oo@
oi t! rioYn
\i oi€
O€-466<f F-. ol
O,--\o c-
aa
f-OnO\hf-1d\o-$oo\oacl6t al
@@o\ooF-o*\ood A ..ioY@noor
qol
\oalaF-F-$$o@$no\
o{ al
C)Oc) .+€n ct\ h hc)oo oo* t 6l r- F-ooo N*i n h 6mn€ \ox- F- + r60.61 O-+ € 6l h€.cl 1 d.oo\ "-i q -q \ONo O1 Oi \O 6l O nO\iO € 6 + O\sh9^oN n
e9sa e6 6 @ @\
ro\
r
r€€
c.l
o.lo{
\o*sN
F-+
o
o
o&
oo
ooaH
o\
$x
o-q 6!E9^on!=
^-OXu ia.2&6ff PE:iE ?Vz4H
Qo-m
€o.=
J
,l
o
o
o&
G
o
o
€
o
oo
F-
J*
rooe= E&Ba
h=Ei.=i-: B-86.;>ebJE}!-oo
.N^9o
r==
Oodz&>
:f,n\o
N><^ENtrxHE aa &5
- aN
6Er.o:::*Y=
h tri;R9s
3d5NZ>
=oo9e&tr:o
Oodz Ee
d- do60FCAF
-No
N
(-)(,
N
F
UU)
J-S r.rFJz>rI]OOr!rA |+E.] UJU)
,l
FoF
(,z
D
O
a
2s.lo
oi
ci
a!l
ETzqo
o
f.t
o
o
.Yi a
.q>EE Ps
d=a
rE;
L=
9:ioxf.l A
o{+()c/)
@-
F-6€
d}
€a€\o
g
\
q
U\
q)
q)q
q)
a-
o(.)
CD!-l
qJ
L
U\)r{
o'
q)
A
q)
q)
a\
\i
I
q)
.a
S
F,-
@
lo$
BgNF<EO,i@Ll 6-P€ or
EqEc(Lo
fr
b!l-
trl
5IooZ
0)a(Eo
\o
I
q)
ooB[.
c;z
--€cxuJ
s.q)
q
U\\)
a)
q)
a)q
a_
Uq)a
(.)
Uq)
tf
s\
*i
I
q)
^a
F\
@
Ets*4O-g().;=
8,2Iq)6az
c)
Oo
C)c)
a
se
I
(Jq)
ri
I
q)
o
0
U
c)
q)
6)
6)ilEc)a
<.)o)
crl
o
o)
ao
o)
q)
I
O
e)o)z
bo
Iq)
c,)
oq)
ri
b€o;0€6SZJ
-E -B
96ob
EE*s5H
a2()
O
oo\o
\olr)
ao
N
6
r-o\\o$@o\
c.t
6
g E $AE
E E S }t:>thOoC) = .l Lilo.=o)bHXa.: a0.)E
JAH-t-Oo IJr -!i>Er(.)tr
okf!
F-O-ooooooo^ vI \a\ co o\O\Naoq c\ c')oOO\(r)S*t
@6@
\o(r)oNr-F-\o \o \ocooN\ooooc{Nco
f-*i$ C.l ra)
@6@
olco
co
Nc\l
c.l
N
a()
o
oa
,(
=fno\
clt
@
,r?€
r-
4
N
li(.)
o
0
O
k()a
0)
C)
0.)&!(.)a
o
C)0)
t-l
cO
F-t'-c\(r)
aO
la)@r-
cl
(i
cd
t-t
(.)
()
()&
oNatr(.)
o
0
(-)
+ro
k
o
C):
&
on
\
o\
\No\q
oo$
@
ra)lr)\o^t\oc\r-$
@
o)00
cO0r
$xE
a)qo.
a()
()
C)&
0)
9.
<)
C)
l-1
O
-o
llo$
BENF;<bO .-: orLl (E-P€ cr
EqfltrO-6
fr
br
E
tsIoo2
o)a(!
t_)
F,\
I
q)
bo
a.
ciz
=€tx
UJ
\q)
q
U\q)\q)
a,)
\)
\)
a-
oUq)a
U\
UqJ
Ia\
o'
q)
o
q)
q)
l]r-)
!-e\
\I
s)
-a
b.\
@
I
F
F
z
o
!E
i€EE
E}e32"7g-
=of
L
o
tI]
Lo
c(J
Eo.
.E> I
'E ca
-=i- eE
.2Za>oo<a E
!>.
-9i.,1 E
z
o
c
i
i2
a
F
..:
I
!
6
j
^.4jv
Es
O
d6
rt E?v g;'
lEs5=E
s!:
€
.l
=
!a
F
qj
T
q
-
1Jri3l deil +;Ll .a^el !S-5t gEl ooQItl e9ld
.01 pB
=t ez;t =a{ sg
AlsI€4tE--
=l{==3l iHx>t {'
669=
n
!s
.i
r- .'l
ri
Floi
F{
e3
F
3
S=UE€ir 9g.i E.Jq<Et!it > xa.
d
;o
tr
!N
?
a
9l z2,,tlSEt
rl x;"<et q,
I
-5
I
6
G
6
Schedule 758 - Monthlv Decouoline Deferral
Schedule 75B specifres the method for developing the Monthly Decoupling Deferral for electric
service. The calculation of the monthly decoupling deferral for January 2015 is shown in Table
1-5 for both decoupled groups.6 In the full version of this table (Table 1-6), the monthly
decoupling deferral amounts across 2015 sum to the annual total decoupling deferral for 2015.
For the electric residential group, deferred revenue for 2015 is $7,167 ,748. Deferred revenue in
2015 for the electric non-residential group is negative $2,373,472.
Table l-5. 201 5 Electric Deferral Calculations
Avista Utilities
Electric Decoupling Mechanism
Developrent of Electric Deferals (CalendarYear20l5)
Line Nc
Revised
Ju-15Source
(a)
Resi&trlirl Group
ActualCustom6
Monthly Decoupled Revenue perCustorer
Decoupled Revenu€
Actual Base Rate Revenue
Actual Basic Charge Revenue
Actual Usage (kwhs)
Retail Revenue Credit (gkwh)
Variable Power Supply Payrrts
Customr Decoupled Payrnts
Resilential Revenue Per Custorer REceived
Defeml - Surcharge (Rebate)
Defeml - Revenue Relaled E)penses
btercst on Defeml
It&othly Reiidential D.f.rril Totrl!
Cumlative Res idential Defeml (Rebat€ySurcharge
NoFRca i&Dtirl Group
ActualCustomB
Monthly Decoupled Revenue per Customr
Decoupled Revenuc
Actual Base Rate Revenue
Actual Basic Charge Revenue
Actual Usage (kWhs)
Retail Revenue Credit (gkwh)
Variable Power Suppt Payrents
Customr Decoupled Paymnls
Non-Resilential Revenue Per Customr Received
Defeml - Surcharge (Rebate)
Defeml- Revenue Related Epenses
25 lntercst on Defeml
nhothly Non-Rcsidertial Deferral Totals
26 Cumulative Non-Residential Defeml (Rebate/Surcharge
I
3
4
5
6
7
8
9
l0
ll
t2
l3
t4
l5
l6
t1
l8
t9
20
2t
22
2i
24
o)
Revenue System
Appendix4 Page 3
(l) x(2) $
Revenue System S
Revenue System li,
Revenue System
Appendix4 Page I $
(Ox(A $
(4)-(5){8) $
(3)-(e) $
Rev Conv Factor $
FERC Rate
Avg Balance Calc $
,(r0)"(12)) $
Revenue System
Appendi\4 Page 3
(14) x(ls) $
Revenue System $
Revenue System $
Revenue System
Appendk4, Page I $
09)x(20) $
(17)-(18){21) $
(16)-(22) $
Rev Conv Factor $
FERC Rate
AvgBalanceCalc $
$
r@3) -(2s) $
(c)
l07.lll
$78.8r
r6,33t,t82
:5. t 0t,8J5
1.76 r.l0.l
17.1.()(6.95.1
0.021080
5,775,223
17,565,2t7
$&1.?6
( r,234,035)
56,0t9
3.25%
(1,595)
$o,r?961t)
(l,l79,6lr)
1S.059
$356.03
t\482,17t
16.25ri.9,10
I,590.7:J
162.655.5Ii8
0.021080
3,48,780
|,239,431
$320.59
1,2/t4735
(56,414)
3.25%
1,606
25 TotalCumlativeDefeml (13) + (26) $
t,rE1 927
1,t87,927
8,3 l6
6 Only one month is shown here to keep the table readable on the page.
Exhibit No. 1
P. Ehrbar, Avista
Page 20 of 224
Page 1-8
o
The sequence of the line numbers in Table l-5 implement Schedule 75B. Actual customers each
month (Stepl of Schedule 75B) corresponds to Line 1 for the residential group and Line 14 for
the non-residential group.
Decoupling Deferrals (Step 2 of Schedule 75B) corresponds to Line 3 in both tables. It is
calculated by multiplying the number of Actual Customers (Line l) by the Monthly Decoupled
Revenue per Customer (Line 2). Actual Revenue collected in a month (Step 3 of Schedule 75B)
is shown on Line 4.
The Actual Basic Charge Revenue (Step ) is shown on Line 5. The total revenue collected
related to the variable power supply (Step 5) is shown on Line 8. This is the product of Actual
kWh Sales (Line 6) and the Retail Revenue Credit (Line 7).
Actual Decoupled Revenue (Step 6) is calculated by subtracting the Actual Basic Charge
Revenue (Line 4) and the variable power supply revenue (Line 8) from the Actual Base Rate
Revenue and is shown on Line 9.
The Monthly Residential Deferral Total for each month (Step 7) is shown just below Line 12.
This is the difference between the Actual Decoupled Revenue (Step 6; Line 9) and the Allowed
Decoupled Revenue (Step 2; Line 3) plus any interest on the deferral.
Interest on the deferred balance accrues at the quarterly rate published by the FERC. The
Monthly Residential Deferral Total for January 2015 is negative $1,179,611. In Table l-6, these
values are cumulatively incremented by month over 2015 on Line 13 and the electric deferred
revenue for 2015 shown on Line 13 at the right is $7,167,748. This is the Residential value
given by Avista on page 2 of 5 in the Electric Decoupling Rate Adjustment filing in compliance
with Commission Order No. 05 in Docket No. UE-140188 on August 31,2016.
Continuing with the electric analysis, identical procedural steps were applied for non-residential
customers beginning in Line 14 and yielding a non-residential annual deferral amount of
negative $2,373,472 in Line 26. The net deferral of $4,794,276, including electric residential
and electric non-residential, is shown at the bottom of Table l-6.
Exhibit No. 1
Page I-9 Case Nos. AVU-E-1 9-0_ and AVU-G-1 9-0_
P. Ehrbar, Avista
Page21 of224
lt(!so-N
E3iLf (ENP€ oZEaEr3lc tLl6
f,r
b5
ln
BIl6lo2
lorlalcp
S
I
q.)
bo
Sq.
c;z
--IEx
tJJ
\\.qJ\-U'a
Q
s_
Uq)
r<
q)
a_o
q)
q)
Ll
t-)
(t.t
\cj
I
-a
F,\
@
6l:
!
z
A
{
E.e!
-2 1os
tr>
3C
U?
8.r
o3
z
I
t.
6-aq
.f
8T
r,.:J=dd:dddd'J;;q;6e3^ --=
:=!e:FiE5"td r 6 o.i..i-l
--9-asa
djJodict-!
-o:5-S6
NO*-*N@
=r1=oi+=6ddidj:,-:o{d
ssSGsSi
q€96ndQ
- = t. = + 6 = = @ 6 6
j-'3'+j
.i jj=ddgdddi
-<PGn"{6
iirio;i-l
e-I6->"o
;i jodi j i-t
d j t o Fi..i-9
-^9-GSo
cj.jeeij j
-!
sH:o
3dg:
=-
^Po=bR$6$
J
''6$"9
FNE
-,li9
=
r&F
-::ci
3
.r!*
,.,si
s
,ii3.-,e&
=
=.9 €":xa€
d
:qs.'.IEa!
;9E
;1 3;j
:,{$-,.aE
I
,'u9j
-: !1d-ugj
Jtr-
,-u3
$
I5^
EEi
I
Ig9.o.E 8.Yr.Es
{p'i rqJ -9 Er c It
-nxE
E€ E€:d
oo9-i'\"o
ooPo6>"o60!60hh
ij.jo+d
=69-AS-
.ijjodio6-l
<aEo?So
€17-.FfSBSE'i'd
-:jJodid-l
aca
on96-S-
a-.toao-g9F-qo---*=6oX+F6-
6 c I a6 s -O6:hh6-
;jCoeioi_:
o s E o 6S -
-; :;odd :-f
6d*dG'S6
;it=oidX.dci'ri$_:
- - I -,as *
d jio <jj :
--h-as-
it*o;j
cr! F
E39
-.9-
?
=.:ipE3i3d-J-t:.:F-q d- |
"I
...1,;;;8^3 ^E.ed
-. ;:e ri95
&,!&h = ^i"*v
I
Iaie
.!9
{ !q.gb+9
t5E
d9-6
sssg
!
o
=frq;:i vi X48jNGt'
,*-
=3e9
!ER
E3s
-]
o!o
53S.,!
=RojEs
PAl
-q-
86S
;
1:q3-g
!
58d:
:qs
:6;:
E&
9; I
AEdE
E
3
I
q
6a
l+6:N!
A
.9
3e5l, >., H:5;;
;
e
E
E
EEE; :q q g f ^NoaaLaY
a66;.J;9g9il=
a i a q "F4ed a =
;
P6
5a
!'6
dta?gE
=E -E ?c <aEgio
'
g 3 F H: E EJta;etE;*$fl?E! 3J€ E9;9e-o6Raed d 6 i* LE ,E?;:iEgEE::*is3+EE
2
?t
E
I
7.
5a
E
F
?E
6?s
=
=o
F
E
;
.bs!;
s4*EE
+*^9 Ic <EE9idH; 3e5!jE&EpEBlEEtc Ei6a s9 G 3Ed!,cEEirtgt.c&Ee.{;;e
tt;;tE*ce!xxp!3eEE
6
Notural Gas Groap 1 (Residential) and Group 2 (Non-Residential)
For natural gas, following steps in Schedule 175A, Decoupled Revenue per Customer (by Rate
Group) is developed. Calculation of Decoupled Revenue per Customer (by Rate Group) is
specified in seven steps in Schedule 175A. These steps are implemented in Table l-7 and Table
1-8.7 Monthly Decoupled Revenue per Customer for Group l: Residential and Group 2: Non-
Residential are then used to develop the Monthly Decoupling Deferral for natural gas, following
the steps in Schedule 175B.
Schedule l75A - Decouoled Revenue oer Customer
Step 1: Step I is to enter the Total Normalized Revenue, which is equal to the final approved
base rate revenue approved in the Company's last general rate case, individually for each Rate
Schedule. Table l-7,Line 1 shows initial Total NormalizedNet Revenue. In addition,Line2
shows Settlement Revenue Increase. The sum of Line 1 and Line 2 is shown on Line 3 as the
Total Rate Revenue (January 1,2015). This corresponds to the full value specified in Step 1.
Step 2: Step 2 is to determine the Variable Gas Supply Revenue. This Variable Gas Supply
Revenue is shown on Line 6. It is the product of Normalized Therms by rate schedule from the
last approved general rate case (2015 Rate Year) from Line 4 times the PGA Rates from Line 5.
Step 3: Step 3 is to determine Delivery Revenue, which is entered on Line 7. To determine the
Delivery Revenue, the Variable Gas Supply Revenue is subtracted from the Total Normalized
Revenue.
Step 4: Step 4 is to calculate the Basic Charge Revenue. Because the decoupling mechanism
only tracks revenue that varies with customer energy usage, revenue from Fixed Charges is
removed. Basic Charge Revenue is the product of the number of Customer Bills in the test
period (2015 Rate Year) on Line 8 times the Settlement Basic Charges (Line 9). The result,
Basic Charge Revenue, is shown on Line 10.8
Step 5: Determine the Allowed Decoupled Revenue. The Allowed Decoupled Revenue is equal
to the Delivery Revenue (from Line 7) minus the Basic Charge Revenue (Line 10). The
resulting Decoupled Revenue is shown on Line 11.
Step 6: In Step 6, Decoupled Revenue from Line 11 is put on a per customer basis. The
Decoupled Revenue (by Rate Group) is divided by the approved Rate Year number of customers
(by Rate Group). This determines the annual Allowed Decoupled Revenue per Customer (by
Rate Group) as shown in Table 1-8.
Step 7: Step 7 is different from the other steps because it converts the annual Allowed
Decoupled Revenue per Customer (by Rate Group) into monthly values. The assignment of
monthly values is carried out by modeling monthly therm use (by Rate Group) in relationship to
the annual therm use for the rate year. This modeling is shown in Table 1-9.
7 All tables in this section are attachments or parts of attachments to the Electric and Natural Gas Decoupling Rate
Adjustment filings of August 31, 2016.
8 For natural gas minimum charges are treated like fixed charges.
Exhibit No. 1
P. Ehrbar, Avista
Page 23 ol 224
Page 1-l I
o
In Table l-9, therm use for Group I (Residential) for 2015 is shown in Line 4 and for Group 2
(Non-Residential) in Line 8. Both monthly therm values and the annual therm values are shown.
Below the monthly values, percentages (Lines 5 and 9) are shown. Lines 14 and 18 show the use
of these percentages, applied to annual Allowed Decoupled Revenue per Customer (by Rate
Group) to generate monthly values.
These monthly values are then taken forward to be used in the implementation of Schedule
175B.
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 24 of 224
Page 1-12
o{'
EN<b
'i()(!N€ c.r<o)qf
(L
cft
I
q)b.
ciz
=s-cxIJJ
U.Eo
o
o6
E
r!o!a
o
a
N\t
1
N*r
clo
t6r
6
d+
ooo N r 6 h I roOO r h \t N @ hooo 6 h \t * 6 +
hoh N € o O 6ror € o o 6 d @ohh I € h 6 6 h
6@-foN@oh€<t@fh N
o
o
o
oa
o
oN$6r;€-
N€
a*oroetcl.: c\o€n
9-
N
ooi
N60ON€- c\€
o
$*
d $,eE e5; ia6&
aP@
LEE
ooo
o6a
@60
O
s
6r
6di oi6a-1
N
N€
6€6N 6o\q
NNN
6
6
a
J
J
o
o
od
e.o
oAr
OO6N6O-i9-i
o-o.lr€
cdp
I
d,=h9 o;nd-itroEO
taFc o-E.!tso
Ha.9Xri !za->
+hQ
*6o$d*rt:rbaho
a6
NOE
6
a
rr
+r
Nr
00€r666--
<f€t€r6
6iO
€-6o@€
o99
ax^E i-ttrx
EE a{ g5
- aN9<
.Ecr.ad6&,e?htrj;
RB3
Eve
=oo9e&EEgOo6z Ed=9=j!Jtr#t:
Ou)€Fl*
Qd8Eo
OF
@NIJ.] -
<NUNtrl-I"i
a-
3pq
l6f-
aoF
Z
ci
O52'.irr IOU.q
-]
O
J-
2v,El
<=tr rrtzlxo
5A
J
FoF
L
z
dE.=o
Ea
az5.9 uE.:E>
=EEE*O
-oE.!: !? o.2xa,zd,
d=!i
6:
!
o
Io
\q)
q
U\q.)
a,)
q)
q)
a<
A)\
oUqr)aa
\,
u\
R
q)
\o
q)
q)at-)
c\
N
I
q,)
-a
Fr
@
l(U$o-Noi;sT<EO ,:<oi.81
1Ed:qflc(Lo
Ioo
UJ
l
cioz
c)ooo
\
I
q)
boBo-
c,z
=IExut
q)
o
q
U
F\
a,)a,
A,)
q)
!
vUq)lt
4
!
t
a\
od
I
q)
-a
-!t\
o
@
.+
o.l
c.l
@
cacao\
@
F-o\
@
+
O
\o
@a
o\
F-c.]
oo
@
(d
oF
q
6tz
IEKaY
Eiao=EUoa,o--o.: tr-
=_= g
!s39
=oq,|.;ax
1oE16IO
l,/E
9oo
zz
o
9
o
C)
o
I
z
o
q)
q)
o(h
o
Fl
!o
hx
OBe..t aoC)9b
39o1A6u6
o;.iod,-:d +Iod0.)
^O=/=o.5oo60)oczo
6lO
6
ol
o
o
oa
o
v)
o
rL
(!o
Eo
o
(g
c0
Ea
frr
oA
$ooci\o * a.t
c.l oO\ r,E$
s, s1
$\oooo\ooNr\ .:
coo00O\nN
v c.l$
oe)
osgo./\- H c..lh(Dv-o.,J\v.atr^!()-c>oqo. d.q
+
(t
z
c.l
c.l+o\
a.l$
l(5!to-NUi <NFi<b[9 t'F-Lr- (ENP€ c.r2,E(')81flcO-(!
Io
bil.-til
tsI6o2
OJoo[)
!a
I
s)$Q.,
c;z
=€txuJ
J
F
F
6
z
z
J
z
L
E(J
.-z=.1 uE.:tr9
2Z
=
.8
qq
E
I
U
:
d
-l
E9
-8.
O
I'dE aE;8iJa-
NAS36-Y=E
q
q
e
t
:
dO;-
=7
- F"il f!fle:l
HA
EI*l E
ll r,Edl {5d dB
4sI9il 5-rgEl:
q=
:€
q
L
i
€-q
+S
;+.1
nS
oS
€"
nl
.l
cI
=-.
P
z
b
.a6o
E
I
dc.E
sl ExE<Y3:5t.'
tsd.3
=a:
sE5Ei.:E.'I
9
E
.:EEI ?4tI;l Ed E_d .:E4 E;.,EIIE€E!e3>{4q,
€-
.6
i!
a
!J
6I
o
I
\q)
4
U\q)
q)
a,)
U
a<
sU
Uq)
t)
d
$
\
R\
L
a-
\)
\)Ll
Sa\
o\
I
q)
-a,vF.\
@
o
Schedule l75B - Monthlv Decoupline Deferual
Schedule 175B specifies the method for developing the Monthly Decoupling Deferual for natural
gas service. The calculation of the monthly decoupling deferral for January 2015 is shown in
Table 1-10.e In the full version of this table (Table 1-11), the monthly decoupling deferral
amounts across 2015 sum to the annual total decoupling deferral for 2015. As shown in Table
1-ll,theannualtotaldecouplingdefenalforResidentialnaturalgasis$5,311,558. Theannual
total decoupling deferral for Non-Residential natural gas is 91,736,736.
Table 1-10. 2015 NaturalGas Deferral Calculations
Lire No.
Crtsgor)Sonrce Ro'ised
Jen-15
(a)(b)(c)
Residential Grouo
I Actual Cugouers Rcrtrruc Sysenr 150.806
\Iorthly Decoryled Retsruc pcr Cwloulcr Appendix 5. Paee l s{8. l{
3 Dccorplcd Releuue (l) x (:)s 7.259.{J5
Acrual Usage Re\-enue S-Yseul :0.-116.016
{Acnral Bas€ Rate Retenue (Exchdirrg Oas Cosls)Rer'atlue S_YgeIlr s 9.163.509
5 .{ctul Fired Chuce RErenue Re\Hrue S!'geul s 1.357.1s.r
6 Crtlollrer DBc ouDled Pavnrenrs (d) - {-r)s 7.806.155
Resilential Rertrue Per Ctstonpr Recched s5l.-46
1 Deferral - Surcharce (Rebale)(3) - (6)s (5-16.800)
s Defenal - Revenre Related Expenses ReI Conl'Faclol s 2.r..r95
FERC Rat€-1.f 59o
9 hrterest on Deferral Alc Balance Calc S (70n
lloolhlv R0sidetrtial Defelr{l Totals s (52-j.012)
Oumrlatirt Resllertial Defernl (RebatelSruchargel0 r(7) - (9))s (s23.012)
\on-Rtsideotial Grouo
ll Acnral Cugoners Re.\Buc SYseu
l:lr{onthly Docorpled Rer-eoue per Cstonrer Anoendix 5. Pace -1 s6{2.:{
l-1 Dccorplcd Ra'arue (l l) x (12)s 1.683.941
Actual Usage Rertnue Systenr 6.9?6.t0t
l.l Actrnl Base Rate Rerernre (Excluding Oas Costs)Revorue Sysenr s 1.739.453
l5 Actrnl Fited Charqe Reverrr Revarue Svstem s 23 l.552
l6 CuionleI Decoupled Payurenrs (l{)-(lJ)s l-507.901
Non-Resileotial Rertow Pcr Crstourcr Reccitcd s515. l0
ll Defcrral - Srucharge (Rebatc)03) - 06)s 176.039
l8 Defenal - Rertnw Related Expenses Rev Cout Factor s (7.880
FERC Rate -l.l50o
l9 hrterest on Deferral .!rre Balance Calc S ll8
\tonthh'.\oD-ResldeDtlal Deferral Tolals s 168-18r
Crurnrlarirr Noo-Residcntial Deferral (Rebate)/Surcharge20 Y{lr-) - {19)s 168.381
:l Total Crunulatile Natrual Gas Defenal ilo); (:0)S (35.1.631)
e Only one month is shown here to keep the table readable on the page. The full natural gas deferral table is shown
in Table 1- I I .
Exhibit No. 1
P. Ehrbar, Avista
Page 28 ot 224
Page 1-16
6
The individual steps in the Schedule 175B procedure are shown in both Table 1-10 and Table
1-1 1.
Step 1: Step 1 is to determine the actual number of customers each month. For Group l:
Residential, this is shown in Line I of Table 1-9 and Table 1-10. For Group 2: Non-Residential,
this is shown in Line I I of Table 1-l l.
Step 2: Step 2 is to multiply the actual number of customers (Line 1 for Residential; Line 11 for
Non-Residential) by the applicable monthly Allowed Decoupled Revenue per Customer (Line 2
for Residential; Line 12 for Non-Residential), which was developed in the Schedule 175A
procedure. Allowed Decoupled Revenue for Residential is shown on Line 3. Allowed
Decoupled Revenue for Non-Residential is shown on Line 13.
Step 3: Step 3 determines Actual Revenue collected. For Residential, this is shown on Line 4.
For Non-Residential Actual Base Rate Revenue (Excluding Gas Costs) is shown on Line 14.
Step 4: Step 4 shows the amount of Actual Fixed Charge Revenues included in Actual
Revenues. This is shown on Line 5 for Residential and on Line 15 for Non-Residential.
Step 5: In Step 5, Actual Fixed Charge Revenue (Line 5 for Residential; Line 15 for Non-
Residential) is subtracted from Actual Revenue (Line 4 for Residential; Line 14 for Non-
Residential). The result is shown on Line 6 for Residential and on Line 16 for Non-Residential.
At this point in the calculation all fixed charges have been removed, leaving only variable
charges. In Table 1-10 this is shown as both Customer Decoupled Payments in total and as
Revenue per Customer received.
Step 6: In Step 6, the difference between the Actual Decoupled Revenue from Step 5 (Line 6 for
Residential and Line 16 for Non-Residential) and the Allowed Decoupled Revenue from Step 2
(Line 3 for Residential and Line 13 for Non-Residential) is calculated. The resulting balance
(Lines 7,8 and 9 for Residential and Lines 17, 18 and 19 for Non-Residential) is the Deferral
Total.
Within Step 6, Line 7 for Residential and Line 17 for Non-Residential is the Direct Deferral
(which is either a Surcharge or a Rebate).
Revenue Related Expenses are stated on Line 8 for Residential and Line 18 for Non-Residential.
Below this, the Federal Energy Regulatory Commission rate of interest (FERC Rate) is stated.
Then, the result of the Average Balance Calculation is stated.
Line 9 (for Residential) and Line 19 (for Non-Residential) show the amount of Interest on
Deferral. Below this, the result is the Deferral Totals.
For both Residential and Non-Residential, the Deferral Totals are positive, which would result in
a surcharge.
Exhibit No. '1
Page 1-17 case Nos. AVU-E-19-O_ and AVU-G-19-O_
P. Ehrbar, Avista
Page 29 ot 224
l(u\t?EN?ir(g t'oJE?-aEaqUIEEfc(Lo
Ioo
UJ
f
ooz
o)ooO
oo
I
q)b.
tr
ciz
=€Exul
!
. a,)
a4
U
!
E
a'
qJ
6o'
q)
\)a
Sa\
I
\).a
F\
@
z
A
j
E
T
I
z
!
E/
Eo
2J6 -^9-9 =F
1a=6i
dZ
z9
.4 1
jEi
vi
,d.id
F1
3B$
vtoid
..i
3
-9cado6oqi=.ici +€do6
cii
*raSal-:o€6d
i;oN6<
1t.:l-"dv
TFASd> .r o o
Xd€664.:ri s6:.-d j j
666d6:666-r.:44-€v
.i
9-GS-
i=fiBe-.ij
4a6
!6as-
a?'-o4--
9-AS6
37I; 3eS-<i j j
9-6S-
;€F-
€ j j
6tr-
_64 j
sEs:83-Fj
=Io
j
i:i3-d3j
-.s6.i6qi
er
aa
n 6'S o
ao".!.1
ae
,d
a
s-s3q
6eS-
69Fs;i8+3.i
a3R-.\
'l6nE
-e-
;i ER
.f
-t6
-3-
;cJ-6
9aiSo
o o-Nuo-
ri
-F-49*iS+
ci
tr66SF
ae-iei=ci+n.i:
rccosrhdF;ddd=rj.icij
EfroSo
;.i:i
s-6s-
r o-6J.i:
ndaso
r.qaq
*o5S-
dr:-{
55=eS;S*;
-cr)o-66q!a.la",-!
p6-sF
jjjt'
.,9*
65
5AP
e*
-996c99 q
.i:-i6g
q=.'I
q
.i 3 .i
j
ij q;
d6€j
E P^
9.. 16 E:
G
EG&
=
o
#hh2- s-{5sgr!r. :E9Ebbb= !uEdiaai- -irsddd e. e
A2;
s9,-FE99E#,€!.6' !- E 9-o:-iEFee5 EdSI-ag"l cE9 E R* E E E 3<- 6Et 9 e-sIisEi+ +Ei3:EEre ;CqqE9T.b.b sETXXJiEE C:
I
-9
ogEc
=e-89B d-.
EE+9zEza
vlE
rsa
-Es
6q6
..1
;frR
6 ;6;3
-Ss
.,Ho
a ;3$
,d
^ iSoi
Fj
Es
.:N
6 E3
I
g
eOe8-9ce.t 6
trE+9
E6vo->=o
i:E
I+
H
i
4
gSg tod*#fie o{E sq a q t ,au E
a i A ir eddd e. e
_E&
sq.-9cE;;6E9EE9 -a 9 =d I E L to E;5EdEr !u9d dd; a
& eE a &; =.EE-E9dEE 'EE.E:Hrts6 eai"EiaE?1 ?-e,-inE E a== 9bq q E 9E b i g:iixS EE cs
Ea
E
a?5;z
=o
E
F
E
6S
=Ia
90as*
oo+N6
lx-es+
.).1-r6
PoCs^9€ddr
+-€n-:aF
*-as-
c 3r,njcij
o
2015 Earnings Test
The decoupling mechanism, in Schedules 75D and 175D, provides for application of an earnings
test, separately for electric and for natural gas.
Schedule 75D - Electric Earnings Test
According to Schedule 75D, the decoupling mechanism for electric is subject to an annual
earnings test based on the Company's year-end Commission Basis Reports that reflect actual
decoupling-related revenues and various normalizing adjustments. As shown in Table l-l2,Line
3, the calculated rate of return on a norrnalized basis in 2015 is 7 .40%. This exceeds the 7 .32o/o
allowed return established by Order 05 of Docket No. UE-140188 (Line 4). Excess Earnings
(Line 6) is $1,113,401. A Conversion Factor is applied in Line 7. When the 50Yo Sharing is
applied, the 2015 Total Earnings Test Sharing is $898,901 (Line 10).
Table l-12. 2015 Electric and NaturalGas Earnings Tests
Line
Number
2015 Commission Basis Earnings Test for Decoupling
Categorv Electric Natural Gas
I Rate Base $ l,338.806.000 s 272,971,000
2 Net Income $ 99.1 14,000 $ 16,783.000
J Calculated ROR 7.40%6.15%
4 Base ROR 7.32%732%
5 Excess ROR 0.08%-1.17%
6 Excess Earnings $ 1,1 13,401 $
7 Conversion Factor 0.619312 0.619450
8 Excess Revenue (Excess Eamings/CF)$ 1,797.803 $
9 Sharing %50%50%
l0 2015 Total Earnings Test Sharing $898,901 $
For decoupled electric customers, the earnings test sharing amount is split between residential
and non-residential customer groups in proportion to their contribution to total normalized
revenue (see calculations in Table 1-13).
Table 1- I 3. 20I 5 Electt^ic Earnings Test Sltaring Adjustntent
Revenue From 2015 Normalized Loads and Customers at Present Billing Rates
lL Residential Revenue 5 2L6,224,542 4g.58o/o
L2 Non-Residential Revenue S 219,883,826 50.42%
13 Total Normalized Revenue S 436,108,368 100.00%
Gross Revenue Net of Revenue
Earnings Test Sharing Adjustment
L4 Residential
15 Non-Residential
16 Total
Adjustment RelatedExpenses
5 44s,679 5 424,638
5 qsz,zzz 5 43r,824
s 898,e01
Exhibit No. 1
uase Nos. AVU-E- l 9-0_ and AVU-G-1 9-0_
P. Ehrbar, Avista
Page31 ot224
Page l-19
o
Schedule 175D - Nutural Gas Earnings Test
According to Schedule 175D, the decoupling mechanism for natural gas is subject to an annual
earnings test based on the Company's year-end Commission Basis Reports that reflect actual
decoupling-related revenues and various normalizing adjustments. As shown in Table l-l2,the
rate of return on a normalized basis in 2015 is 6.15%. This is less than the 7 .32% allowed return
established by Order 05 of Docket No. UG-140189 which established the decoupled rates in
effect in 2015.
Since the normalized return is less than the allowed return, the Earnings Test has no effect for
Natural Gas customers for 2016.
2015 Three-Percent Annual Rate Increase Limitation
Decoupling annual rate adjustment surcharges are subject to a3o/o annual rate increase limitation
(there is no reciprocal limit on rebate rate adjustments). The test is to divide the incremental
annual revenue to be collected (proposed surcharge revenue minus present surcharge revenue) by
the total "normalized" revenue for the two Rate Groups for the most recent January through
December.
Normalized revenue is determined by multiplying the weather-corrected usage for the period by
the present rates in effect. If the incremental amount of the proposed surcharge exceeds 3ol0,
only a 302 incremental rate increase will apply. Any remaining deferred revenue will be carried
over to the following years.
Schedule 758 - Electric 3% Rate Increase Test
The Electric Incremental Surcharge Test is shown in Table 1-14. Specifications for the test
limits the Residential Surchargeto 3oh with the remainder defened to the following year (Line
23). For Non-Residential customers, there is a Rebate of 1.4% (Line 24). The Residential
Electric Carryover Deferred Revenue is $875,657 (Table 1-15, line 25). The Non-Residential
Electric Carryover Deferred Revenue is $0.
Exhibit No. 1
uase Nos. AVU-ts-l 9-U_ anO AVU-(,-] 9-U_
P. Ehrbar, Avista
Page 32 oI 224
Page l-20
e
Table I - 1 4. 201 5 Electric 3% Incremental Surcharge Test
Line No,3% lncremental Surcharge Test Electric
November 2016 - October 2017 Usage
Residential
Non-Residential
Proposed Decoupling Recovery Rates
Residential
Non-Residential
Present Decoupling Recovery Rates
Residential
Non-Residential
lncremental Decoupling Recovery Rates
Residential
Non-Residential
9 lncremental Decoupling Recovery
10 Residential
11 Non-Residential
lncremental Surcharge %
12 Residential
13 Non-Residential
3% Test Adjustment (1)
t4 Residential
15 Non-Residential
3% Test Rate Adjustment
76 Residential
17 Non-Residential
Adjusted Proposed Decoupling Recovery Rates
18 Residential
L9 Non-Residential
20 Adjusted lncremental Decoupling Recovery
2l Residential
22 Non-Residential
Adjusted lncremental Surcharge %
23 Residential
24 Non-Residential
1
2
2,465,787,4U
2,t54,719,7N
So.m3oc
-s0.m143
s0.0m0c
s0.0000c
So.oo3oc
-s0.00143
$ 4,316,t13
S 7,397,362
s (3,081,24e)
3.42%
-tryo
(s10,626)
-s0.00037
So.ooom
s0.00263
-50.00143
3,4c,3,772
6,485,O2L
(3,081,249)
3.Wo
-1..Wo
3
4
5
6
7
8
S
s
Notes
(1) The ca rryover ba I a nces wi I I differ from the 3% a diustment a mounts due to the
revenue related expense gross up partiallyoffsetbyadditional intereston the
outstanding balance during the amortization period.
Exhibit No. 1
case Nos. AVU-E-19-0_ ancl AVU-G-I9-O_
P. Ehrbar, Avista
Page 33 ol 224
Page l-21
o
Table l-15. 2015 Residential Electric Carryover Deferred Revenue
Line
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
L7
18
19
20
2L
22
23
24
25
26
Residential Electric
Calculate Estimated Monthly Balances through October 2017
Ending Balance lnterest
3.25%Qt20L6
3.46%Q22076
3.50% Q3 2015
Amortization
Dec- 15
Earnings Sharing Adjustment
Adjusted December Balance
Ja n- 15
Feb- 16
Mar-15
Apr-15
May-16
Ju n-16
Jul-15
Aug-15
Sep-15
Oct-16
Nov-15
Dec-15
lan-!7
Feb-17
Mar-17
Apr-17
May-17
Ju n-17
lul77
Aug-77
Sep-17
Oct-17
57,t67,748
$424,6381
56,74J,rL}
56,761373
56,779,58s
56,798,046
56,877,647
S5,837,30s
S6,8s7,019
56,877,079
55,897,077
s5,917,193
SG,937,3G8
S6,391,343
Ss,699,18s
Ss,01s,346
54,4s9,sss
S3,902,19s
53,4s3,624
S3,M4,859
52,6st,z4o
52,779,Os9
5t,7Ls,sgz
S1,330,s2s
iafi,ast
s18,263
s18,312
S18,362
S19,601
s19,6s8
5t9,714
s2o,oo0
S2o,os8
s20,116
52o,L7s
S1s,4os
S17,GoG
s1s,603
5L3,797
5L2,r76
ito,7Lz
S9,463
S8,29s
s7,034
ss,671
54,436
Ss,zrs
Ss6s,43s
5709,764
5699,442
ss69,s88
Ss69,s37
s4s9,283
5418,2t7
5407,924
5479,216
s459,148
S389,493
s4s8,081
Total 532r,074 56,789,727
Summary
27 2075 Deferred Revenue
28 Less Earnings Sharing
29 Add lnterest through 10/31/2017
30 Add Revenue Related Expense Ad
31 Total Requested Recovery
32 Customer Surcharge Revenue
33 Carryover Deferred Revenue
s7,t67,748
(5424,538)
532t,674
S29s,894
57,360,678
56,48s,02L
s87s,6s7
Exhibit No. 1
P. Ehrbar, Avista
Page 34 ot 224
Page 1-22 uase Nos. AVU-tr- | v-u_ ano /\VU-(,- t Y-u_
O
Schedule 1758 - Natural Gss 3% Rate fncrease Test
The Natural Gas Incremental Surcharge Test is shown in Table l-16. The test limits the
Residential and the Non-Residential Surcharge each to 3Yo. For both the Residential and the
Non-Residential Groups, there is an additional revenue amount that is deferred to the following
year.
Table l-16. 2015 I'lalurol Gas 3?6 Increntental Surcharge Test
39( IrKrem€ntal Surcharge Terl
Line No.
November 2015 - October 2017 Usage
I Re5idential
2 Non-Rgidential
Proposed Decoupling Recovery Rat6
Residential
Non-Residential
Present Decoupling Recovery Rates
Residential
Non-Residmtial
lncremental Decoupling Recovery Rates
Residential
Non-Residential
9 lncremental Decoupling Recovery
10 Residential
11 Non-Residential
lncremental Surcharge %
L2 Residential
13 Non-Residential
3% Test Adjustment (l)
14 Residential
15 Non-Residential
316 Test Rate Adjustment
16 Resdential
17 Non-Residential
Adjusted Proposed Decouplirg Recovery Rates
18 Reidential
19 Non-R$idential
Adjusted lncremental Decoupling Recovery
Residential
Non-Residentlal
Adjusted lncremental Surcharge %
Residential
Non-Residential
Notes
3
4
5
6
7
8
Natural Gas
I 19.200,013
s2,601,464
s0.04872
90.04872
s0.036r3
7,70?,916
5,807,425
1,900,491
4.qt%
(7sL,
s0.02
4,597,823
3,48&984
1,10&839
s
s
s
s
s
20
21
22
2t
24
(1) The carryover balanceswill differ from the 3% adiustment amounts due
the revenue related expense tross up panially offset by additional interest
the amorti.ation
Exhibit No. 1
the b3lance
Page l-23 Case Nos. AVU-E-19-O_ and AVU-G-I9-0_
P. Ehrbar, Avista
Page 35 of 224
6
For Residential Natural Gas, the Carryover Deferred Revenue is $2,261,112 (Table l-17, Line
33).
Table l-17. 2015 Residential Natural Gas Canyover Deferred Revenue
Line
No.
Residential Natural Gas
Calculate Estimated Monthly Balances through October 2017
Ending Balance I nterest
3.2s%Qt20t6
3.46%Q22016
3.s0% Q3 2016
Amortization
t
2
3
4
5
6
7
8
9
10
\L
72
13
t4
15
t6
t7
18
19
20
2L
22
23
24
25
26
Dec-15
Earnings Sharing Adjustment
Adjusted December Balance
Jan-16
Feb-16
Mar-16
Apr-16
May-16
Ju n-16
Ju l-16
Aug-16
Sep-16
Oct-16
Nov-16
Dec-1.6
Jan-17
Feb-17
MarL7
Apr-17
May-L7
Jun-17
)ull7
Aug-17
Sep-17
Oct-17
Tota I
Summary
27 2015 Deferred Revenue
28 Less Earnings Sharing
29 Add lnterest throueh 1O|3U2OL7
30 Add Revenue Related Expense Adj.
31 Total Requested Recovery
32 Customer Surcharge Revenue
33 Carryover Deferred Revenue
ss,317,198
So
ss,317,r.98
ss,331,s99
ss,346,038
ss,360,s17
5s,37s,974
5s,397,474
5s,4o7,ozo
5s,422,790
ss,438,607
5s,4s4,46e
$s,47o,979
Ss,086,191
54,s2t,334
53,934,687
53,4s7,978
s3,0s4,220
S2,81s,6s4
52,686,s72
s2,61s,31s
52,s61.,284
s2,s13,130
52,4sO,24s
52,z6t,Ltz
5r4,4OL
514,44o
st4,47e
s1s,4s6
s1s,s01
s1s,s4s
s7s,770
s1s,816
s1s,863
s1s,909
s1s,373
s13,991
5L2,314
s1o,76s
se,483
58,s48
s8,012
57,727
s7,s38
s7,38e
$7,228
s6,861
S399,ss9
5s78,847
ss98,961
s487,474
s473,24L
5247,Lt4
5t37,094
s78,978
s61,s70
sss,s43
570,LLz
S19s,994
5268,4o2 s3,324,488
Ss,317,198
So
5268,4O2
s164,496
Ss,7so,096
s3,488,984
52,26L,LL2
Exhibit No. 1w-G-19-0_
P. Ehrbar, Avista
Page 36 of 224
Page l-24
6
For Non-Residential Natural Gas, the Carryover Defened Revenue is $770,314 (Table 1-18,
Line 33).10
Table l-18. 2015 Non-Residential Natural Gas Carrvover Deferred Revenue
Non-Residential Natural Gas
Calculate Estimated Monthly Balance through October 2017
Line
No.Ending Balance
5i.,736,736
So
lnterest
3.2s%QL20L6
3.46%Q220L6
3.s0% Q3 2016
Amortization
t
2
3
4
5
6
7
8
9
10
LI
L2
13
L4
15
76
t7
18
19
20
2t
22
23
24
25
26
Dec-15
Ea rnings Sharing Adjustment
Adjusted December Balance
Ja n-15
Feb-15
Mar-16
Apr-1G
May-16
lun-L6
Jul-15
Aug-16
Sep-16
Oct-16
Nov-16
Dec-1"6
)an-ll
Feb-17
Mar-L7
Apt-L7
May-17
)un-77
)ul17
Aug-17
Sep-17
Oct-17
5L,736,736
sL,74L,44O
5r,746,Ls6
s1,7s0,88s
5t,7ss,934
5t,760,997
5L,766,o74
5t,77t,22s
5t,776,39t
St,zBt,s72
$1,786,769
5t,662,289
S1,so8,oss
5t,3s3,347
s]-,223,77O
s1,113,406
5r,o42,272
S998,010
S9Gs,7s4
5934,22s
s900,960
58s9,ss3
$71o,31[
s4,704
54,71^6
54,729
ss,048
ss,o53
ss,o78
ss,1s1
ss,166
ss,181
ss,1e6
ss,o23
54,6t7
54,167
s3,7s3
s3,403
s3,13e
52,e7t
s2,860
52,767
52,672
52,s64
52,373
5t29,soz
s1s8,8s0
S1s8,87s
S133,330
S113,768
574,273
547,233
s3s,115
534,29s
S3s,938
543,s7o
S91,613
Total
Summary
27 2015 Deferred Revenue 5L,736,736
28 Less Earnings Sharing SO
29 Addlnterestthroughl9/3t/2lt7 590,341
30 Add Revenue Related Expense Adj 552,075
31 Total Requested Recovery 5t,879,152
32 Customer Surcharge Revenue 51,108,839
33 Carryover Deferred Revenue 577O,3t4
seo,341 S1,0s6,763
r0 The difference of $5,640 between the deferred revenue of $5,3 I 7, I 98 in Table I - I 7 and the deferred revenue of
$5,311,558 in Line 9 of Table l-11 is the balance from a prior account associated with a previous decoupling
mechanism.
Exhibit No. 1
P. Ehrbar, Avista
Page37 o1224
Page l-25
Decoupling Mechanism - 2016 Electric (Schedule 75) and Natural Gas
(Schedule 175)
In this section, we review analysis of data from the test year from October 2013 through
September 2014 (a historical test year), which was used to develop amounts for revenue recovery
for calendar 2016. Recovery occurred from November 2017 through the end of October 2018
(the second rate year). The decoupling mechanism is designed to capture all fixed cost assigned
for recovery through volumetric rates that is not actually recovered due to lower sales than
expected during calendar 2016. This cost is recovered by allocation to customer bills according
to a model. The decoupling deferrals total is based on comparison of the value of actual sales in
calendar 2016 to the value of normalized sales (from October 2013 through September 2014) on
a per customer basis.
As specified in Schedule 75 and Schedule 175, calculations were carried out separately and in
parallel, for Residential and Non-Residential accounts. For each of these groups of accounts, the
sum of monthly deferral amounts over calendar year 2016 is the cumulative deferral (rebate or
surcharge). The cumulative deferral (with adjustments for prior year carryover balance, interest,
and revenue related expense adjustment) is collected through the decoupling tariff on a
volumetric basis from November I , 2017 to October 3 I , 20 I 8.
Electric Group I (Residential) and Group 2 (Non-Residential)
First the electric service analysis is reviewed, then the analysis for natural gas service.
Schedule 75A - Decoupled Revenue per Customer
For electric service, following steps in Schedule 75A, Decoupled Revenue per Customer (by
Rate Group) is developed. Calculation of Decoupled Revenue per Customer (by Rate Group) is
specified in seven steps in Schedule 75A. These steps are implemented in Table 1-19 and Table
l-2o.rr
Step 1: Step I is to enter the Total Normalized Revenue, which is equal to the final approved
base rate revenue approved in the Company's last general rate case, individually for each Rate
Schedule. Table l-19, Line 1 shows initial Total NormalizedNet Revenue. In addition, Line 2
shows the Allowed Revenue Increase. The sum of Line I and Line 2 is shown on Line 3 as the
Total Rate Revenue (January ll,2016). This corresponds to the full value specified in Step 1.
Step 2: Step 2 is to determine the Variable Power Supply Revenue. This value is shown on
Line 6 and is the product of Normalized kWh (12 ME September 2014 Test Year) from Line 4
and Retail Revenue Adjustment from Line 5.
Step 3: Step 3 is to enter Delivery and Power Plant Revenue. This is constructed by subtraction
of Variable Power Supply Revenue (Line 6) from the Total Normalized Revenue (Line 3) and is
entered on Line 7.
ll All tables in this section are attachments or parts of attachments to the Electric and Natural Gas Decoupling Rate
Adjustment filings of August 3 I , 201 7 for the 20 I 6 deferral year.
Exhibit No. 1
0
Page l-26 wo_
P. Ehrbar, Avista
Page 38 ol 224
6
Step 4: Step 4 is to Remove Basic Charge Revenue. Because the decoupling mechanism only
tracks revenue that varies with customer energy usage, revenue from Fixed Charges is removed.
Basic Charge Revenue is shown on Line 10. Basic Charge Revenue is the product of the number
of Customer Bills in the GRC test year on Line 8 times the Allowed Basic Charge (Line 9).
Step 5: In Step 5, the Decoupled Revenue is equal to the Delivery and Power Plant Revenue
(Line 7) minus the Basic Charge Revenue (Line l0). Decoupled Revenue is shown on Line 11.
Step 6: In Step 6, (see Table l-20) Decoupled Revenue is put on a per customer basis. The
Decoupled Revenue (by Rate Group) is divided by the approved Test Year number of customers
(by Rate Group). This determines the annual Allowed Decoupled Revenue per Customer (by
Rate Group).
Step 7: Step 7 is different from the other steps because it converts the annual Allowed
Decoupled Revenue per Customer (by Rate Group) into monthly values. The assignment of
monthly values is carried out by modeling monthly kWh use (by Rate Group) in relationship to
the annual kWh use for the rate year. This modeling is shown in Table I -2 I . Kilowatt hours for
Group I (Residential) for the test year is shown in Line 3 and for Group 2 (Non-Residential) in
Line 6. Both monthly values and the annual kWh values are shown. Below the monthly values
(Lines 4 andT) monthly percentages are shown. Lines 11 and 14 use this percentage model,
applied to annual Allowed Decoupled Revenue per Customer (by Rate Group), to generate
monthly values.
The monthly values developed following the steps in Schedule 75A are then taken forward to be
used in the implementation of Schedule 75B.
Exhibit No. 1
Page l-27 Case Nos. AVU-E-'19-0 and AVU-G-19-0
P. Ehrbar, Avista
Page 39 of 224
ortgN<6t'o([rt€or-c o)1f(L
ooo.l
I
q)b.
q.
I3,)
't)
oc(o
If,,
IJ
f
aoz
(l.,o(5)
ciz
=€txul
:\q)
U
v)
U
s-q.)
q)
U
q)
a<
!q)
UAJ
r-.1
U
(.)\)r.\
o'
q)
a_o
a)
q)a\c
S\
o\
I
a)
-a
r!ts\
@
r
B-E6-f6-uaahalFo
-@hao€.r.! t'drii#
=.nlAnir,U €--:"r'o fo\'d *<0d4rf
6z
al-oFoh6rE6-OhOJh @ r al a,
-Oral\Ori rl rf O\ rf'2 od d ei'U h",'d atU
o;
9a
E
EIiA9aIEE.E &.9.
fagrfilE;g$>A.t*<<r0()<
9:!r9
BS;9e9OnOouo
a
-AEfi=.EI E6 Gt
< E:aYc1Ef;5EIFBdi).LrP4:tE3E&F,I.s6oJziEi, eedo(,
993
9!,.EE"
E8o
EllEo€a
Iul
\o
atIh-
@
.bt'oi@\orf6
8
r^ato\ri
f \o ro<) \o6d I 8-3h 8-oNetFn666doO6!toltof ci' ri' o joFalh
et & cr6 yt
os*6o+6GO- O\- o- 9_€60rorrraah€ohe.igg66
69i9'
2U
bd!s
ao
90@dO@
'!. od oq.
at at
e88*8*ocf
al o \o
h- od q.
G€
\o
?I
ttO^F6IE
R+-€Rqo -ag 8,9AE6
=6 t,o.r&7'd o
E?8
aEa('}<ra
€o\9
83
-i ci'.r. q.
at Oi
stf.
+F
3
t€8Aei' dh6o{oi' I
G-J
FI
9Iiix
g
A
o
E€
E'.zoo
r
Lrftr10rnr-=.id!,larO?Yiet
\o-r
Ri6 dfrod
o
33hE.tor'
ctOhod = oi'ctY6.r c, a!.Oro
L
rao5cttO
ni'(a
O\Ood=F9rto6'r
of
@iclt\6rSER€oFriel'6Or\ovi
\
I
u"l
F+f^f
O-9
-R5 frosf'
EIfi01tfan'9 tr5iuie6olQND..E1#
EgE7-d>
qaaaaE<f ei o'cl-o
r.:-d
sttt\O@
eijd'\Ov\O
t
a!
848oooei od .r
i-Ooriod
er E{aho6t.,t -. -:.dHOFvr
t" E" g"
+r€
.+- tai ..j'
cl cl
888ooocfdcf
OrHooiodid\vatf !t
g6=iis^d r9r tro5€'lrr8-i
RS r.
E$E
rr Ysi.: 9l;UisFgEdcEdcf. .t p,
6=6?<.ts
ra
aa)rt-
o
iJEgl<F4vF
9-
3 rrra-)qt *J(,aOB.J*,dV
al
U
al
Oo
at
IUo
Et-Sqr2tr{oOUJ
trj l,do
-l
FoF
ods
3A
U
oItr.to
()
F.l
o
U'J
at!
7,lJ.Io
zl.l
I
&
E9.2 .r,gg!;i4
PgIEaC
aI.1. *
!9l.l Ea
o
da
o
Bo
Ito*
BENF<Et9 ---l_r- (E $P€ orZEa818c(La,
obli
l+DI
oz
oa6t)
o\c\
I
\)bosq
ciz
=
=xuJ
L.q)
a
U
:..\)
\)
q)
q)
q)
$oUq)
r_l
()\
Q
AJ
Ia
o'
\)
\)
q)a\o
c\
\I
q)
*a
,!F,\
@
?
;',.aa
e7zan
6
g
sd
oa
o7.
o
FI
3l.l
E
Eo6
HYEi!o9.e8q A
=caet'.=.2fr
.-E7ltoalda.ie
ETtr:I
o
E
Eto!
ba
o€al\ot6oat
aoat
\odr
cl
E H *AEv;€a-j E.: u
Es a3E;EIEO9E E6*;ori
O-?\oo.tO\o\. ao- qctalr6t*o\
wi oi ri
oorHFdco-qe-ro6do,o\o
--doi
ei
cial
al
at
a!
IA
I
o60\Gaclho- co- +
1+
\
\o cl -loro\\o. -l r60rf. t'
oPeAG,
-'HatEE=Elv4l l'Hc
-l
oat
o\o
HE
a5-()Ex-E&9x2aaa9(J !,6!o 6dedoIr3Ht9* Booo--udfla
6'
-o
6
l(5to-Noi;lT<E(9 t'c!_r (E{J€ or1Ea:1fc(L(E
Ioo
uJl
ttoz
0)o(5o
en
I
q)
u0
a_
ciz
=€txtrl
o
c
c
a
J
AE
I,^lvI
kd&1ci!
TE36
${
9.
t
d
Fi
alJH l:.erl :;I EeTItBT"1fi*4 7,elT?i
!,ti
IIIi
-a9a
s6
3-q
,!
a
3$
I
rl
TE40!'
t?
frc
!ii
II;
c.6
c.
,
I
n
!
rle !E!,2
a?
3c
i?s iF
f = t ifl3 f *.TE'.:
?
3
5
4
o
IA
d
I
,
c
I
I
7,
!4
I3
.!
.g
!l.l
EE0EI
E:Et*3 E'E';3
'E EEr?l
f; rEr4E
'E .2.uetr3
c
6
EtEI
\q)
4
U
L.q)
v
p
qJq
U\
oUq)
t-]
(.)\
Uq)
L!
>.
o
o'
C!)
$
q)
q)a\o
\
*ia\
I
q)
-a
t\
@
6
Schedule 758 - Monthlv Decouoline Deferual
Schedule 75B specifies the method for developing the Monthly Decoupling Deferral for electric
service. For Group I (Residential), the calculation of the monthly decoupling deferral for
January 2016 is shown in the top part of Table l-22.12 For Group 2 (Non-Residential) the
calculation method is shown in the bottom part of Table l-22. In the full version of this table
(Table l-23), the monthly decoupling deferral amounts across 2016 sum to the annual total
decoupling deferral for 2016. For the Electric Residential, deferred revenue for 2016 is
$10,288,205. For Electric Non-Residential, deferred revenue for 2016 is$1,967,777
Residential Decoupling Deferrals (Step 2 of Schedule 75B) conesponds to Line 3 in the top part
of Table l-22 and the top part of Table l-23. It is calculated by multiplying the number of
Actual Customers (Line 1) by the Monthly Decoupled Revenue per Customer (Line 2).
Residential Actual Revenue collected in a month (Step 3 of Schedule 75B) is shown on Line 4.
The Residential Actual Basic Charge Revenue (Step ) is shown on Line 5. The total revenue
collected related to the variable power supply (Step 5) is shown on Line 8. This is the product of
Actual kWh Sales (Line 6) and the Retail Revenue Credit (Line 7).
Residential Actual Decoupled Revenue (Step 6) is calculated by subtracting the Actual Basic
Charge Revenue (Line 4) and the variable power supply revenue (Line 8) from the Actual Base
Rate Revenue and is shown on Line 9.
The Monthly Residential Deferral Total for each month (Step 7) is shown just below Line 12.
This is the difference between the Actual Decoupled Revenue (Step 6; Line 9) and the Allowed
Decoupled Revenue (Step 2; Line 3) plus any interest on the deferral.
Interest on the deferred balance accrues at the quarterly rate published by the FERC. In Table
l-23, these values are cumulatively incremented by month over 2016 on Line 13 and the electric
deferred revenue for 2016 shown on Line 13 at the right is $10,288,205. This is the Residential
value given by Avista on page 2 of 6 in the Electric Decoupling Rate Adjustment filing in
compliancewithCommissionOrderNo.05 inDocketNo. UE-140188 onAugust 31,2017.
For Electric Non-Residential, Schedule 75B specifies the method for developing the Monthly
Decoupling Deferral for electric service. In the full version of this table (bottom section of Table
1-23), the monthly decoupling deferral amounts across 2016 sum to the annual total decoupling
deferral for 2016 (for Electric Non-Residential) of $ 1,967,777 . This is the Electric Non-
Residential value given on Page 3 of 6 in the Electric Decoupling Rate Adjustment filing in
compliance with Commission Order No. 05 in Docket No. UE-140188 on August 31,2017.
Since deferred revenue is positive, Electric Non-Residential receives a surcharge.
The calculations and the Excel programming are identical for Electric Residential and Electric
Non-Residential.l3
12 Only the first few columns of the table are shown here, to keep the table readable on the page.
r3 New rates became effective January 11,2016. Defened revenue calculations for the first l0 days of January 2016
were calculated at the rates prior to the change. January I lth through the 3 I't was calculated using the new rates.
Exhibit No. 1
P. Ehrbar, Avista
Page 43 ol 224
Page I -31
Table l-22. 2016 Electric Deferral Calculations
AtiS. Ulilitlrs
Dccoupllr g :Ue(t il tsrx - fT-lSO2o.l Brsc ef fecth'e lj lU 20 16
Devdopme[t of lVA Electdc Defa'rals (Crlender Yeer 20lQ
(
s
s
s
I
,
3
4
5
6
1
I
9
Actud Bu Rac Rataue Raaue Sy:tem
ActuC Boic Chugc Rauuc Rauuc Sy:tco
Aotd Usge (kIl.b, Rauu. Slstlo
Atbcf,Eot {,
Radl RocaucCrcdt(S'k\t) Pagc I
\:riablc Powa Supply Palmots (6) r (7)
Or*omaDccoplcdPqrat: (.1) - (5) {8)
RcsidatiC Rauue Pq Cu*ma Rceirtd
I)daC . Surchrgp (Rcbac) (3) - (9)
DdoC -RauucRdacdErpas Ra'Cm'Frrq
FERCR l.
htseq o DdoC ArgBducc Cdc
llortlly Rcidctiel DCmel Totels
Groulaivc Rcidatial Ddad
Swcbrga(Rcbae) Bdnce E(10) - (11))
(N)
Rcllatiel Grorp
Actud Cu:tmsr
Mothly Deurpled Rouuc ps
Grioma
Dccoplcd Ratouc
l\or-Rsidariel Grop
Actual Cun@ss
\tqtblv Drcqpled Raouc pa
Qrctoms
Dccoplcd Ro:ouc
o)
Rauuc Systca
Attacloat {,
Pas.3
(l) x (2)
Rauuc Sylm
Attactoot 4.
Pagc 3
(t{)r (15) S
3l'o
67,165.11
s?8.8r
5,193.368 S
s 8,069.977 S
s 582J73 S
88.?8!.350
0.0]109 s
r,8?r.532 S
5,616,071 S
s83 6r
612,70{) S
t4.6{9 S
I 1,39? t0
s3J6.03
4.057,716 S
5.689,t 16 S
5lt.!06 s
57,401,939
0.02108 s
1,110.05,r s
3.966.555 S
s3J8.03
9l.t9l s
({.1{0) s
6Soo Pro Ratcd
(c)
tJI.050.11 :08,1t?
588.3:s85 l
t!,{J?,9{6 S l?.75t,3t3
16.9{6.9Jt S :S.0t6.927
t.::2,93{ S 1,805,358
186,4{2,936 l;5,115.:86
0.016{r s 00r?9:
3.059,539 S {,93t.061
l:,66,1,{38 S t8,:80,509
ss9:9 ss-
(:06,{9r) s
g,{?{ 7s :{,1:i
s
s (505
13.933.90
s362.51
8,676,3 t6
35,i3 I
s ll,?34.06:
u,9{7,I.t3 s l?,636,158
r.076,:63 S t.588.t69
120.5+6.1?l 177.9{9,ltt
10
u
s
s
tl
t3
l{
IJ
l6
17 AaudBmRacRoaue RouucSl,lcm S
18 Aaual Buic Chugc Rouuc Rouuc Systco S
19 Aotd UsSs(k$}s) Rauuc S.v:tco
Artrctoar {,
:0 Raail Rorare Crcdit (S'k\lt) Pagc I S
It \:aiablcPowaSupplyPaycDti (19)x (10) S
22 ChromcDccupledPayaots (17) - (18).(2t) S
No-Rcsidotid Ra'oqc Pq Cucea Rccir:d
:3 Ddaal - Suctasr (Rctac) (tO . (:l) S
l{ Ddsal -RouucRdacdExpas Ra'CovFrrq S
FERC Rsr.
25 ht rert 6 Dd6d ArgBalacc Cdc
llolrlly Nor-Rcidctirl DC6rrl Totd3
Cloulairt Nm-Rcddatial Dcfoal
16 Swctrgc'(Rcbac) Balacc [(23) - (]JD
!1 Torrl Coqrlerive El(lric De{erral
0 016{t
t,978. l6]
8,891,71?
0 0l 792
3. r 88.2t 7
r:.8s9,1?:
s3
(|]5.]
5,?89
il
s3?t.55
(:16,.101) s
9.918 'S
s
+
s
s
Exhibit No. 'l
0
Page l-32 P. Ehrbar, Avista
Page 44 ot 224
l(E$
BgNT<E(J t'()i3f1catltEI(5
Io
O)
LIJ
l
aozoaoi.)
enoA
I
q)
boq
ciz
=€sxtu
3
I
!
L
5
3x
l'
^t*;
a
ri
>F
.iL
99-3e
:B!
i,g
.?4E'
tsE+93a
F
rl-l f-3l=ll3sl fllE
:Lt ti
3
3
ts
d9=
3ta
F
e
g.
a
xt
e4
-.9
as
d9
I
3A
frsEgg
*Hgp
3Es
6Eva
E_3
;E= ,.:.i seslE6$g's E :,"": Eqgf $8"*
3rs5F85r=Jd9di;J{j
gg sxn9a9qq =o-q=----iQ E:F*'=t' .d
; E- = ; ?BBi=8i 0i3!i E€F;i i.5HEjJi -'_E-,i";-
2-gEE" "E:9-*E -!.s,i:;! atF-b!i$f i E3-: -et ??E:E!gEI Er! ie;E?r E;= =; EE E E 4aa q.E;;de ;di &9d+E3 z
Rr. 8.3-En ac
.:
E3fi3Fnd
={ded+'
Ees5s6$
=E'df;r'{-
i 3 ! ?.;; serEg$*"- f, d ;'."b =Ega{g;=-
! S e "zza iEt=EA*" e n ,,ia =$$s+=-
- 3 - --ata-*=; E ;.=* 3.=::iEF9,. E F i!; eEsBt?3d i j + *Ja-
Eetr,id39
=Egnd ;
dd
!: = ;== iFs:EiS.': d I y.i =cigdd;.4 ? 3F: EEEHg=
isf
F-
r: I + rsi EFlixg$.'H d -.i =d{ICd;.. . = = r.= = P =u H g
9-
-=-aU.UiSgEElrdd
q=R
43,
888
=:. i
=-;
=aeF9a
: E = ra i: p : ;iI- 4 2 tli
=
E i { a5i'; d ? FgEd :-i
_:
; f; E 3E.-, u x \a
eE*
"sp 3
=:i P.-u{!
:rExr:d
:: C a-eR $
BqE
? E g;
3i
'! !li " i; !! E Ed qq Ed 5E;} ;
:E
Itg
EI HIsl Hl
"L;laEfF
RXBg
s?
JEed
$q
EEi3
5I9d
i!
;E;il
rF-
I
P
!
!
?
t
i4';E
2T
q
sEEFEA*F-ccajj;*-a
;x*4e9f*-Rs8;d;d
--a9-AS-E=B+ggH3sfgEJ=dd
-FNaN6*-3de+;8*FE{deSgdg
d='
3E$3As**=rdngt';d
HxFdsai€;i
sF€3Et$*5t'dxgi4i
3$ElAp$*sicadcdd
F 5 $ =:1 3**5*Ffisd D E g!r -dd3*d;d
o + t -I dd
= I * z;+ =sBf sgde: D i ira -ddxid.
: I = an4 sEsFSE*Si a; ::-r, -iiEa{d-
d t' qti.: +d
i. *" 'z eg:i a i'6aaE:tI 6:;I
i6E$6:_.€EAaP.< y i i
9-:e E:.E 9! isEia6J EIFiEib= E+i:e! E?E:i;&Ex EE g 9at3?Y r
i: EE;:&E gi: &s6cEE !
=x 3!$g n?
=3ps$sn
+4:
Eg;a3i
E:ic3gqi9Bdi
74t) on<-
-2
ggF
cas
r; -i
t g;
-3
xi;;9
l E x i=2i A d {i=l
d .i-.i
-96
i B <t iii
: ,::t !
aR*
3F
E
E
=E sh:6,
^;bS
i3g
134*
-a-i*5
rqa.{
:Ex."59
=
:3=
4.8
4;9aa3
1;*Erf.:
R-BsEl
!i
- 6Iae
E !- *
e3O :E
I E !,j E
"9 E6 :2
iA
g
\.\.q)
a
UL
Qq)
tr)
q)
\
AJ
qJa\o
!'c\
*;\I
q)
F\
@
6
Naturul Gas Group I (Residential) and Group 2 (Non-Residential)
For natural gas, following steps in Schedule 175A, Decoupled Revenue per Customer (by Rate
Group) is developed. Calculation of Decoupled Revenue per Customer (by Rate Group) is
specified in seven steps in Schedule 175A. These steps are implemented in Table 7-24 and
Table l-25.14 Monthly Decoupled Revenue per Customer for Group 1: Residential and Group 2:
Non-Residential are then used to develop the Monthly Decoupling Deferual for natural gas,
following the steps in Schedule 1758.
Schedule 1754 - Decouoled Revenue per Customer
Step 1: Step 1 is to enter the Total Normalized Revenue, which is equal to the final approved
base rate revenue approved in the Company's last general rate case, individually for each rate
class. Table l-24, Line 1 shows initial Total NormalizedNet Revenue. In addition,Line2
shows Allowed Revenue Increase. The sum of Line 1 and Line 2 is shown on Line 3 as the Total
Rate Revenue (January 11,2016). This corresponds to the full value specified in Step 1.
Step 2: Step 2 is to determine the Variable Gas Supply Revenue. This Variable Gas Supply
Revenue is shown on Line 6. It the product of Normalized Therms by rate schedule from the last
approved general rate case from Line 4 times the PGA Rates from Line 5.
Step 3: Step 3 is to determine Delivery Revenue, which is entered on Line 7. To determine the
Delivery Revenue, the Variable Gas Supply Revenue is subtracted from the Total Normalized
Revenue.
Step 4: Step 4 is to calculate the Basic Charge Revenue. Because the decoupling mechanism
only tracks revenue that varies with customer energy usage, revenue from Fixed Charges is
removed. It is the product of the number of Customer Bills in the test period on Line 8 times the
Allowed Basic Charges (Line 9). The result, Basic Charge Revenue, is shown on Line 10.
Step 5: Determine the Allowed Decoupled Revenue. The Allowed Decoupled Revenue is equal
to the Delivery (from Line 7) minus the Basic Charge Revenue (Line l0). The resulting
Decoupled Revenue is shown on Line 11.
Step 6: In Step 6, Decoupled Revenue from Line I I is put on a per customer basis. The
Decoupled Revenue (by Rate Group) is divided by the approved Test Year number of customers
(by Rate Group). This determines the annual Allowed Decoupled Revenue per Customer (by
Rate Group) as shown in Table 1-25.
Step 7: Step 7 is different from the other steps because it converts the annual Allowed
Decoupled Revenue per Customer (by Rate Group) into monthly values. The assignment of
monthly values is carried out by first calculating the dishibution of monthly therm use in the test
year. This calculation is shown in Table 1-26.
14All tables in this section are attachments or parts of attachments to the Electric and Natural Gas Decoupling Rate
Adjustment filings of August 31, 2017 for the 2016 deferral year.
Exhibit No. 't
P. Ehrbar, Avista
Page 46 of 224
Page l-j4
6
In Table l-26,therm use for Group 1 (Residential) for test year is shown in Line 4 and for Group
2 (Non-Residential) in Line 8. Both monthly therm values and the annual therm values are
shown. Below the monthly values, percentages (Lines 5 and 9) are shown. Lines 14 and 18
show the use of these percentages, applied to annual Allowed Decoupled Revenue per Customer
(by Rate Group) to generate monthly values.
These monthly values are then taken forward to be used in the implementation of Schedule
1758.
Exhibit No. 1
Page t-35
Page 47 ol 224
os
.EN<Et'@(Et€orECD18(L
ciz
=€ExIU
$ar=snti h. 6. e" o- oi;"r89;9E o.-. 6
g6d
o(c.2. 6
6FF66
= a 8_8.43,=6--ts6
5i-F.o-\
,l d-
&
3-eaplt
a& v,?,
.E *F3
!EEg
==':3
@
cJ
!Eofr(J
0
a,7.&
r6
B
Ua
E9
E.?
9"
N
ts
Q
ao
.r
6q@
99al+6Ar16rooio+.;6-66€€do+-6h6-0
,
&!
"La
o
5*A
ro6orci 1
-, e'
I,6
F6+trR.
rEJl ,rs&EAAE
- u&iir 9
EE6e'6 H,= d(J<o€69
t
o
ots
?6t
q
o
+@_
B!>E^ u2?i 2
d lib n
E€:* 6
frgX EHE i
;iE E*E EEoir I-'o d
i*E E:E E
0Fo
eB$
-i -'
saRHfrf
F' .d
=ggF^O Oro9dt
8d|^.€
E
o
q
o,o
.l
E. E.E.
;;
ooo8.8-8-
-.
oo6848d;,., +o!--
aa8ooo6 F,;
?- 6-6-
ooo888{eei
eeeBEAF;i6. @. r!
ao
g
d
5oI
gotrE)dd
Z
c'
'/.U(,
o
U
dc
v.(,
t
D,
E
H
(J
Fa;
l,aada
FNeoU!T?{;e3)a_2
qEE
OI3E!6zv,.?o
E
A
!
Ee
\\)
4
Ur.q)6
qJ
q)
q)
a-
oU\)a4
r\
\
GE
o'
q)
a_o
q)
q)a\o
c\
\a\
I
q)
t'-
@
\o.o
I
q)
bosa-
l(ESo-NUi <NlT<b[, .i o)Lr.(5tP€ orZEaElfcO-6
fr
b5
E
ts
r6oz
o)aot)
F.\eF)
I
q)
bn
q.
ciz
=€ExlU
o
Orr-!f\i
@
\o
ei
n
ea
GoF
\O r' O\codo9\o.\f.t al ol\O O\ clctd<>didiarcl
@vr@
..l t-- \Of-. cA v:
.j.:oiO\ O\ \Octoo\ei6o
6@6
larorru iD ru ='a = o,Jc-69854.9-A&6"upx'6Odfi={r-otkL
= 6l&ruk6t\&
5
tr{
at
C)
O
c)a
o
(ga*
\o!t+r-- 00 00a.nei.ft cl ao\ot'l ,.i
@6
F- tfi \Ora, O\ O\.qA-{
dA)V)O\+rncldci
@@
()lrcO.€-i aC
=Dd
Hil
xacl8*gEzo!!3e?gil- 7*9Eep(J !,
EZE'oT=(u:,(.)o. ,u o.dr' 6o;9(J ;i {'oFo
a al ..)
U
I
G
Ali3
6t&
oz
al
Ct
Et,
a)
0
F.J
Go
G
cz
ae
.-EEEA=O Il
s&
=o)Y'"te,a1lGO,\JA
-EGO
=4,En7,2
aa,
ao
9
{,a
0/
\-\)
o
v)
(-)\q)
q)
\)
q)
q)
$
(-)\)a
4S
s.
R\o
s\
a-;\
I
ED
^a
F\
@
O\t.ENit
o tJ.)€orto)18o-
oOen
I
UboR
Q.,
ciz
=IExLrJ
ST
I
t
?
{a
il
t
a*
€
j
a: tr
l6I.a
t.is
{ri
te$
I$e
E
e
a;
j
::A;
- tr6HIiI
drld
I*r
T"E&
{{x i
,{"?"
B!
q"
AE
3e<ja
.i
si
8;:
4
a,.i
*3i
I
IEi!t'
I!,
,
A6.x{Ia
!!Etillr5":Il;
!6
8r
nd
aaJ
aeI
ari
a(
di
a
a.'-i
-
I,-l,lhl!.
.I
II!,
g
6
EIt-.{!I€liTiIdlcg?i
s
!
5
a
,.
{!?
{II
E
?)
c
b
Ia
6
,r
,
4
I
^
I
7
4
E3
.I
I
;?,
&E
"!93:s4UEAn?,2: {.
=a9t== a53i
f; E!qOr,38
| .1.CEi=7.6
I
E
E
oa-
EF
r.6\
o
q
U\q)
a.
q')
q)
q)
$q)
a-
Uq)a
Fi
\
L
=
,o
q)
\
U
q)a\o
\
\cJ\I
q)
-a
F-
@
,{..ni 69=39:3Pl;=9
0
Schedule I75B - Monthlv Decoupline Deferual
Schedule 175B specifies the method for developing the Monthly Decoupling Deferual for natural
gas service. For Group 1 (Residential), the calculation of the monthly decoupling deferral for
January 2016 is shown in Table l-27.ls In the full version of this table (Table l-z8),the monthly
decoupling deferral amounts across 2016 sum to the annual total decoupling defenal for 2016.
As shown in Table 1-28, the annual total decoupling deferral for Residential natural gas is
$7,152,977. The annual total decoupling deferral for Non-Residential natural gas is $2,002,654.
The individual steps in the Schedule l75B procedure are shown in both Table 1-27 and Table
1-29.16
Step 1: Step 1 is to Determine the actual number of customers each month. For Group I
(Residential), this is shown in Line 1 of Table l-27 and Table l-28. For Group 2 (Non-
Residential), this is shown in Line 11.
Step 2: Step 2 is to multiply the actual number of customers (Line 1 for Residential; Line l l for
Non-Residential) by the applicable monthly Allowed Decoupled Revenue per Customer (Line 2
for Residential; Line 12 for Non-Residential), which was developed in the Schedule 175A
procedure. Allowed Decoupled Revenue for Residential is shown on Line 3. Allowed
Decoupled Revenue for Non-Residential is shown on Line 13.
Step 3: Step 3 determines Actual Revenue collected. For Residential, this is shown on Line 4.
For Non-Residential Actual Base Rate Revenue (Excluding Gas Costs) is shown on Line 14.
Step 4: Step 4 calculates the amount of Actual Fixed Charge Revenues included in Actual
Revenues. This is shown on Line 5 for Residential and on Line 15 for Non-Residential.
Step 5: In Step 5, Actual Fixed Charge Revenue (Line 5 for Residential; Line 15 for Non-
Residential) is subhacted from Actual Revenue (Line 4 for Residential; Line 14 for Non-
Residential). The result is shown on Line 6 for Residential and on Line 16 for Non-Residential.
At this point in the calculation all fixed charges have been removed, leaving only variable
charges. In Table 1-28 this is shown as both Customer Decoupled Payments in total and as
Revenue per Customer received.
Step 6: In Step 6, the difference between the Actual Decoupled Revenue from Step 5 (Line 6 for
Residential and Line 16 for Non-Residential) and the Allowed Decoupled Revenue from Step 2
(Line 3 for Residential and Line 13 for Non-Residential) is calculated. The resulting balance
(Lines 7, 8 and 9 for Residential and Lines 17, 18 and 19 for Non-Residential) is the Deferral
Total.
Within Step 6, Line 7 for Residential and Line 17 for Non-Residential is the Direct Deferral
(which is either a Surcharge or a Rebate).
15 Only one month is shown here to keep the table readable on the page. The fu1l natural gas deferral table is shown
in Table l-28.
16 New rates became effective January 11,2016. Deferred revenue calculations for the first l0 days ofJanuary 2016
werecalculatedattheratespriortothechange. January llftthroughthe3l'twascalculatedusingthenewrates.
Exhibit No. 'l
Page l-39 Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 51 ol 224
o
Revenue Related Expense are stated on Line 8 for Residential and Line 18 for Non-Residential.
Below this, the Federal Energy Regulatory Commission rate of interest (FERC Rate) is stated.
Then, the result of the Average Balance Calculation is stated.
Line 9 (for Residential) and Line 19 (for Non-Residential) show the amount of Interest on
Deferral. Below this, the result is the Deferral Totals.
For Residential, the Deferral Total is $7,152,977. This result is reported by Avista on Page 2 of
5 in the letter of transmittal from Patrick Ehrbar to the Commission dated August 31,2017. For
Non-Residential, the Deferral Total is 92,002,654. Since both are positive, both result in a
surcharge. This result is reported by Avista on Page 3 of 5 in the letter of transmittal from
Patrick Ehrbar to the Commission dated August 31,2017.
Exhibit No. 1
P. Ehrbar, Avista
Page 52 ol 224
Page l-40
.lrlst. utllties
Dccooplh8 flcctrDbD - rC-I5010S Brs. ?ftcflh'? t/lU20l6
Dclelopm cDt of lv A Nrtur. I c,. s Dcf.rrr b (crhtrdr r Y cr !' !0lO
Lire Io,Sour(e
lto Srcd
fta 16
3l?r
Oldh$
68! o
l{c Be*
(a)
R.Gid-dd Crclp
..ldulCusloffi
\lmthly Dcorplcd RawepE CusoG
Drdplrd R6uuc
.rcud L-s+e $afcmeioC onlf ).{culBsRrcRoqw
Erdud6cs cots)
.{au:l Fircd ChagrRawc
QrrwDcmdcd Ptmcos
Rciddil Rcwc Pa Cucffi Rcdrrd
DdcC -Su&rgc(Rdetc)
Ddad -EmcRdtGd Erpcos
o)
R6BeS)sa
AE dmr5,
P.!c 3
(l)r (:) S
R6m. slsio
AtedmtS,
P.fF l
(l l)r (12) s
{9,316..r5
s.lt. l{
tr17{,,161 3
859.35
s6.r2.2.r
ssl.908 s
r03.585.55
s6l J9
6,369,157 S
1 r,656.00,
69e17N S
9.13,{27 S
6019.27? S
sJ8,.l0
319,880 3
(t{,608)'s
t,8o'r,65
s751.65
tJ60.069 3
(c)
I Sl.9r:
357.tt
q7{r,6r9
:0. l 58,861
10. l::.561
t.l9:6.1s
8,929.885
s5t..l0
(r86,!66)
8.066
t-25.tLs (:r l)
S (l?3,,1,a1)
s (l 7s..1Jt)
1,66{
t?l 7.71
l.9t t.977
I
3
.l
5
6
7
I
RaucSF@ 6.50:,859
Ra@.S,sra 3 1,319,859 S
R6mcs!r@ S .l{9,251 S
({)-(5) S LE80.608 S
s5t.10
(3).(6) 3 (506,r.16) S
Ra'Com.F=tu $ 22,67.1 S
FERC Rr.
Arg BCae C:lc
q(7) - (e)
hca oo DcferC
\b!rlh' Reidqli.l De{snl f d.k
ormJ rir? De&rrt (R6ilc) BCae
Itor.B.dd.ai.l Grery
ldDdCNoffi
llmdly Daooqlcd R6@c pc Cwom
DroplcdRcur
Ad0d L's+e (hfcEriot oall')
.{rtul B*RlcRouu
(Erdud6G6 Corr)
.,laul Fircd ChrgrRawc
OrimaDcflglcd Pslms[r
lioa.Rcidatil Rawc Pc qr$@ Rcied
Dd.ni -Sudngc(R6:tc)
DdoC -RowcRdecdE:pos
tuscl oo De&rrd
Ib!aLl!' So!-Ssi.lathl Ddni Tolrls
Olmd tirr Dcftrrd surda!(Rcbate) Bal&a
Tohl [e-mail] Dd6r.l (R6.ae)
9
l0
ll
l2
t3
l.l
t,
l6
l7
!8
t9
l0
2l
Ra@cslrsm S
Rsacsl1ld 3
il.r) -tr5) s
(lr).(r6) s
R6'Com'Frttr S
FERC R,.
.{rjBJaeCalc
E(!7) - 0e))
(r0) +(S)
612,7{9 S
78,9:0 S
56r,E29 S
t656.1 l(lr.t:l) s
53.r 3
l,lE.r,04l s
t655-t I
l76.0lE s
(&039)'S
6.913.97.r
t.99:.Jt l
:.1.r.6s I
1,7,1?.t70
t656.r I
t0{,t07
(7.s0,
3.15.n
t)49,112
165,?3t
s
s
s 212
S
s
156.815
I i6,8lJ
s (11,624
6
Table l-27. 2016 Natural Gas Deferral Calculations
Exhibit No. '1
Case Nos. AVU-E-19-0_ and AVU-G-'I9-0_
P. Ehrbar, Avista
Page 53 ot 224
Page l-41
l(E$
3gNf<EC,itl(Ert.J€ oraEa:rflEo.o
Ico
r.ll
ooz
o)ooi.)
c\\I
q)
bosq,
ciz
=€ExIUI
c
'i
J
I
{
II
I
!I
I
:5
:.t
a!
EI
,ei!8I.!ta
!d6
'i:4"..l<Z?
*igt,l.E
:
i-
.i
4t
i.;jll:l
c't
,irir-a.i \
TIET:iIl*i i rili!
tti!. a,
III
le$!3 lt
, ! r l{
r f i iii, ] :ia z aat , t,
I,i-=I T B'lir:ci
,,!iI
t4
,,
ill.il?-1.
-t_
lfl.lr;;.1 = :r
,|::
'li',l;:
.
i[;
l'.
,E
.a
. .l a:
,:I
iit
a
t I ttl. I I r,a't , tr'z. E L.lia: I ll:az a a a
ii
i ii!i!t:i,rl
IIit;i!rlror l
lil r,rr:Y',,)
a
to
EaI
R\I.U
vaa
!\,
L.
k
q)
o
q)
q)o\o
sa\
od\I
q)
-a-vF,\
@
2016 Earnings Test
The decoupling mechanism, in Schedules 75D and 175D provides for application of an eamings
test, separately for electric and for natural gas.
Schedule 75D - Electric Earnings Test
According to Schedule 75D, the decoupling mechanism for decoupled electric customers is
subject to an annual earnings test based on the Company's year-end Commission Basis Reports
that reflect actual decoupling-related revenues and various normalizing adjustments. As shown
in Table l-29,Line 3, the rate of return on a nornalized basis in2016 is7.Sloh. This exceeds
the 7.29o/o allowed retum established by Order 05 of Docket No. UE-150204r7 lline 4). The
Excess ROR is 0.220 , corresponding to Excess Earnings of $3,218,417 (Line 6). A Conversion
Factor is entered on Line 7, which is divided into the Excess Earnings to produce Excess
Revenue (Line 8) of $5,193,843. When the 50% Sharing (Line 9) is applied, the Total Earnings
Sharing for electric is$2,596,921 (Line 10).
Table l-29. 2016 Electric Earnings Test
The Electric Total Earnings Test Sharing amount is then split between residential and non-
residential customer groups in proportion to their contribution to Total Normalized Revenue
(Table l-30). The split is 50.62% Electric Residential and 49.38oh Electric Non-Residential.
The dollar values for the split are $1,314,495 Electric Residential (Line 14) and$'1,282,427
Electric Non-Residential (Line l5). These values are adjusted to remove various revenue related
expenses by dividing them by the Gross Up Factor derived in Table 1-31 (1.048963). The final
17 Page 6, Paragraph 5 (Commission Determinations) in Washington Utilities and Transportation Commission v.
Avista Corporation dba Avista Utilities, Dockets UE-150204 and UG-150205 (Consolidated), Order 05, Final Order
Rejecting Tariff Finding, Accepting Partial Settlement Stipulation, Authorizing Tariff Findings. Service Date
January 26,2016.
Exhibit No. 1
Page 1-43 Case Nos. AVU-E-19-0 and AVU-G-'I9-0
P. Ehrbar, Avista
Page 55 of 224
5 Excess Earnings
7 Conversion Factor
8 Excess Revenue (Excess Earnings/CF)
9 Sharing 7o
10 2015 Total Eamings Test Sharing
Line No.
5 1,442,726,0N
3
4
5
Electric
1 Rate Base
2 Net lncome s 1o8,40s,ooo
s
5
Calculated ROR
Base ROR
Excess ROR
7.5t%
7.29%
o.22%
3,2t8,4L7
0.619660
5,193,843
50%
s
6
0
values for the Electric Earnings Test are $1,253,138 for Residential Electric and$1,222,566 for
Non-Residential Electric. These values are shown on Line 14 and Line 15, respectively, in Table
1-30. These are also reported on Page 2 of 6 (Residential) and Page 3 of 6 (Non-Residential) of
the Letter of Transmittal from Patrick Ehrbar to the Commission for the Electric Decoupling
Rate Adjustment, Tariff WN U-28, Electric Service, dated August3l,2017.
Table 1-30. 2016 Electric Earnings Test Sharing Ad.justment
Revenue From 2015 Normalized toads and Customers at Pres€nt Billing Rates
11 Residential Revenue S zzr,:sg,mo 50.62Yo
72 Non-ResidentialRevenue s
s
277,949,WO 49.38%
13 Total Normalized Revenue 441,348,000 100.00%
l4
15
16
Earnings Test Sharing Adjustment
Residential
Non-Residential
Total
Gross Rarenue
Adjustment
S 1,314,4955 t,zsz,qzt
Net of Revenue
Related Expenses
S 1,253,1385 t,zzz,soo5 z,sgo,gzt S z,4ts,lu
Table I -3 1. Derivation of 2016 Electric Gross Up Factor and Revenue Conversion Factor
.{1'tSI.{ LTILIIIES
Rcroor Colresior Frctor
$ $hbgaa - f,kdrk $'it!D
T\r'EL\:E IIO\TIIS E]-DED D6obtr 31, :016
Lirc
:{o.D6crtlio!Faaot
I R6ac
E{).r*
! UrcolLctal€s
3 Comb*u Fea
,l \*LSbEloo Ercir Trr
5 Torrl Erpa$
6 NelOpalbgbsreBG&GFIT
7 F€dal hmTu@ 35t6
t REITIiLTCONTERSTO}iTTCTOR
9 Grca Up Foctd
1.0400@
oauilE3
o0a20u
aoJs{-oj
AOkit;7V
a 9i3-rrj
aJj-rodl
ao19.i&
2015 @mmisio 86is CoErsim Factq with Un@lle€tible s€rvke Cd,etion
Exhibit No. 1
Case NoS.AVUI-E:I9.0 enilAVU-G-1 9-0
P. Ehrbar, Avista
Page 56 ol 224
Page 1-44
0
Schedule 175D - Naturul Gas Earnings Test
According to Schedule 175D, the decoupling mechanism for natural gas is subject to an annual
earnings test based on the Company's year-end Commission Basis Reports that reflect actual
decoupling-related revenues and various normalizing adjustments. As shown in Table l-32,the
rate of retum on a normalized basis in 2016 is 8.56%. This is more than the 7 29% allowed
returnr8 (Line 4). The Excess ROR is 1.27% (Line 5). The dollar value of Excess Earnings is
$3,628,723. This is adjusted for various revenue expenses by dividing by the Conversion Factor
(0.619798) given in Line 7 for Excess Revenue of $5,854,687 as shown in Line 8.
Table l-32. 2016 NaturolGas Earnings Test
With the Sharing percentage set at 500/o, the 2016 Total Earnings Test Sharing is $2,927 ,343
(Line l0). The Conversion Factor on Line 7 of Table 1-32 is developed in Table l-33.
r8 Page 6, Paragraph 5 (Commission Determinations) in Washington Utilities and Transportation Commission v.
Avista Corporation dba Avista Utilities, Dockets UE-150204 and UG-150205 (Consolidated), Order 05, Final Order
Rejecting Tariff Finding, Accepting Partial Settlement Stipulation, Authorizing Tariff Findings. Service Date
January 26,2016.
Exhiblt No. 1
Page l-45
Page57 of224
927s
2015 Commission Basis Earnings Test for Decoupling
6 Excess Earnings
7 Conversion Factor
8 Excess Revenue (Excess Earnings/CF)
9 Sharing %
10 2016TotalEarningsTestSharing
Line No.NaturalGas
286,597,000
3
4
5
24,524,000
l- Rate Base s
s2 Net lncome
s
s
Calculated ROR
Base ROR
Excess ROR
8.56%
7.29%
1.27%
3,629,723
0.619798
5,954,697
50%
6
Table l-33. Derivation of 2016 Natural Gas Gross Up Factor and Revenue Conversion Factor
f,iDe
No.
AVISTAUTILITIES
Rewnue Conwrs ion Faclor
Wshitrgton - Gas S)stem
TWE-VEMONTIIS E{DE) December 31, 2016
Descriptioo Frclor
I Rewoues
Eqcnse:
2 Uncollectibles
3 Comission Fees
4 Wuhington Frcise Tq
5 TotalE)eense
6 NetOpemtingIncomBeforeFIT
7 FedenllncomTu@35o/o
8 REVENTJECPI.T\iERSIONFACTOR 0.619798
9 Gross Up Factor L.W729
2016 Commission Basis Conversion Factor with Uncollfftible Service Corrstion
1.000000
0.006183
0.002000
0.038282
0.046465
0.953535
0.333737
The split between Natural Gas Residential and Natural Gas Non-Residential is developed in
Table 1-34. The split is modeled on contribution to revenue. Stated in percentage terms, the
split is 76.15% Residential and23.85yo Non-Residential (Lines 1l and l2). At the Gross level,
the dollar values are $2,229,293 Residential and 5698,050 Non-Residential. When expressed net
of various revenue expenses (by dividing by the Gross Up Factor from Table l-33, Line 9, the
values are $2,125,710 Natural Gas Residential and $665,616 Natural Gas Non-Residential.
These values are also reported Page 2 of 5 for Residential and Page 3 of 5 for Non-Residential in
Letter of Transmittal from Patrick Ehrbar to the Commission for the Natural Gas Decoupling
Rate Adjustment, Tariff WN U-28, Electric Service, dated August3l,2017.
Table l-34. 2016 Natural Gas Earnings Test Sharing Adjustntent
Revenue From 2016 Normallzed loads and Customers at hesent Bllllng Rates
11 Residential Revenue S U0,176,m0
12 Non-Residential Revenue S 34,499,m0
13 Total Normalized Revenue S 144,675,m0
Eamings Test Sharlng AdJustment
14 Resid€ntial
15 Non-Residential
16 Total
Gross Revenue Net of Revenue
Adjustment Related
5 2,229,293 5 2,r2s,7t0
Exhibit No. I
Page l-46 Case Nos. AVU-E-I9-0_ and AVU-G-19-o_
P. Ehrbar, Avista
Page 58 of 224
6
2016 Three-Percent Annual Rate Increase Limitation
Decoupling annual rate adjustment surcharges are subject to a3o/o annual rate increase limitation
(there is no reciprocal limit on rebate rate adjustments). The test is to divide the incremental
annual revenue to be collected (proposed surcharge revenue minus present surcharge revenue) by
the total "normalized" revenue for the two Rate Groups for the most recent January through
December.
Normalized revenue is determined by multiplying the weather-corrected usage for the period by
the present rates in effect. If the incremental amount of the proposed surcharge exceeds 3oZ,
only a 30lo incremental rate increase will apply. Any remaining deferred revenue will be carried
over to the following years.
Schedule 758 - Electric 3% Rate Increase Test
The electric Incremental Surcharge Test is shown in Table 1-35. Following the specifications for
the test limits the Residential Surchargeto 3Yo with the remainder deferred to the following year.
However, division of the Revenue from 2016 Normalized Loads with Customers at Present
Billing Rates (Line 1) by the Incremental Decoupling Recovery (Line 6) results in a value of
2.0Yofor ElectricResidentialandavalue of 0A% forElectricNon-Residential(Line7). Since
these values are both less than 3Yo, no adjustment is necessary for either Electric Residential or
for Electric Non-Residential. For both Electric Rate Groups, the Carryover Deferred Revenue is
equal to zero.
Table l-35. 2016 Electric 3% lncremental Surcharge Test
3j6 lncr:m.ntrl Surch.rEa T..t
Line No.
1 Reyenue frorn 2016 Normalired loads and
Customers et Pr6ent Eilling Rates (Note 1)
2 Nowmber2017 - October 2018 Usage(kWhs)
3 Proposed DecouplinS Recovery R6tes
4 Preaent DecouplirB surchar8e Reoovery Rates
5 lnffemmtal Decouy'ing Recovery Rates
5 lnsementaloeroudingRecovery
7 lncrem€ntal Surcharge %
8 3%Test Adju*ment (Note 2)
9 396 Test Rate Adjustment
10 Adjuted Proposed DecdplirE Rtreery Rates
11 AdiEtedlnsementalDeoplingRerwry
12 Adi6ted lncrem.ntal Surcharge 96
Notes
Rpsidential Non-Residential
5 223,399.0@ S 2uB49,o@
2,4sL572p67 2,7@p28828
50'00445
s0'00253
s0.00182
S 4453,583 S
zoDr
ss
So.omoo
5000445
s 4,463,683 s
LW
864,012
864,012
(1) 2016 t{ormalized Revenue derived frorn UE-17O485 Rwenue Model with
rat6 adrusted to refltrt Aug6t 1, 2017 present rates
(2) The orryover balance$ will ditfer from the 3% 3djustment amounts due to
revenue related expense Eross up partially offset by additional interest on
the amortization
Exhibit No. 1
balance
Page l-47 Case Nos. AVU-E-I9-0_ and AVU-G-'|9-0_
P. Ehrbar, Avista
Page 59 of 224
o
Schedule 1758 - Notural Gas 3% Rute Increase Test
The natural gas Incremental Surcharge Test is shown in Table 1-36. The test limits the
residential and the non-residential surcharge each to 3Yo.
For the natural gas residential group, there is an additional revenue amount of $718,577 that is
deferred to the following year because of the test (Line 8). For the natural gas non-residential
group, the surcharge is less than3Yo so the defened revenue carried forward to the following
year is equal to zero.
Table l-36. 2016 Natural Gas 3(% Increntental Surcharge Test
3% lncremental Surcharge Test
Line No.
Revenue From 2016 Normalized Loads and1 Crnor"r, at Present Billing Rates (Note 1)
2 November 2017 - October 2018 Usage
3 Proposed Decoupling Recovery Rates
4 Pres€nt Decoupling Surcharge Recovery Rates
5 lncremental Decoupling Recovery Rates
6 lncremental Decoufling Recovery
7 lncremental Surcharge %
8 3% Test Adjustment (1)
9 3%TestRateAdjustment
10 Adiusted Proposed Decoupling Re€o/ery Rates
11 Adjusted lncremental Decoupling Recwery
12 Adjusted lncrementalSurcharge96
Notes
Residential Non-Residential
s 110,175,000 s
L24,577,6t9
S0.06157
5o.02927
So.orzso
4,023,857 $
3.65%
(718,s77) $
-s0.00s77
s0.0ss80
3,305,044 s
3.m96
34,499,000
56,682,4tL
so.orgoa
s0.02108
s0.01796
1,018,016
s0.03904
1,018,015
2.95%
s
(
)
(1) 2016 Normalized Revenue derived from UG-170486 Revenue Model with billed rates
adjusted to reflect August 1,2017 present rates.
(2) The carryover balances will differ from the 3% adjustment amounts due to the
revenue related expense gross up partially offset by additional interest on the outstanding
balance during the amortization period.
Exhibit No. 1
P. Ehrbar, Avista
Page 60 of 224
Page l-48
6
Decoupling Mechanism - 2017 Electric (Schedule 75) and Natural Gas
(Schedule 175)
In this section, we review analysis of data from the test year from October 2013 through
September 2014 (a historical test year), which was used to develop amounts for revenue recovery
for calendar 2017. Recovery will occur from November 2018 through the end of October 2019
(the third rate year). The decoupling mechanism is designed to capture all fixed cost assigned
for recovery through volumetric rates that is not actually recovered due to lower sales than
expected during calendar 2017. This cost is recovered by allocation to customer bills according
to a model. The decoupling deferrals total is based on comparison of the value of actual sales in
calendar 2017 to the value of normalized sales (from October 2013 through September 2014) on
a per customer basis. Note that the 2017 deferral year uses the same test year and decoupled
revenue per customer as 2016.
As specified in ScheduleT5 and Schedule 175, calculations were carried out separately and in
parallel, for Residential and Non-Residential accounts. For each of these groups of accounts, the
sum of monthly defenal amounts over calendar year 2017 is the cumulative deferral (rebate or
surcharge). The cumulative deferral (with adjustments for prior year carryover balance, interest,
and revenue related expense adjustment) is collected through the decoupling tariff on a
volumetric basis from November 1, 2018 through October 31,2019.
Electric Group I (Residential) and Group 2 (Non-Residential)
First the electric service analysis is reviewed, then the analysis for natural gas service.
Schedule 75A - Decoupled Revenue oer Customer
For electric service, following steps in Schedule 75A, Decoupled Revenue per Customer (by
Rate Group) is developed. Calculation of Decoupled Revenue per Customer (by Rate Group) is
specified in seven steps in Schedule 75A. Electric tables for 2017 are attachments or parts of
attachments to the Tariff WN U-28, Electric Service, Electric Decoupling Rate Adjustment filed
August 17,2018.
Step 1: Step 1 is to enter the Total Normalized l2I|i4E September, 2014 Revenue, which is
equal to the final approved base rate revenue approved in the Company's last general rate case,
individually for each Rate Schedule. Table l-37,Line 1 shows initial Total Normalized Net
Revenue. In addition, Line2 shows the Allowed Revenue Increase. The sum of Line 1 and Line
2 is shown on Line 3 as the Total Rate Revenue (January ll,2016). This corresponds to the full
value specified in Step 1.
Step 2: Step 2 is to determine the Variable Power Supply Revenue. This value is shown on
Line 6 and is the product of Normalized kWh for the test year from Line 4 and Retail Revenue
Adjustment from Line 5.
Step 3: Step 3 is to enter Delivery and Power Plant Revenue. This is constructed by subtraction
of Variable Power Supply Revenue (Line 6) from the Total Normalized Revenue (Line 3) and is
entered on Line 7.
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 61 ot 224
Page 1-49
6
Step 4: Step 4 is to Remove Basic Charge Revenue. Because the decoupling mechanism only
tracks revenue that varies with customer energy usage, revenue from Fixed Charges is removed.
Basic Charge Revenue is shown on Line 10. It is the product of the number of Customer Bills in
the GRC test year on Line 8 times the Allowed Basic Charge (Line 9).
Step 5: In Step 5, the Decoupled Revenue is equal to the Delivery and Power Plant Revenue
(Line 7) minus the Basic Charge Revenue (Line l0). Decoupled Revenue is shown on Line 11.
Step 6: In Step 6, (see Table 1-38) Decoupled Revenue is put on a per customer basis. The
Decoupled Revenue (by Rate Group) is divided by the approved Test Year number of customers
(by Rate Group). This determines the annual Allowed Decoupled Revenue per Customer (by
Rate Group).
Step 7: Step 7 is different from the other steps because it converts the annual Allowed
Decoupled Revenue per Customer (by Rate Group) into monthly values. The assignment of
monthly values is carried out by modeling monthly kWh use (by Rate Group) in relationship to
the annual kWh use for the test year. This modeling is shown in Table 1-39. Kilowaff hours for
Group I (Residential) for the test year is shown in Line 3 and for Group 2 (Non-Residential) in
Line 6. Both monthly values and the annual kWh values are shown. Below the monthly values
(Lines 4 andT) monthly percentages are shown. Lines 11 and 14 use this percentage model,
applied to annual Allowed Decoupled Revenue per Customer (by Rate Group), to generate
monthly values.
The monthly values developed following the steps in Schedule 75A are then taken forward to be
used in the implementation of Schedule 75B.
Exhibit No. 1
Page l-50 P. Ehrbar, Avista
Page 62 ol 224
UaSe l\OS. AVU-tr- lY-U_ anQ t\VU-\r- lY-U_
Ito<'
EEi[l c <.rL! (o(O
EEelc (Ltbi
Eb
Eloloz
lq)
l6t-)
I
I
I
t_,
I
a)bo
tr
ciz
=€cxuJ
at-En-r@<uatfraaF6
-€s@6,;-=r€.rFt
=ffhrF6E @-al! rd
q^i
2,
N-OFOF6F9hi- O- 6- O- djs@Faao
-oF.rI'E d o.a* 5."'tl d
&
!c6Et6E: -r $E Et.!;*:;gsaca><6(J<
9539
a
ae
!FailtrUt
b!
7
Ei
:;9Orodoa
a
Ei rI*g4a.a<di'Yr(6VeE61'nd, ('
6.u 5T;E6Er6 a)x)E
EE3d(,()
p
so
a
E
!{2
N
I
^l
o
Ito'
9.
6
oo'tFq
6
sa
0-
I
)Iitd!
gao
o
.r. qi €-
000
-- ci o-
o.
do9qdi c.
r60
o.d6-
9.raq.to d
6060do
f
!
F6
R.
d-0 -cgcrn!F=o t5 -&
=., o
:ro lEotF
E 'Yfot5= d()<lo
-o9
3
ei
o
I-
IF.
3
I
dl6
to
.t
o
o.
o
i
J
J
i
6tc
EaI,ordl
!
a
atl
r =ei-9'
r; dtoi1
.!o.l
Go'= d
.!o.l
a
i=Yi
6' oi
.i=o'too@oi
;=da-oF.
9.
G
F^cI
;.i Eeoy@ zir Ed>-t^aa ut
5 x-.tn !-l3 y&
.!JoE:EE g.E'ad>
888o. o" o.
d"d
ooo888+6od
1o, 19v9
g.9 3.
ooo888
1-.o.
oooaaa
d!11
ooo660ooo
-i ei €;
888o. o. a.do.rq-i@"
tt
,F=6:^c qE? H6
R !-'
CS ^4od
H I.Eai;EIEspbI GdE*€z tdg€tts<F
F)!24:a,d,<:Ea9F
9.,
7,3a5oo
,9,a
7,;E-)*du
Ue.t
7fto;ouaoJ
U>".o-.
4
H6r(,
S-tsJ2,JrOOu
pv
IFo
!,Hit?Ef,94
-t{.! z, .r'5:15E EnrEao-
132.9eL!
;A
o
Ecaoz
ae
tU
q
U\q)
q)
q
s_
Q
a
U
Uu
r..r\-
\U
p
ei)a
F.-
sa\
F-.a
I
q)
^ars
@
Ito*
tsgNE<6[, t'$s-39Z,Eol( uJ -!Efo .Ilc (Ll6
fr
bi
F;1I
B
Kt;lo2
| 0.,lal6D
c\(-)
I
Ub.Bq
ciz
=€sxtJJ
&
o7,
a
,t
q)
f.l
oo
E\J.2u
50.^t2.I5e-
= ao{,
:7 e,6l ,:
!9r=rf.aZ's.!
EE
EE
o
o9
s.q)
o
Va
U\q)
\)
q)
q)q
S
$
Uq)a
U\
Uq)
tJ
f.\sc\
od.a
I
q)
F-
@
.+
c>Noio,-oiI!9m<=.ao-6lUm
^|.Jc-lL-NetI:o= N2AE.io () o
O=ir-.JO;);*;8sEs=+e+E616.2BEilaAEO*
N6
at00
alr
ctI
6al
qr
F
ro\
9) Va
oo
@. v,-tf,FONaq-o
o99G
CLGjoa v-Otlt-vIAa
-4 iDail
E E &,F
9 EE ie!a99{5 62a;
Of iarl.
6
TL
o00o.
N
ai
\oo\o
n+
t
o-f,.t clOr6el elat -t
ri o.
I (E.{'
?.E N#itC crri,3 9.acatl*c(LG
Ic,)
JJ
l
aoz
o)
o)
eo!n
I
q)
boq
ciz
=€txtU
\
er)
4
U\q.)\
q.)
q)
\)
a_
Uq)
r_l
U
L
(..)q)
t.L]\
\
o'
q)
o
q)
q)
l-)
f.-
\
o\oo
I
qJ
t\
@
,t
!I
q
'q
B
c
I
g
-i
CG
I.tt4
fis?E?t
iTEE3ersf
iq4
!!r
E
t
;
t
d
ai
fl-iut a.6
! EaffueI
/tt! ?H
fuir
-o:i
E$
s.E
:
3S
gd
xf,
1
o- .:
E-c
E
o
P
.!
[:
a
Iu-!!
-r'tf-'
t
3
4
il!
3,;
E$
EgFI
6::
ri
dhq
:$
F--
.aF
I
I ;',
7i t:
!
!!a tE:'o
3',
a
i^ trn. 7Fr .il
ixif
6
9
6
5
i
o
to
d
i
a
I
,!
4
7
e
I.t
.3!!&a
t
3uI
.E?:+alel9E*&itt
f; rE'rva
":6 .I
tsi.
-Eer.l t
o
E
Eao
A
i;
6
Schedule 758 - Monthlv Decouoline Defenol
Schedule 75B specifies the method for developing the Monthly Decoupling Deferual for electric
service. For Group I (Residential), the calculation of the monthly decoupling deferral for
January 2017 is shown in the top part of Table 1-40.1e For Group 2 (Non-Residential) the
calculation method is shown in the bottom part of Table 1-40. In the full version of this table
(Table l-41), the monthly decoupling deferral amounts across 2017 sum to the annual total
decoupling deferral for 2017. For the Electric Residential, deferred revenue for 2017 is negative
with a value of ($2,092,790). For Electric Non-Residential, deferred revenue for 2017 is
$1,735,911.
Residential Decoupling Deferrals (Step 2 of Schedule 75B) corresponds to Line 3 in the top part
of Table l-40 and the top part of Table 1-41. It is calculated by multiplying the number of
Actual Customers (Line 1) by the Monthly Decoupled Revenue per Customer (Line 2).
Residential Actual Revenue collected in a month (Step 3 of Schedule 75B) is shown on Line 4.
The Residential Actual Basic Charge Revenue (Step 4) is shown on Line 5. The total revenue
collected related to the variable power supply (Step 5) is shown on Line 8. This is the product of
Actual kWh Sales (Line 6) and the Retail Revenue Credit (Line 7).
Residential Actual Decoupled Revenue (Step 6) is calculated by subtracting the Actual Basic
Charge Revenue (Line 4) and the variable power supply revenue (Line 8) from the Actual Base
Rate Revenue and is shown on Line 9.
The Monthly Residential Deferral Total for each month (Step 7) is shown just below Line 12.
This is the difference between the Actual Decoupled Revenue (Step 6; Line 9) and the Allowed
Decoupled Revenue (Step 2; Line 3) plus any interest on the deferral. Interest on the deferred
balance accrues at the quarterly rate published by the FERC. In Table 1-41, these values are
cumulatively incremented by month over 2017 on Line l3 and the electric deferred revenue for
2017 shown on Line 13 with the value of minus $2,092,790. This is the Residential value given
by Avista on page 2 of 5 in Tariff WN U-28, Electric Service, Electric Decoupling Rate
Adjustment filing in compliance with Commission Order No. 05 in Docket No. UE-140188 on
August 17 ,2018. Since the value is negative, Electric Residential does not receive a surcharge.
For Electric Non-Residential, Schedule 75B specifies the method for developing the Monthly
Decoupling Deferral for electric service. In the full version of this table (bottom section of Table
l-41), the monthly decoupling deferral amounts across 2017 sum to the annual total decoupling
deferral for 2017 (for Electric Non-Residential) of $1,735,911. This is the Electric Non-
Residential value given on Page 3 of 5 in the Electric Decoupling Rate Adjustment filing in
compliance with Commission Order No. 05 in Docket No. UE-140188 on August 17,2018.
Since deferred revenue is positive, Electric Non-Residential receives a surcharge. The surcharge
is adjusted by the Earnings Sharing Deduction, the Prior Year Residual Balance, and by Revenue
Related Expense Adjustment for a final Customer Surcharge Revenue Amount of $1,170,966.
'e Only the first few columns of the table are shown here to keep the table readable on the page.
Exhibit No. 1
Page l-54 P. Ehrbar, Avista
Page 66 ol 224
0
The calculations and the Excel programming are identical for Electric Residential and Electric
Non-Residential.
Table 1-40. 20 I 7 Electr"ic Deferral Calculations
^\'isre ftilities
Ilecoupliag lleclaaism - UE-I502(X Base cffectite lill/2016
DerCopuert of I1 -l Electric Deferrels (Crleoder l'car 2017)
Lile r-o.Sourte JrDlT
I
3
(r)
Re*leotietGroup
Aetud OrEonrs
Itouthly Decorpla<t Rertauc per
Cusonrr
DeorpledRertaua
!lo+Residodel Group
.{crurl CunoErs
Ilonthll' &coupled Rcvcaua pcr
Custorrx
Dccoupled Rertaue
(b)
Rrrtnue $'*m
AEafulcat {,
hge 3(l)i C)
(c)
212,1 3{
s88J2
t 18.73626r
.l
6
?
8
9
Acnal Besc RateRcrtoua Rcrtnuc $$an g 19977..1.10
AaudBasicCbagrRcrtaue Rfllsrrc$'icm S 1,836,153
4s61 gegr (lc[ts) F.crtaua $*ro 330,{10875
Att &ncot {,
Rct il R.rau.Cndrt 6L\\l) Page I S 0.016,11
tadabl! Poslr Sr.Fply Prlumts (6) x O) S iJ2!:08
CusocrDccoupltd Palcats (.1) - (5) -(8) S 22,719,078
Residcorirl Rcttmrc Per CugomcrReaiEd tl07-10
De&rral - Surchargp (tubate) (3) - (9) ! (3981"817)
Dc&rnl - Rcwnuc Relrkd Erpcoses Ret'CmvFaoor t 181,732
FERC Rete 3.509'.
Intcr?s oa DcFrral .ArgBalaaa Calc S (5j{2)
llonth'Re*latielDe&rnl Totels S (3r0sr2t)
Cwnuluirt Rcidratirl &fcr.l
Surdaga(Rcb:ta)Balaaa I((10)-(12)) S (3S05,628)
l0
t1
l2
l3
ll
l5
l6
Rcrtauc $sco
Attrc!0ent.l,
Prgc 3
(l.l)x (15)
3 5,883
s362Jr
t r3,008,00r
17 Acual Berc RetcRcrtauc R.rtntt ${.dr S 18,19!j80
18 ActudBaicChrrgpRertoue Rertmrc Sirrcor t l,566Jil
19 Aoltal Usrg? (tdlAs) RnrmE $s.rn 135988,820
.{tudlocar 4,
20 R.teil RcrlaucCtldil (l't\la) hge I 3 0.016{l
2l vziable Pov,tr Sryply Patc.ots (19)r (20) S 3,052,077
22 Cu*orarD*ouplcdhlurus (14 -08){ll) S 13J7,1,152
]-oa-R.sidatid Rrirmrc kr Cuto@cr Rccehrd tl78 )9
23 Dturat-$rctrgr(Rrbilc) (16)-(22) S (566,151)
2.1 DeErnl -RcrturcRdatcdErpcoss RerCmrFecor S :5975
FERC R:re 3-50':i
:5 htcrcr @ I)ctrnl ArgBrlaoe Cdc t O88)
llonth$r-on-Reeidahl Ildcrnlfods S (5{096{)
Cuurlatirt Son-Rcsideotid De$rral
26 Sur<tn9(Rrbata)Balaae E(C3)-(:5)) t (5{096{)
27 fotelCumledtefl€flricllc{crr.l (13)-(2q S (.lj.l6-i9l)
Exhibit No. 1
P. Ehrbar, Avista
Page67 of224
Page 1-55
\ot-)
I
q)b.
tr
ciz
=€tx
LrJ
Lrr.qJ
q.)'
L_l
(.)
U
ra
o'
q)
qJ
a
F.-
!.c\
*-i\I
ql
F\
@
E
z
Jo
5
E
!
>d
.lr
.EE
i5i
.9 E'E
<l;
.E<
A'
$E
.E:
z
E
dd
s55=8S*s
={*eg**"dd
Ex3r
raa
E
-P-ai ;i8-eg t'
, N E ---s.1s3; = = i:.:i AtrP;;r,^-v: = d rti- =dj=ddJd- u 9q i-ia =gssn-d i-! {d
:3 R iZ; EFtsFCE.*!, E H =Li =ignEdr"'=-
r R E Eii? if af ES'$p.r E f, il: ===aJ=:i-- a U Iti 3h*BErr"'t' I-a dd
: E B V,\i" E$EEAETT" e I r€i =e8.*=+ i jr dd
* E E '.4-=;, E:FEFe*=: i d E;? =f {;$9"*+ J ja d+
- h o --s=-6*<
.- = d =xrEdaidd'' u li St7 XB&HBg; rri+ dd
? 6 s =;4 =*=$=f 5x
.:5 d v:i.' =d:I6d;', U R '!T' EFABBg:-
r.6 s z.i sRFeRgEF
V: = J =JJ!JJ:]" 4; "_".:- =8r433 "
d d: j d j:-
,r f: =i',i gFEI8E*51I d .i ,= =dd!dJir" d s t;14 =s8ss- "d .ri4 d{
i E e ri,i !8$.i;a*t, E E r j; ECES$e"=d ='8 {f
I 3 E" Er.i cE"r.$Et*€, r d i jE EdHegs"-
HBE
-i
ddr!
iSe
-oot=6S6-fiEgtr+d+g
FE6:^<e
j i+= jdxd+id
1a 9 0 nn h'
r;:;+;4;5{fr'i;i:*t;$c33i-! dd
ii:sili:HF$Anr!qsRsg{;+j-a {d
F=6-^SAr,::3&a=;skR_- _-'ri =ddxdd;d! - 3 dd 3
i F E ,ri:':EEsiSe*3
= a g =5+-d€aS5;ge ! :i-! dd C
FF-6-{-_
r !1-s6agi;s-=
n l: 'r.;=EHIEg{A
= 3; !i.i6e;AdH;g
6 d dd
z = - -5-a-<5: a { si-.;EEde+xid{
c i (::1 dd
r tr - -F-;-la: a a '.;;agFltr;Hh: E I r;rECgai*;gc n-9 dd
i: A :L-;sES?SA*$! E d 5i9E:5;BFSd=a I a4ooQ
- a - a9F^-:&:.a S ;1',4R5il;tri3.t I A , : J = d d I d d J d
6 d ;d i
*h E'{
L;Ai3a4
:3g
*
EI
"l
=
d9
Rts.ci=
g6
EE
e-
TEJP
:gi!
.i
39
EF
i39J
a6
.rEs
^-=.
nat
3:;;3id3d
-s -;d;daE'
;ic=aH26
E
d
d
q
a
x
a
j
a
GF
sts-
e
EF.dn
3Hiis
9gdd
s6€R
6A
it
;.&-a? a!gY S:ge u.EiE ;E6i s:HE et-:'6"r ;
=!t<<9d-,6?
gEE!.a E:9 ^.* qa:.e-i:: . 3EehE€9 = IiEiE"Et =l't: i E & E : E lEtYf,:E?7 ,zYd
t i E* E E EEEi* s d # 6 6 ! rdE
Ez
9h
i.!!
9 E 22vaEa,i'+ = A,i6B E: EE3
eEza9&^
E?I222
b
: &-
= nEa ^ ii ir i i6 0< EA 6i F.z z?5 i
:E
...;99PE a ;,6 Faaa{€: 6geg =n n n i t h ; ( r, E ,;;;EE; 6-iXE 3
iE3*'d
&.q!
eE iSqj!; i iEr^9!{&E+ d iJsiEE!ea"* =laiJ E-fi,: gF 5; E:EP
* f ;: I Ec t g E {EissE9A:Eoe ,&!,i,
=;==; F E E E i=e9::: ! E;.; E E 5 E Eii i i & s,i e E E ! s-€
^ a{e
-?i=EE-tt;E
eS
^r -{ E
{ E 3d 5'; Et I
9E E
iE
los
3gNr<EC t'coi394Ea:18c(L(o
Ic
')J
l
o;oz
c)o(Ut)
o
Naturol Gas Group I (Residential) and Group 2 (Non-Residential)
For natural gas, following steps in Schedule 175A, Decoupled Revenue per Customer (by Rate
Group) is developed. Calculation of Decoupled Revenue per Customer (by Rate Group) is
specified in seven steps in Schedule 175A. These steps are implemented in Table l-42 and
Table 1-43. Monthly Decoupled Revenue per Customer for Group 1: Residential and Group 2:
Non-Residential are then used to develop the Monthly Decoupling Deferral for natural gas,
following the steps in Schedule 175B.
Schedule 175A - Decoupled Revenue oer Customer
Step 1: Step 1 is to enter the Total Normalized Revenue, which is equal to the final approved
base rate revenue approved in the Company's last general rate case, individually for each rate
class. Table l-42, Line 1 shows initial lzlll4E September 2}l4Total Normalized Net Revenue.
In addition, Line2 shows Allowed Revenue Increase. The sum of Line I and Line 2 is shown on
Line 3 as the Total Rate Revenue (January ll,2016). This corresponds to the full value
specified in Step 1.
Step 2: Step 2 is to determine the Variable Gas Supply Revenue. This Variable Gas Supply
Revenue is shown on Line 6. It the product of Normalized Therms by rate schedule from the last
approved general rate case from Line 4 times the PGA Rates from Line 5.
Step 3: Step 3 is to determine Delivery Revenue, which is entered on Line 7. To determine the
Delivery Revenue, the Variable Gas Supply Revenue is subtracted from the Total Normalized
Revenue.
Step 4: Step 4 is to calculate the Basic Charge Revenue. Because the decoupling mechanism
only tracks revenue that varies with customer energy usage, revenue from Fixed Charges is
removed. It is the product of the number of Customer Bills in the test period on Line 8 times the
Allowed Basic Charges (Line 9). The result, Basic Charge Revenue, is shown on Line 10.
Step 5: Determine the Allowed Decoupled Revenue. The Allowed Decoupled Revenue is equal
to the Delivery (from Line 7) minus the Basic Charge Revenue (Line 10). The resulting
Decoupled Revenue is shown on Line 11.
Step 6: In Step 6, Decoupled Revenue from Line 11 is put on a per customer basis. The
Decoupled Revenue (by Rate Group) is divided by the approved Test Year number of customers
(by Rate Group). This determines the annual Allowed Decoupled Revenue per Customer (by
Rate Group) as shown in Table 1-43.
Step 7: Step 7 is different from the other steps because it converts the annual Allowed
Decoupled Revenue per Customer (by Rate Group) into monthly values. The assignment of
monthly values is carried out by first calculating the distribution of monthly therm use in the test
year. This calculation is shown in Table 1-44.
In Table l-44,thermuse for Group I (Residential) for test year is shown in Line 4 and for Group
2 (Non-Residential) in Line 8. Both monthly therm values and the annual therm values are
shown. Below the monthly values, percentages (Lines 5 and 9) are shown. Lines 14 and 18
Exhibit No. 1
P. Ehrbar, Avista
Page 69 of 224
Page l-57
6
show the use of these percentages, applied to annual Allowed Decoupled Revenue per Customer
(by Rate Group) to generate monthly values.
These monthly values are then taken forward to be used in the implementation of Schedule
175B.
Exhibit No. 1
uase Nos. AVU-E-I9-U_ and AVU-G-I9-U_
P. Ehrbar, Avista
Page 70 ol 224
Page l-58
(E\iES
<tst'-(oF-€or
TE
,l,)
9l
ocG
If,
,)
rl
f
oozooo)
o\!n
I
q)b.
\
oz
-_acx
LU
P
so
Uo
Eo
83
\o
,
o
r'Fe-o6
d.
oF
E g.g.
+r'
ooo666666@oaorodiO
-.
s.tE
ci r;
lrl
$o
H
5,o
B
3EkL5
aU
?,
It-
IUJo(,J
o3
o ooo
g<+d666Y@H-66(5 h"€_ €.O..i?''831;9.O 9--- Orlar6
€h
4E2o
hFFftOao6+6
=6"€.14a.=@<HrO:-.t6@E - r- <>-r-"E o6-'d at-Udc
q
Or
\o.
r_.
ts
o.*6o_fl
F
dl
aal
at
aa r-
t--?6+O\o-; qOoh
d66€6F.t-os-Hs.4o-1r@
ooo888oo' n: vits\ot1 @".1
60F+6@q.o. +-
€6ro-i \o'
rts-€oo6
afreh-oqoaatt
EEg.I @.1
6- 6. d!
6
ooo r € a al t666 6 € o .a600 6 @ h on:+-' \o' J o\ oi + vihd@ @ 6 6 @ 66@6 6 0 @ .r
OOr s \O , - (}t 6+-6 h € @ 6d
ot\,
aa
at6-
oal
@
aa@6.€€
oro60\osiFOO€ei -i
t,)
sI.t1dh
D-v
tr.lo
i=
EH
fH
n(,
n
E
n7,
-iEEda
a4aoEc>;a
E3,toa,OE*,?&63
FT
=oE9
.2. a
o
E
E
o:E
l;A
c,t-oF
all:]€
E]
EE]gEdo9ElLJ "ql
E T FE E]
rEEEH>ada>1<<dro<1., - + -.oJ
a
Ei,&E
a
o
o
b
t
For+68rBe0 -agY"g*hE
-,'ilE3 *E!6A* qlto.t6?,f,€o9
6c-.t
E-J
)I
Ed
E.:.?or
3
u
F.axga&,3<vH8 6'=e E
d o-4E?$4 =6
d-o.!=oEiE.E:-b
:Ealr=^d B\o
=THR S-:-E;BS ^[$E:l-E;EEE:pBg ie?deEzEc34F<F
Lq)
4
U\q)
q)
q)
a<
q)
U
a,)r_l
U
e
o'
q)
\o
q)
qr)a
b.\
\
N\
I
qJ
-a
F.\
@
l(Eto-N
+atO sor!-r OFP€ oraEaElecO_(!
Io
6i
ul
5I6o2.
d)ao
L)
\o
I
a)bo
Q..
ciz
=€ExuJ
to r,a-\o \o
MN\o o\c{ -
g-f-tr66q" ql .i6dm\oAqr;\>6
.A (A
it
7.
FVI\O66Gmo\:€atO\tm6N.N
aA ta
q
(,6!
A.G
;HNvIfia
=>.4iDHe
I
,,)
x
.h
E
orr6ri -q
O\ O\.'l o.d\o
g!&,
fiodi?.#Q iU&lt o4'1,
^uG
>l!odto
f&
€
€Z
Ei
E6
rEU2.,*&
-^qz>.
oEGO()E
Fo
=4)zz
o
q)
q
U\q)\q)
q)
q)
a<
sq)
oUq)a
d
Re
F..r
S\
"a;\t.
I
q)
d
o
o
oa
o
rhT
€
<>e!6oL
LLr lllH=20N9
oL)eEu-Ao"etES
o a) oo=o&vd Itoiuood'or'-oazao;iool:o
NM
@
Itov
Hii
HEFlc (L
HID
H
11t6lo2
lorl6l6t)
I
I
\o
I
\)b.
tr
oz
=€ExIJJ
es
d-;
F
c
q
g
J
CF
lo
T4
t
. isi,it8{
iss
E8
o
&
&:
.t
q
a
:3
s 165iIi
TI ts
IrIclE!1j}
lsi $
ae
=uE-
ei{o.t
.l
P26';
;rd
,t
a
eit
!5
q.'
a{
I!ino
f-'.
!I
i3
Et..pJT,rll,Iiri
i{i
Bf,
g"
rl
xa
a.
a{
B-
tlrl
BA
a
{-s
Br{$.::
e*
9=a
IrT*f403'
i
,r{IE! x"{ i,g
f,r{i,1X1''
3
g
3
13
I
t
5
1l
1
E
tA
d
a
,
a
I
I
a
I3
a
.t
o
i
a
7:
I
E
5E:U
ET
^>. t
isesi;; U!Ea+
'3a,-o,2,
i:
",eI
EIAo!
t,I
L.q)
q
r,\q)
q)
q)
a<
Qq)aq
t,,
E\{
.o
\
a)
o
q)
\0a
f.\
\
\\I
q)
NF\
o
d..ar 69=i1 alI9t=9
,:.
6
Schedule l75B - Monthlv Decouplins Deferral
Schedule 175B specifies the method for developing the Monthly Decoupling Deferral for natural
gas service. For Group 1 (Residential), the calculation of the monthly decoupling deferral for
January 2017 is shown in Table l-45.20 In the full version of this table (Table l-46),the monthly
decoupling deferral amounts across 201 7 sum to the annual total decoupling deferral for 2017 .
As shown in Table l-46,the annual total decoupling deferral for Residential natural gas is
negative 51,972,082. The annual total decoupling defenal for Non-Residential natural gas is
$840,286.
The individual steps in the Schedule 1758 procedure are shown in both Table 1-45 and Table
r-46.
Step 1: Step 1 is to determine the actual number of customers each month. For Group I
(Residential), this is shown in Line I of Table 1-45 and Table 1-46. For Group 2 (Non-
Residential), this is shown in Line 11.
Step 2: Step 2 is to multiply the actual number of customers (Line 1 for Residential; Line 11 for
Non-Residential) by the applicable monthly Allowed Decoupled Revenue per Customer (Line 2
for Residential; Line 12 for Non-Residential), which was developed in the Schedule 175A
procedure. Allowed Decoupled Revenue for Residential is shown on Line 3. Allowed
Decoupled Revenue for Non-Residential is shown on Line 13.
Step 3: Step 3 determines Actual Revenue collected. For Residential, this is shown on Line 4.
For Non-Residential Actual Base Rate Revenue (Excluding Gas Costs) is shown on Line 14.
Step 4: Step 4 calculates the amount of Actual Fixed Charge Revenues included in Actual
Revenues. This is shown on Line 5 for Residential and on Line l5 for Non-Residential.
Step 5: In Step 5, Actual Fixed Charge Revenue (Line 5 for Residential; Line 15 for Non-
Residential) is subtracted from Actual Revenue (Line 4 for Residential; Line 14 for Non-
Residential). The result is shown on Line 6 for Residential and on Line 16 for Non-Residential.
At this point in the calculation all fixed charges have been removed, leaving only variable
charges. In Table l-46 this is shown as both Customer Decoupled Payments in total and as
Revenue per Customer received.
Step 6: In Step 6, the difference between the Actual Decoupled Revenue from Step 5 (Line 6 for
Residential and Line 16 for Non-Residential) and the Allowed Decoupled Revenue from Step 2
(Line 3 for Residential and Line 13 for Non-Residential) is calculated. The resulting balance
(Lines 7, 8 and 9 for Residential and Lines 17, 18 and 19 for Non-Residential) is the Deferral
Total.
Within Step 6, Line 7 for Residential and Line 17 for Non-Residential is the Direct Deferral
(which is either a Surcharge or a Rebate).
20 Only one month is shown here to keep the table readable on the page.
Exhibit No. 1
r"g
Page74 of 224
6
Revenue Related Expense are stated on Line 8 for Residential and Line 18 for Non-Residential.
Below this, the Federal Energy Regulatory Commission rate of interest (FERC Rate) is stated.
Then, the result of the Average Balance Calculation is stated.
Line 9 (for Residential) and Line 19 (for Non-Residential) show the amount of Interest on
Deferral. Below this, the result is the Deferral Totals.
For Residential, the Deferral Total has is a negative $1,972,082. This result is reported by Avista
on Page 2 of 5 in the Natural Gas letter of transmittal from Patrick Ehrbar to the Commission
dated August 17,2018. For Non-Residential, the Deferral Total is $840,286. This result is
reported by Avista on Page 3 of 5 in the Natural Gas letter of transmittal regarding Tariff WN U-
29, Natural Gas Service, Natural Gas Decoupling Rate Adjustment in Docket Number UG-
1 40 I 89 from Patrick Ehrbar to the Commission dated August 17 , 2018.
Exhibit No. 1
P. Ehrbar, Avista
Page 75 ol 224
Page 1-63
0
Table 1-45. 2017 Natural Gas Deferral Calculations
Arbte L:tilities
Decoupling }lectrntsul - UC-150205 B.se effectire Yll/2O16
D6'elopmeot of \YA ltrturrl Gls Deftmb (Crlmder Yeer 20f 7)
Line r-o.Source
Pro Rered
Ja n-I l
)
3
(a)
Rcidrotiel Group
Actual Cu{om€rs
\loably Decorryl.d R.rauc pct
Customer
Decoryled Rcltouc
Actnd Uiagp (idoreetiooal onll)
Acud Brs Ratc Rcrtauc.l @xcludes Grs Co$s)
5 Actud Fixcd Cbergc R*taue
6 CustomerDeoupledPalunts
Resideatid Rcrcaue Per Cugomcr
Receirtd
7 Deeral - Surchrrgc (R.brtc)
8 De&ral - Rcrtaue Relatcd Expcases
(3) - (6) s (3,137,! 15)
R.r'CoatFacto( S 1.t3,266
FERC Ratc 3.50c4
.4rgBalaocc C:lc S (.1,366)
s (2rr8rr5)
r((7)-(e) s (2,998,215)
(b)
Rcrtmr S1:rem
Att chrt 5, Prg?
3
(l) xG)
Rcrtauc S1':rcm
(o
r56..115
t61.49
3 9.618,092
t7,3m.256
Rcrtauc S-r'stem
Rrrrau Slscm
({) - (5)
s t{,178.1.r3
s l,{2:,936
3 12,755,207
s8l J4
9 latereE on Defcrrd
ilon6ly Resideodel Ddernl f oab
Cumuladl! Rcsid.odal Defenal
Surcharp (Rcbatc) B alaocel0
n
t2
l3
Soa-Resideatiel Group
Acnnl Cu*ooers
\f otrthl!' Dccolplcd Rcrau. per
Custoorer
Dccotpled R.crtauc
Rcrtnuc S1!em
.{ttecknmt i, Pigr
3
(ll)r (12) S
Rertnuc Sy*em S
Rerauc S.rtem S
(1{1 - 115' S
(13)- (16) S
RerCoavFactor S
FERC Rate
1,866
s753.65
2, r 59,958
9,021,8?8
t.68?.1 09
319.691
2,367,1t7
3826.0.1
(207,159)
9,,17,1
3.i09,6
t.l
ti
l6
Actual Usagr (iaformrtio8.l od])
.{ctud Ba* Rate Rcrtrue
(Excludcs Gas Coss)
Acud Fird Chargc Rcrtnue
Customer Deoupl cd Paycats
Non-Rcsidcadd Rerroue Pcr
Cu*oma Rcch'ed
De&ral - Surhrrge (Rcbrte)
Defrral - Rertauc Related Erpca*s
Iotel Cumuh th'e I r nrrel Ges
Dderrrl
lotereionDeferral ArgBdaaccCalc S (:gg)
Ilonrhl.r lton-ResHenttel De(rrel f orels S (fr8,!7t)
cumulatirtNon-RcsidtottalDcferrd r(17)-(r9)) s (19g,37.r)Surdruge. @ebatc) Eatao
7I
It
l9
20
!l (10)* (20) S (3,196,189)
Exhibit No. 1
case Nos. AVU-E-19-0_ and AVU-G-I9-0_
P. Ehrbar, Avista
Page 76 of 224
Page l-64
l6\t?Erl
?EEg t'F-i,3:1rd:18c(LG
I3
c)
J
l
aoz
q)oo)
ta)\o
I
a)bo
\
@
ciz
=pExtu
a+
I
i
j
ti
t_a
ei
?
Esa*
{?4&
'2. tzl
EE
E
Isi
irEs:Itr, !,EE Ee"d
.ia
sBa,,IF,
a'a
8.t
ace$A,tr€
aa.r6i:
- .l '!X}H E.,fi F' Ii;
gi oad{
66
:, i n ,, PE;: E :t9'j6 I ; F 6 ro.ltr i af o id-i rl oiiler A ; :i93 Ji R=
6 -66 -6e 4 e ;- fr*e ? nfida-1q?ni--fn=q" r * ; bFr d ".;
6 6A6 6-
h n * s E:s: s3*''E$ i,pqhSs=-
s Q a :? B;:4 6xis- 1! or q ao-r- :l r-{-n"53'j
"hE3sE'-
E fre:1
3 Adr38+heEz
oflo
r'6 a. +;d-,eFqn
eia he i*ii f,
6[IT-E $5ZI:F E9ttdd
*{o
E{- 5
ce
a,
sa
^ sl a
6eBr Po9FG=re .,
sQs ite iaA'8doi
gQs F6,= i- .qe E.-s. n9:
sQr 7,6CF€e gag E:
F6
6
ET
tI
d
t,3
,
E
Ea
$
7I,
F
g
6
z
I
.,,
F
J.,
F
t
t
t.
I
4
F
€
rl
,6
F
et
g
6
g.
ei
a
3
8"
o.
t3
,.t
8.
3.a
<J
.{
st
aa_
E
a
61
9.
et
6
3"
*
.t
I
ei
ga.t .tqc
o
e
ts E.I {!! o?EETE gi! "$ei cBe,7t
E7Eeaz
I
- A';';# Bf "
eBs B
3! 45.
aEa 8
e: 3 a".a
566
d, -: :a ;di;F..a -- a..l
- 9 .{ - --h ,;i a d 6 i6b -'
^ P3:'e r:l*ag F €- a --tF-6r
6 666
a
E sHd
* -6;?
L
.q)
'\.1q
!U
\
AJ
q)
I.\
\
\cj\I
q)
-a
F\
r 5 E fr rax 3 aFS;; B e- ; rsE * si-I .--
n
' 66i 66s E e ;r eaa i =64!a- ,l dr 1 ,.q^" g vr-.F." ii P s 333; *tr-- € - dq$ -
ti
6 666 66t4 * a r,sa8a3i
" A ?' ,i sBs fr F-i-
.i
E *${
sc,
.', :8 ;I o-ia
o F - o 3 6-;?
6 I A v. ioF 6 FGF"ag g gf,iihg ';
;4J-i
6$c E p r Hca € sidc 9 9 6 .i1 F e €.tn.i E ; .d ridoi !I -ivq.; 3- 4 Eaq... --
6 606 66si::r r.Le63ai!6- E F- ,_ '..tq:! F.-h''ts$: b fimfisg,.
iE= slr{jj. ,fEi:
issg$lsgiri
;!;13:.
,; i-'ida
4 rSi
.:
q
3al+
-cv=!l
Elfi.} a
?r-iiirdd4
?!U
:as--.tn9a
13
I
oEE. eoo6fi
4
tataE4gd{ieP"€i
--:EE*{dd
o
I6
4,
{dt
!
3
-E
ETT,tftr
IE$I-
EtSP-! d-
'irlj:;te Etrl5
FoRE'
(r
.E .Egg
ciEE
Retrf9
!,,=
47,E12"
E;2-
= oE'= EaJ--dc !1.0ac9TYC{ao
T.&at4
a
d
.t
.,
2017 Earnings Test
The decoupling mechanism, in Schedules 75D and 175D provides for application of an earnings
test,2l separately for electric and for nafural gas.22
Schedule 75D - Electric Earnings Test
According to Schedule 75D, the decoupling mechanism for decoupled electric customers is
subject to an annual earnings test based on the Company's year-end Commission Basis Reports
that reflect actual decoupling-related revenues and various normalizing adjustments. As shown
in Table l-47,Line 3, the rate of return on a normalized basis in 2017is 7.41%. This exceeds the
7.29% allowed retum established by Docket No. UE-150204 (Line 4).23 The Excess ROR is
0.l2yo, corresponding to Excess Eamings of $1,852,833 (Line 6). A Conversion Factor is
entered on Line 7, which is divided into the Excess Earnings to produce Excess Revenue (Line
8) of $2,986,551. When the 50%o Sharing (Line 9) is applied, the Total Eamings Test Sharing for
electric is$.1,493,276 (Line 10).
Table l-47. 2017 Electric Earnings Test
2017 Conmlstlon Basb Ea.nlnts Test for Decoupllnt
Rate Base
Net lftome
calculated RoR
Base ROR
Excess ROR
Excess Earnin8s
Corussion Factor
Exc6s Rerrerue (Excess EarningVCF)
Sharing %
S r,8s2,833
5 29855sr
2r Information on the background of the Earnings Test is limited to information provided in the Tariff. In response
to Data Request 092, Avista states that "[t]he calculation of excess earnings was agreed upon as part of the
SettlementprocessinDocketNos. l40l88andl40l89. Allinformationregardingtheexcessearningstestis
included in the Tariff Schedule 75D."
22 Rate of Retum is not related to the operation of the 3Yo cap. In response to DR 091 , Avista states that "Rate of
Retum (ROR) is net income divided by rate base for a given annual period. The combination of three elements,
namely revenues, expenses, and rate base, determine the resulting ROR. Changes to the relationship among all of
these elements will impact the actual or normalized actual ROR achieved each year. The 3oh cap impacts the timing
ofamortization ofprior year deferred revenue and as such does not impact earnings or rate base during the
amortization period because surcharge revenues from customers are offset by deferred revenue amortization for a
net income impact of $0 and the deferred revenue on the balance sheet is not included in rate base."
23 Page 6, Paragraph 5 (Commission Determinations) in Washington Utilities and Transportation Commission v.
Avista Corporation dba Avista Utilities, Dockets UE-150204 and UG-150205 (Consolidated), Order 05, Final Order
Rejecting Tariff Finding, Accepting Partial Settlement Stipulation, Authorizing Tariff Findings. Service Date
January 26,2016.
Exhibit No. 1
Page l-66 UASE NOS. AVU.E.] 9-U_ ANO AVU-U-] V.U-
P. Ehrbar, Avista
Page78 of 224
Une No.
I
?
Electric
s 1,513,705,000
5 tt22o2@
0.
3
4
5
I
7
8
9
0
0
The Electric Total Earnings Test Sharing amount is then split between residential and non-
residential customer groups in proportion to their contribution to Total Normalized Revenue
(Table I -48). The split is 51.09% Electric Residential and 48.9lYo Electric Non-Residential.
The dollar values for the split are $762,867 Electric Residential (Line 14) and $730,409 Electric
Non-Residential (Line 15). These values are adjusted to remove various revenue related
expenses by dividing them by the Gross Up Factor derived in Table l-49 (1.047725). The final
values for the Electric Eamings Test are 5728,117 for Residential Electric and $697,138 for Non-
Residential Electric. These values are shown on Line 14 and Line 15, respectively, in Table
1-48. These are also reported on Page 2 of 5 (Residential) through Page 3 of 5 (Non-Residential)
of the Letter of Transmittal from Patrick Ehrbar to the Commission for the Electric Decoupling
RateAdjustment, Tariff WNU-28, Electric Service, datedAugust17,2018, inDocketNumber
uE-140188.
Table l-48. 2017 Electric Earnings Test Sharing Adjustment
Rerrcnue From2OtT Normallzed Loads and Customes at Present Billlng Rates
11
12
13
Residential Revenue
Non-Residential Revenue
Total Normalized Revenue
Eamingr Test Sharing Adjustment
t4 Residential
15 Non-Residential
16 Total
S 23t,2l9,o47 51.09%
5 221,381,43s 48.91%
S 4s2,600,482 100.00%
Gross Rerrenue Net of Ranenue
Adjustment Related Expenses
5 762,867 $ nA,nts zgo,aos s egz,tEe
s
Exhibit No. 1
P. Ehrbar, Avista
Page 79 ol 224
Page 1-67
s
uase t\os. AVU-tr- tY-u_ ano AVU-u- I Y-u_
6
Table l-49. 2017 Derivation of Electric Gross Up Factor and Revenue Conversion Factor
Ure
No.Dsqiptiotr
I Ronrc
f,rpan:
2 l.lncollcuiblc
I Qmnictiro P6sg
1 ['ashinsoo Ercisc Tax
5 Tonl Exparc
6 ll€t Opuitirg Incomc Bcbre FiT
7 Fcdcrrl lncmc Trx@ 35q6
8 REVENf,TECO\TERSIO:\ TACTOR
9 Gross Up Factor
Fddor
0_00501 I
0.
o.
0.0t5J5 t
0.
0.
0.6:0t9:
.7 Commission Basis Conversion Factot
Schedule 175D - Natural Gas Earnings Test
According to Schedule 175D, the decoupling mechanism for natural gas is subject to an annual
eamings test based on the Company's year-end Commission Basis Reports that reflect actual
decoupling-related revenues and various normalizing adjustments. As shown in Table l-50. the
rate of return on a normalized basis in 2017 is 8.32%. This is more than the 7 .29Yo allowed
return (Line 4). The Excess ROR is 1.03% (Line 5). The dollar value of Excess Earnings is
93,226,615. This is adjusted for various revenue expenses by dividing by the Conversion Factor
(0.620530) given in Line 7 for Excess Revenue of $5,199,773 as shown in Line 8.
Table 1-50. 2017 Natural Gas Earnings Test
With the Sharing percentage set at 50o/o, the 2017 Total Earnings Test Sharing is $2,599,887
(Line 1 0). The Conversion Factor on Line 7 is developed in Table 1-5 1 .
Exhibit No. 1
s
2017 Commission Basis Earnings Test for Decoupling
5 Excess Earnings
7 Conversion Factor8 ExcessRevenue(ExcessEarnings/CF)9 Sharing %
10 2017 Total Eamings Test Sharing
3
4
5
1 Rate Base
2 Net lncome
s
S
Line No.Natural Gas
s 313,U4,0m
s 26,os7,om
8.32%
7.29%
1.03%
3,226,615
0.620530
5,t99,773
50%
Calculated ROR
Base ROR
Excess ROR
Page l-68
Page 80 ot 224
o
Table I -5 I . 2017 Derivation of Gross Up Factor and Revenue Conversion Factor (Natural Gas)
A1'ISTA t-TILITIES
Ra'auc Coorcrsior Frdor
\v$hhgtoB - Grs Slrtr[
TNEL\:E IIONTHS E\DED Dcmbo -11. !017
DBriplim Fador
Ll!.
r-o.
I Revcnc
E'rPas*3 Uacollccriblcs
3 Commisioo Fa?s
4 Wr![hffoo Excisc T.x
5 Total E:pasc
6 N.tOp€nthglacocBcbrtFlT
7 Fodaal Incomc Trx @ 35?'c
8 REVENUECONVERSIO}IFACTOR
9 GrossupFact6
2017 Cormissbo Basis Cmrtersion Factor
o,0050t
0.95466r
0..13.r I 3 l
The split between Natural Gas Residential and Natural Gas Non-Residential is developed in
Table l-52. The split is modeled on contribution to revenue. Stated in percentage terms, the
split is 77.II% Residential and22.89Yo Non-Residential (Lines 11 and 12). At the Gross level,
the dollar values are $2,004,793 Residential and $595,094 Non-Residential. When expressed net
of various revenue expenses (by dividing by the Gross Up Factor from Table l-S2,Line 9, the
values are $1,913,898 Natural Gas Residential and $568,113 Natural Gas Non-Residential.
These values are also reported Page 2 of 5 for Residential and Page 3 of 5 for Non-Residential in
Letter of Transmittal from Patrick Ehrbar to the Commission for Tariff WN U-29, Natural Gas
Service, Natural Gas Decoupling Rate Adjustment, dated August 17 ,2078 in Docket Number
uG-140189.
Table l-52. 2017 Natnral Gas Eornings Test Shoring Adjustntent
Renenue From 2017 Normallzed loads and Customers at Present Bllllng Rates
11 Residential Revenue S 104,202,001
L2 Non-Residential Revenue S 30,930,843
13 Total Normalized Revenue S 135,132,844
77.1
Eamints Test Sharing Adlusbnent
L4 Residential
15 NonResidential
16 Total
22
Gross Revenue Net of Revenue
Adjustment Related Expenses
s 2,004,793 s 1,913,898
13
2,599,887 2,482,O7t
Exhibit No. 'l
P. Ehrbar, Avista
Page81 of224
Page 1-69
0
2017 Three-Percent Annual Rate Increase Limitation
Decoupling annual rate adjustment surcharges are subject to a3o/o annual rate increase limitation
(there is no reciprocal limit on rebate rate adjustments). The test is to divide the incremental
annual revenue to be collected (proposed surcharge revenue minus present surcharge revenue)24
by the total "normalized" revenue for the two Rate Groups for the most recent January through
December.
Normalized revenue is determined by multiplying the weather-corrected usage for the period by
the present rates in effect. If the incremental amount of the proposed surcharge exceeds 3oZ,
only a 3% incremental rate increase will apply. Any remaining deferred revenue will be carried
over to the following years.
Schedule 758 - Electric 3% Rate Increase Test
The electric Incremental Surcharge Test is shown in Table 1-53. Following the specifications for
the test limits the Residential Surcharge to 3o/o with the remainder deferred to the following year.
However, division of the Revenue from 2017 Normalized Loads with Customers at Present
Billing Rates (Line 1) by the Incremental Decoupling Recovery @ine 6) results in a value of
negative 5.78% for Electric Residential and a value of positive 0.14% for Electric Non-
Residential (Line 7). Since these values are both less than three percent (3o/o), no adjustment is
necessary for either Electric Residential or for Electric Non-Residential.
Table l-53. 2017 Electric 3% Incremental Surcltarge Test
3i lrGnattd trEh..tr Yd
Um No.
I nrwre fm 2Ol7 NomEfiEd teds r.d
ClBffisrt PEslt liiunf Rat6(ttote f]
2 ilomber 2018 - Odo0er 2Of9 Us.Cc (kr rhs)
3 PreDo*dDoudingR$wyR.t6
4 Pccnt Dedpli,tgrrchrBeSffiry ia6
5 lmcrentJ D€@CanB B.@rv iala3
6 lrcttmntdor@Cir8Rffiry
7 lEemilel fufh.r8t 9t
8 316 T€lr A4u*lmt (tlot€ 2)
9 3x T6r ftrle AdGlmnt
lO MFredPmpedoe(dplin!RmsvRacs
ll Adirsaed hcrftrtal DecodingRffiy
12 Adirra€d tErmffalstrd|ar8r f
R.rid.rri.l
5 21r,219,04'
2.384.16q,302
il@.Reidedi.l
5 22r"381,435
2,r6q4ssr6s
301,584
303,584
tlot€5
(1) Revme ftm 2ol7 rcmliz€d leds ard cuffis .l prerl tilirtr
€ft€diE dm€ ,uh 1, 2018,
(21 I}E e.rywer balaE6 ui! diffs tm the 396 adirsmt amnts doe to
rm Elar€d erpem 8rG up oanialy ofGd by additinl int(es{ s
2a To emphasize, this is a test of an incremental surcharge and this test is a key element in the flexibility of Avista's
decoupling mechanisms.
Exhibit No. 1
P. Ehrbar, Avista
Page 82 of 224
-somrrs
so.00443
-!oro55r
(ljt.!79,184, s
-5.7ara
.9
So.om
-sooor15
(B3rtr8r) s
.'.rti
Page l-70
0
Schedule 1758 - Natural Gas 3% Rate Increase Test
The natural gas Incremental Surcharge Test is shown in Table l-54. The test limits the
incremental residential and the incremental non-residential surcharge each to 3Yo.
For both the natural gas residential group and the natural gas non-residential Broup, the numeric
value of the result is negative. Since these values are under 3o/o,no adjustment is applied. For
both groups, there is no deferred revenue carried forward to the following year.
Table l-54. 2017 Natural Gas 3'% Incremental Surcharge Tesl
396 lncrementd Surcharye fest
l-ine No.
Rey€nue From 2017 t{ormalized Loeds andI Crr,o'nerr.t Pr6ent EillirE Rates {Note 1,
2 Norember 2018 - October 2019 usate
3 Proposed Oecoirdirts Recorery Ratet
4 Prcse.t D€coupliru Surcharge Recorery Rates
5 hcrernental OeouplirE Recorrery Rates
6 lrEemental Decoirplirts Recovery
7 lncrem€ntal grchd8e%
8 396TestAdjrrstrnent(21
9 3XTestRateAdjustment
10 Adiu*ed Proposed Decorrplir€ Recolr€ry Rates
11 Adiusted lncremental Decoupling Reov€ry
12 Ad,usted lndernental furdrarge %
Notet
Residential
s 104,202,00r
r26528,897
-so.02720
so.0ss80
-s0.08300
s (10,s0r,8981 s
-10.08%
Non-Residential
s 30,930843
59,004,176
55
s0.00000
-s0.02720
5 (r0,$r,898) S
-10.08%
(l) Revenue lrcm zOL? normalired loads and customers at present billing rates
since June 1,2018.
(2) Ihe carryorer balances will diffur from the 3% adiustment amorhts due to the
rdated expense Sross up partially offset bv additional interest on the outstandinS
the
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 83 ol 224
Page l-71
Audit Statements: Is the Source Data Credible?
Having reviewed calculations for conformance to Schedule 75 and Schedule 175,the second step
in the Task I analysis is to validate the general credibility of the test period costs and revenues,
balance sheets, load projections, and other company financial data. Since this data was audited
by a professional audit team (Deloitte & Touche LLP) that provides an opinion regarding the
accuracy of the data, we are relying on their professional opinion to validate the financial
integrity of the data.
Attachment A to Avista's Response to H. Gil Peach & Associates Data Request No. 015
provides copies of the Report of the Independent Registered Public Accounting Firm for the
Avista Corporation and subsidiaries for calendar years 2015, 2016 and 2017. These opinions are
based on certified audits of the company's accounting practices. Each Independent Registered
Public Accounting Report expresses an unqualified opinion on the Company's internal control
over financial reporting. These opinions validate the data used to implement the Avista electric
and natural gas decoupling mechanisms.
The Deloitte & Touche LLP "Report of Independent Registered Public Accounting Firm" for the
twelve-month period ending December 31,2015 is shown as Figure l-2. Deloitte & Touche
LLP also provided their financial audit opinions of Avista's reported financial statements for
calendar year 20l6and20l7, as shown in Figure l-2 and Figure l-3.
REPORTOF IN DEPENDEI'IT REGISTERED PUBTIC ACCOUNTING FIRM
Tothe Eerdof Ured06endSharehold€rsof
AY6ta CoooraUon
Spolane, wilhi8ton
wehaye audted the acompenyhgconrolid.t€dbelrnce rhae$ of Ayilb Co.pootioo rrd rub'idili.i lthe'Conpmf!:r ol oecember
3L 20lSandml4,xdtherelatedconrdiratednetementtormcom.comtreh0tEiyei,rcongequivandrcde€nauendrcontoalint
htorest3,rrrdttih flows{rertholtherhreeye*rhlheperiode,r&dDecefibnlL20lS.neaefiEnralsE&meatsmtherrl'oolrtibalilyof
theCo{llpeny', mrnagemeilt.0orrc5ponsililityirtoe4raJsanoFniorortherefinandJstetemeflBbasedgnalraudi6.
Wecooduatedouraudits hatro.d3ncewiththestrndards ofthePublaComp.fiyArcouflrhtOve'3afitEoard(Unitedst.ter.Thos€
srndrdi rpqute thct we Cen and per(rm the audit to obirio rsesonaHe asuame rbol h,heth€, the fnandal qalemmrrre free of.n.terill
misstatement. An audit irsludei €raminiflg, m: ren brsh, eviden(e rufportint the amor,trr and dircloJu6 ir the finaoci.l rutem€ntr Ao
rudil aboifibd€sarsersintth€eccoufltintprifiiphsusedandrigndtinestimrt6madebymana$m€nt,eswelaseyaluatinttheoverdl
financiel itatememp.eg€ot.lion.wEb€lievethstorauditsproviderreasonableb:asforouropinior.
lnouropinhq such conrolidatedfinenrialstatements Ferent{atly, in allmaterial rerpccts,thefinanrielpositioo otAvbu CorpooUon
and subidia.iBat Decembe.3l,2015 a6d 201/t, andthe resulti of theiroper.tiors a^dtheiroshfloys fr eedr dthe threeyeac inthe
penodended oet.mba3l,20l5,incsrformtywitheccountidtprinarplest€refallyacceptedmtheUnat€d$at6olAmerite.
Wehaveaboaudtod,inaccordanc€withth€!rarderdsofth6PublicCompenyAccountinsOve,!thtBoa.d(Unit€dSletesl,the
Company'r intenulcodtroloverfinandilreporthgerofD€cember3l.2015,baredontheoiterbgtablBhedinlnterrlalControl-htegoted
Framewo.t{2013} isruedbythe(ommitteeolSgonsoringo.ganir.tbnsoltllelreadwayGmmisson,andorrrepor!dat€dtebruary23,2016
elpre$ed.nunqu.lified oprnimdlfleCompa0l/iinEmal(ontroloverfinaodJreponin&
/s/0eloitt€&Tou(heLl.P
Sertrle, wa$inglon
tebruly 21, 2016
Figtrre l-2. 2016 Financial Audit Opinionfor Calendar 2015
Exhibit No. '1
6
Page l-72 Case Nos. AVU-E-19-0_ ancl AVU-G-I9-0_
P. Ehrbar, Avista
Page 84 ot 224
REPORT OF INDEPENDENT REGISTERED PUBLIC ACCOUNTING FIRM (2016)
To the Board of DirectoBand
Shareholdersof Avista
Corporation
Spokane, Washington
We have audited the accom panying consolidated balance sheets ofAvista Corporation and subsidiaries (the
"Compa ny") as of December 3 1, 20 16 and 20 1 5, and the related consolidated statements of income,
comprehensive income, equity and redeemable noncontrolling interests, and cash flowsforeach ofthethreeyears
inthe period ended December3l,2016. Thesefinancial statementsarethe responsibilityoftheCompany!
management. Our responsibility is to express an opinion on these financial statements based on our
audits.
Weconducted ouraudits in accordance with thestandardsofthe PublicCompanyAccounting0versight
Board (United States). Those standards require that we plan and perform the audit to obtain reasonable
assurance a bout whether the fina ncial statements arefree of material misstatement. An audit includes
examining, on a test basis, evidence supporting the amounE and disclosures in the financial statements. An audit
also includes assessingtheaccounting principles used and significantestimatesmade bymanagement, aswell
asevaluatingtheoverall financial statement presentation. We believe that our audits provide a
reasonable basis for our opinion.
In our opinion, such consolidated financial statements present fairly, in all material respects, the financial
position of Avista Corporation and subsidiaries at December 31, 2016 and 2015, and the results of their
operations and their cash flows for each of the three years in the period ended December 31, 2016, in
conformity with accounting principles generally accepted in the United States ofAmerica.
We havealsoaudited, in accordancevriththestandardsofthe PublicCompanyAccounting oversightBoard
(United States), theCompany's intemal control overfinancial reporting as ofDecember 31, 2016, based ofl the
criteria established in IntemalControl-lntegrated Franewo*(2013) issued bytheCommittee ofSponsoring
0rganizations oftheTreadway Commission, and our report, dated February 21,2017 expressed an unqualified
opinion on the Company's internal control overfinancial reporting.
/s/ Deloitte & Touche LLP
Seattle,
Washington
February 21,
2017
6
Figure l-3. 2017 FinancialAudit Opinionfor Calendar 2016
REPOR' OF INOEPENOENT REGISTEREO PUBLIC ACCOU}{TIHG FIRi'
fo d( tu.t l*6 i rt( *-d of llN6 oa Are ( dFJ6
(ttu n rk riwd lsrr..rlr
U. br. drd rlE e.Wy'ry drdl.lM MR J'G.6o,Asro (fiIdlo rd it dE6|fi. aots)', ri dl)<d.t.rl- $t7rd !ol($.6d (rdd*derffihn(rftGr. smF.l6*. ffm. ryr!...d.r* &na Lr.xh d * k !r5 - dtf(ddllcrilb.r Jl
)0lr.fidrh.clJnl|&k.llq1hd!Blnrdtr.rfr.'fffiCffih'l Lorol.6*n,ih.llsalMrffil.-h.ril8d
ffiE *<,.xd prn o{rh(i.r(tfr r o( lhrh }t. tolt rd:ol(t il ercdl nad {f.r6 -d rB (a[ *xr tu.aa oJl}( [s
)6 '' rlEFi{ rd ah6*6 }I :olr. rcdmsr} hasmrilt xddf6& -t$edn tu tr.dds&,har.
*? b. & # i afl& rft rk drorft A$lx C..te) A.@htorcrtlr Bd (th{arr 9.5l IR'A(I8I6. AAq'r
rehJgmaffil*Ilqsuiildl}.alEr,1, !llr.lcdqffic*.[lrrH6'*dci*ol. !?tEftrriqtd rJrrrl,Edb
ltE ( ffi !a dsteia Ort.idffi of dx Tt*., ( ffi rd d rtfo! A hrsf :q, &l r..rFs.i r q*rld (ll..rm m t(-dF!i*rtr ld@ adcFlrS
a.nL(l|i5
nr fdffi *ft rwbl0 dtLcoi..ttmS'd (k6,saai}6loqr! -Wdk('q-tt frrC
@L-adod.l&w..r..|lU-rtdtli6,atFdrif drF(' OBDdxlwabhrd.Fildqara6|tcrhlr.('o+r!F..rd..c radE US H6J E!ilE b.t.rdtk#rda. Ebrd tttrlr*oarl.SGssisr{ f&i&arOEKdrL
'CAOB
*..!.ixr.d fl ..drr. -s{-..! rd il. .ahb srtk lt A{B rlrx e isat k r !a- d Flan rh. d b & tEGJc
ffi h rl6lE r: lil*d ffi 4 fia oas*l @r rt air ir b Cffi H. k rJt r.(l.d |Bftr.f ,t EriG
D e rL n|lr n erJ ffi d rt. fffi l ffi rL6.r & t6 af d fr{ d tatilrf f6car.6 ll. Gtarl b lk drsai F..n*r itlH.um4. o ! K ba 6td*..rt &i * ffi ' d l}-b6 . rlE fnEd ffi (L rdir rho *luArl
oil*rt rL effi t Fd6 Edr sl rFfrd w nr.L b! tF.FGr - *u. s*.t$t tt 6sru rwc d ll. frtJw trt tdnr tk fi ,s Fdd.. rrrir.tL b tus w
la r'.!li1l1l
kkB :0. :{t t
*t bi. mrdsrhCry'idtr {K l9ll
Figure 1-4. 2018 Financial Audit Opinion for Calendar 201 7
Exhibit No. 1
uase Nos. Avu-E-l9-u_ ano AVU-U-]v-u_
P. Ehrbar, Avista
Page 85 ot 224
Page 1-73
Summary - Task 1
Based on our analysis of three years of data, we conclude that Avista has calculated rates and
deferrals in accordance with the Commission Order approving the decoupling mechanisms for
the first through the third Decoupling Years.
The purpose of the Decoupling Mechanism is to decouple the Company's Commission-
authorized revenues from sales, such that the portion of the Company's fixed costs planned for
recovery through volumetric sales and not otherwise recovered from actual energ)l sales will be
recovered through the mechanism. In decoupling, the revenue requirement for a given year is
first set. The portion of fixed costs collected through the fixed portion of customer bills is not
included in the analysis. Since volumetric sales fluctuate and may not fully cover the fixed cost
component included within the volumetric portions of customer rates, the difference between
acfual decoupling-related revenue received from customers through volumetric rates, and the
decoupling-related revenue approved for recovery through volumetric rates is accumulated in
deferred revenue accounts.
Operationally, this compliance verification was carried out in two steps:
o First, we traced calculations to insure conformance with Schedule 75(A,B,C,
D, E) and Schedule 175(A, B, C, D, E). In carrying out this analysis, we
checked to see that the reported calculations matched the methodological
specifications in each Schedule. Also, we checked for 2015, 2016 and2017
the component Excel spreadsheets introduced as Avista Exhibits for the annual
filings for Tariff WN U-28 Electric Service for Electric Decoupling Rate
Adjustment; and for Tariff WN U-29 Natural Gas Service for Natural Gas
Decoupling Rate Adjustment as filed on August 31,2016, August 31,2077 and
on August 17,2018.
o Second, we have included the opinions of the independent auditor for 2015,
2016 and2017 to indicate the validity of the financial data upon which the
calculations depend.
The overall result in this section of the analysis is that we find the deferrals and rates to have
been calculated by the Company in accordance with the Commission order and the Amended
Petition, as determined by methodological specification in Schedule 75 and Schedule 175.
Exhibit No. 1
6
Page l-74 uase Nos. AVU-E-I9-U_ and AVU-G-I9-U_
P. Ehrbar, Avista
Page 86 ol 224
0
Section 2. Billin of Cost of Service AnaImacts and Recove
There are two primary evaluation objectives associated with Task 2:
o Determine if there were any differences in decoupling tracker adjustments between rate
classes.o Determine if allowed revenues are recovering the cost of service for group one
(residential) and group two (non-residential subject to decoupling)25 and customers not
subject to decoupling.
Each objective is addressed in a separate section. Both sections use the customer classes (rate
categories) customarily used by Avista for cost of service analysis and for decoupling filings.
These customer classes are listed in the table below for electric and natural gas customers.
Table 2-1. Electric and Natural Gas Rate Groups and Custonter Classes (Rate Categories)
Electric Service Natural Gas Service
Rate Group
Customer
Class Code
Customer
Class
Rate
Schedules
De-
couDled Rate Group
Customer
Class Code
Customer
Class
Rate
Schedules
De-
coupled
Residential EI Residential 1,2 Yes Residential GI Residential l0l,102 Yes
Non-Residential E2A General
Services n,l2 Yes Non-Residential G2A General
Services lll Yes
Non-Residential E2B Large General
Seruices 21,22 Yes Non-Residential G2B
Large
General
Services
121 Yes
Non-Residential E2C Pumping 30,3t,32 Yes Non-Residential G2C Intem:ptible l3l Yes (a)
Non-Decoupled E3A
Extra Large
General
Services
25 No Non-Decoupled G3A Excluded
Schedulesl
fi2, 122,
132 No
Non- Decoupled E3B Street & Area
Lighting 4t-48 No Non-Decoupled G3B Excluded
Schedules 2 146,148 No
(a) No customer history for natural gas Rate Schedule I 3 I (Intemrptible) over the years requested (20 I 2-20 I 7)
2s For customers subject to decoupling, the mechanism captures all fixed costs allocated to the volumetric portion of
customer bills. Avista states in response to Data Request 090 that ". . .on a customer basis there are no costs which
are not captured in the mechanism."
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page87 o1224
Page 2-l
o
For reporting and referencing purposes, we have defined a Customer Class Code for each rate
category. The Customer Class Code identifies the fuel in the first character, electric (E) or
natural gas (G), decoupling rate group in the second and a subset of the rate group defined by
one or more rate schedules in the third. Separately for electric and natural gas, and as explained
in the section of the evaluation covering Task l, the decoupling mechanism defines two groups
of customers subject to the decoupling tracker adjustment, residential (Rate Group 1) and non-
residential (Rate Group 2). We also define Rate Group 3, non-residential customers not subject
to the decoupling tariff. The aggregation level hierarchy listed from highest level of aggregation
to the lowest is as follows:
1. Rate Group
2. Customer Class (Rate Category)
3. Rate Schedule
For example, Customer Class Code El is electric decoupling Rate Group 1, residential, and
includes rate schedules 1 and 2. A third character is not necessary since Rate Group I only
includes residential rate schedules. Rate Group 2 is non-residential customers subject to the
decoupling adjustment tariff. There are three customer classes (collection of rate schedules)
included in Rate Group 2 for both electric and natural gas service. Rate Group 3 is used to
identiff customers not subject to the decoupling tariff adjustment. Electric and natural gas each
have two customer classes that belong to Rate Group 3.
Summary of Decoupling Mechanics and Results
Before examining the impact of decoupling by rate class it is useful to take a high-level look at
the mechanics of the decoupling mechanism, actual deferrals, requested recovery amounts and
decoupling rates. Avista's decoupling mechanism allows for the recovery of the difference
between actual revenue and allowed revenue.26 This difference is referred to as the decoupling
deferral balance and is tracked for the two electric and two natural gas customer groups subject
to decoupling; residential and non-residential.
Beginning in 2015, monthly deferrals are accumulated over a calendar year and used with other
determinants to calculate the decoupling rate required to collect or refund the under or over
collected revenue. Decoupling rates become effective in Schedule 75 (electric) and Schedule
175 (natural gas) November 1 of the year following the year in which deferral balances were
calculated. The timing of deferral balance accumulation and decoupling rate adjustments is
shown in Figure 2-1.
26The details of Avista's decoupling mechanism are included in Final Order ("Order 5") for Docket Numbers UE-
140188 and UG- 140189.
Exhibit No. 1
uase Nos. AVU-ts-] 9-U_ ano AVU-Lj-] V-U_
P. Ehrbar, Avista
Page 88 of 224
Page 2-2
6
Time
--------------2018 -------------
JFMAMJJASOND ]FMAMJJASO ND JFMAMJJASO ND JFMAMJJASO ND
Deferral
Year
Deferral Year 1 Deferral Year 2 Deferral Year 3 Deferral Year 4
Rate
Year
Rate Year I Rate Year 2
/_
\-
Figtre 2- L Timing of Deferral Balance Accurnulation and Decoupling Rate
The first deferral year resulted in a deferral balance at the end of 2015 that was used, along with
other determinants, to calculate the decoupling rate in effect during the first rate year (November
7,2016 through October 31,2017). The same process is followed in the second deferral year
and rate year. Any deferral balance carried over from the first rate year due to the application of
the 3Yo cap is included in the calculations of decoupling rates in effect during the second rate
year (November 2017 through October 2018). Details of these calculations are shown in Table
2-2 for the first three years of operation of the decoupling mechanism.
Table 2-2. Summarv of Deferral Balances and Decoupling Recovery Rates
Electric
Residential Grouo Non-Residential Group
Notes 2015 2016 2017 20I5 2016 2017
Deferred Revenue ($)7,167,748 I 0,288,205 -2,092.790 -2.373.472 1.967.777 1.735.91 I
Requested Recoverv ($)A 7.360.678 10,913,950 -2,76s,63s -3,08t,249 864,012 1,170.966
Customer Surcharse (Rebate) Revenue ($)6.485.021 10.91 3.9s0 -2.765.635 -3.08 1.249 864,012 |,t70,966
Carryover Deferred Revenue ($)875,657 0 0 0 0 0
Decouplins Rate (Schedule 75) ($/kwh)B 0.00263 0.00445 -0.001 l6 -0.00143 0.00040 0.00054
Incremental Revenue (Percent)3.00%2.00%-5.78%-1.40%0.40%0.14%
Limited by 3ok Cap?Yes No No No No No
Natural Gas
Residential Group Non-Residential Group
Notes 2015 2016 2017 2015 2016 2017
Deferred Revenue ($)5,3 17,198 7,152,977 t.972.082 1.736.736 2.002.654 840,286
Reouested Recoverv (S)A 5,750,096 7,652,369 .3,441,586 1.879,152 2.212.881 407.7 t9
Customer Surcharse (Rebate) Revenue ($)3.488.984 6,9s1,431 -3,441,586 1,108,839 2,212,881 407,7 t9
Carryover Deferred Revenue ($)2,261,t t2 700.938 0 770.313 0 0
Decouplins Rate (Schedule I 75) ($/therm)B 0.02927 0.05580 -0.02720 0.02108 0.03904 0.00691
Incremental Revenue (Percent)3.00o/o 3.00%-10.08%3.00%2.95o/o -6.13%
Limited by 3%o Cap?Yes Yes No Yes No No
A: Requestedrecoveryisequaltodeferredrevenueafteradjustingforsharedexcessearnings(ifapplicable),deferralbalancecarryover
from prior year (ifany), interest, and revenue related expenses.
B: DecouplingratesScheduleT5(electric)andSchedulelT5(naturalgas)takeeffectonNovemberlstofthefollowingyear. Forexample,
rates shown in the 2016 column have an effective date ofNovember 1,2017
Exhibit No. 1
P. Ehrbar, Avista
Page 89 of 224
Page 2-3
--------------2015 -------
6
Years shown in Table 2-2 conespond to the deferral years and rate years shown in Figure 2-1.
For example, the 2015 column refers to calculations made from data for deferral year one (2015)
and the resulting deferral rates in effect for rate year one (November 1,2016 through October 31,
2017). As a specific example, consider the workings of the decoupling mechanism as shown for
the natural gas residential rate group in2016. Cumulative defenal balances during the year
amounted to $7.153 million. This amount along with adjustments, including the carryover of
92.261million from 2015 requested recovery not amortized into rate year one due to the 3Yo cap,
resulted in a requested recovery of $7.652 million. For the second consecutive year the 3Yo cap
took effect, limiting the customer surcharge revenue expected from the new decoupling rate
(effective November 1,2017) to $6.951 million and resulting in carryover deferred revenue of
S0.701 million.
An important characteristic of the Avista decoupling mechanism that applies to all rate goups
and fuels is evident in the residential natural gas example. Because the 3Yo test is applied using
current rates, including the current decoupling rate, the new decoupling rate will adjust higher
and be capable of amortizing higher levels of requested recovery.27 At some point, even if
weather or other conditions that caused initially high deferral carryovers persist, the decoupling
rate will eventually adjust to a level that recovers 100 percent of requested recovery and
carryover deferral balances will fall to zero. This greatly reduces the possibility of snow-balling
deferral balances even in the face of persistently warm winters over consecutive heating seasons.
This point is well illustrated for residential natural gas customers. Carryover deferred revenue
fell from $2.261million for the 2015 defenal year to $0.701 million in2016 even though
defened revenue and the requested recovery was nearly two million dollars higher for the 2016
deferral year. Heating degree days were 15% less than normal (warmer winter weather) in 2015
and l4Yo less than normal in20l6.
Factors Influencing Use per Customer
Avista relies on volumetric charges to recover a portion of fixed costs for all rate groups and
fuels. This causes use per customer to be an important factor in determining deferral balances
and decoupling rates through the decoupling mechanism. More specifically, changes in use per
customer from levels used in the test year to set decoupled revenue will lead to positive or
negative deferral balances depending on the direction of change, all other things equal. Higher
use per customer will cause negative deferrals and lower use per customer will result in higher
deferrals, again all other things equal.
Two important factors causing use per customer to vary from test year are acfual weather
deviations from normal weather and acquired energy efficiency savings through Avista
programs. There are other factors of course but these two are either known in the case of energy
efficiency or readily measurable in the case of weather.
27 This is a special feature of the Avista decoupling mechanism that makes the mechanism flexible.
Exhibit No. 1
Page 2-4 CeteX otTVU:E:Tg-O_ an d AW -G- 1 9-0_
P. Ehrbar, Avista
Page 90 of 224
6
Electric
The table below shows calculations for estimating these impacts on electric use per customer.
Table 2-3. Electric Use per Customer Variancefrom Test Year
2015 2016 2017
Usage
(MWh)Customers
Use per
Customer
ftwh)
Usage
(MWh)Customers
Use per
Customer
ftwh)
Usage
(MWh)Customem
Use per
Customer(kwh)
Residential
Test Yetr 2,437,s08 207,850 t1,727 2,378,478 205,172 1 1.593 2.378.478 205.172 I 1.593
Actual 2.323.300 207.37 |11.204 2.288.227 209,864 10,903 2,492,293 212,495 l,729
Chmse from Test Year fl 14.208)(479\(s24\(90.25 I )4.692 (689)I 13.815 7,323 136
Percent Chanse -4.7%-0.2%-4.s%-3.8%2.3%-5.9o/o 4.8%3.6%1.2%
Chmge from Test Year Due to:
Weather (33.120)fl60)(73.659\(351)113,472 s34
Cumulative Enersv Efficiencv 0 0 (33.272\(ls9)(6 l,s00)(289)
Non-Residential
Test Yee 2, I 50,843 35,277 60,970 2.144.8s7 34.823 61.s93 2.t44.8s7 34.823 61.593
Actual 2.179.747 3s.26s 61.810 2. l 5 8.998 35.617 60.618 2,1 84,830 35,994 60,700
Chmse from Test Year 28,904 (r2\840 14.142 794 (97s)39.974 1.171 (893)
Percent Chmqe 13%0.0%1.4o/o 0.7%2.3o/o -1.60/o 1.9%3.4%-t.s%
Chmse from Test Yed Due to:
Weather 10,361 294 (7.200\(202)28.8s I 802
Cumulative Enersy Effi ciency 0 (4 I,935)(r.177)(8 r .076)(2.2s2)
The test year used for 2015 deferral calculations was a projection of 2015. The test years for
2016 and2017 both used a l2-month period ending September 2014. Actual usage, customers
and use per customer compared to the test year are shaightforward calculations. Changes due to
weather are also straightforward calculations, the results of which are also shown in Table 2-3 in
terms of total and use per customer impacts. Avista provided the weather impacts and
supporting monthly details by rate schedule showing the deviation in heating and cooling degree
days from normal and the corresponding model coefficient on each weather term. Energy
efficiency impacts are calculated as cumulative savings from Avista programs since the test year.
One way to quickly visualize the results of the calculations shown in Table 2-3 is a plot of each
factor's influence on the percent change in use per customer from the test year. Figure 2-8
presents this information for the electric residential rate group.
5.@6
3.096
l.0}6
-1.096
-3.Or5
.5.096
-7.W
l---- 201s -----l | ---- 2016 -----l
lTotal Eweather EEnergy[fti(iency OOther
| ---- 2017 ----- |
Figure 2-2. Percentage Change in Use per Customer, Electric Residential
Exhibit No. 'l
P. Ehrbar, Avista
Page91 of224
Page 2-5
o
Considering 201 7 results, use per customer was I .2Yo higher than test year assumptions.
Weather impacts alone are estimated to have pushed electric residential use per customer 4.602
higher. The 2017 weather impact was largely offset by a 2.5% drop in use per customer due to
Avista's energy efficiency achievements. The "Other" category is simply the difference between
the total and the readily quantifiable factors of weather and energy efficiency. Other unidentified
factors have pushed use per customer lower and have been lessening in influence over time.
For electric residential customers weather impacts on use per customer can be large and work in
either direction. It is also true that energy efficiency impacts always push use per customer
lower and that downward influence becomes more pronounced the further in time an evaluation
year is from the test year. Cumulative energy efficiency savings will reset with a new rate case
and test year.
Figure 2-3 shows a plot of total and each factor's influence on the percent change in use per
customer from the test year for the electric non-residential rate group.
l.go
O.lNo
-1.0%
.2.0%
-3.0%
-4.O%
I*-ftl
| ---- 2o1s ---- I I --- 2016 --- |
lTotal EWeather @EnergyEffkiency OOther
| --- 2017 ---l
Figure 2-3. Percentage Change in Use per Customer, Electric Non-Residential
Avista's energy efficiency achievements have been the primary factor influencing changing use
per customer in the electric non-residential group. From having no influence in 2015 because
they were implicitly included in test year assumptions, energy efficiency impacts more than
offset weather and other factors in20l7 causing an overall drop in use per customer of 1.5%.
Weather appears to be far less influential in electric non-residential customer usage than it is for
the electric residential group. Other unidentified factors have pushed use per customer higher at
a small but consistent percentage over time.
Exhibit No. 1
Page 2-6 Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 92 of 224
0
Natural Gas
The same analysis of the factors impacting changes in electric use per customer were also
completed for the natural gas rate groups. Results of the analysis are shown inTable 2-4.
Table 2-4. Natural Gas Use per Customer Variancefrom Test Year
As with electric, the natural gas decoupling mechanism used a projection of 2015 as the 2015
test year. The natural gas test year for 2016 and 2017 both used a l2-month period ending
September 2014- Again, these definitions of test periods are consistent with the electric
decoupling mechanism. The calculations shown in Table 2-6 are also consistent with the
approach described for electric and presented in Table 2-3.
Results of the analysis of changes in natural gas use per customer are visually represented in
Figure 2-4 for the natural gas residential group.
6.Wo
4.0$
2.Vo
O.V.
.2.O%
4.O%
-6.O%
-8.0%
-lO.Wo
-l2.Mo
-74.e4
u
pq
0.096
ffi.*,H
I ---- 2015 ----- I I ---. 2016 ----- I
lTotal trWeether EErergy Effkiency trOther
| ---- 2017 ----- l
Figure 2-4. Percentage Change in Use per Custonter, Natural Gas Residential
Exhibit No. 1
uase Nos. AVU-E-I 9-U_ and AVU-G-I 9-U_
P. Ehrbar, Avista
Page 93 of 224
2015 20t6 2017
Usage
(MWh)Customers
Use per
Customer
(kwh)Usage(MWh)Customers
Use per
Customer(kwh)Usage
(MWh)Customers
Use per
Customer(kwh)
Residential
Test Yeil I t7,ltt,207 1 50,1 86 779 120,72t,607 148,995 810 t20,721,607 I 48,995 8t0
Actual 103.436.220 151.254 684 108.796. I 87 I 53.995 706 131,782,922 157,563 836
Chmse from Test Yeil (L3.s74.987\1.068 (95)(r1.92s.420)5.000 fl04)I l.06l.3l s 8.568 26
Percent Chmqe -11.60/o 0.7o/o -12.2%-9.9%3A%-12.8%9.2%s.8%3.2%
Change from Test Year Due to:
Weather fl s.3 18.639)001)fl0.650.43I )(69)4.404.967 28
Cumulative EnersY Effi ciencv 0 0 (360.660)(2)(93 l.t 20)(6)
-- Non-Residential
Test Year 51,764,097 2.s48 20,316 52,606.812 2,584 20,358 s2,606,812 2,584 20.358
Actual 45.886.568 2.65t 17.309 48.208.894 2.770 17.404 55.684.308 2.918 l 9.083
Chmge from Test Year 6.877,s29\103 (3.006)(4.397.91 8)186 (2.9s41 3.077.496 334 fi.275\
Percent Chmge -il.4%4.0%-14.8%-8.4%7.2%-14.5%s.8%12.9%-6.3%
Chmqe from Test Year Due to:
Weather (s,3s7,641',1 (2,021\(3.63 l,036)fl.3I l)1,407,324 482
Cumulative Enersy Efficiency 0 (687.328\(248)(903,662')(3r0)
Page 2-7
6
Weather is clearly the dominant factor in understanding changes in residential therm use per
customer from the test year. The total change in use per customer tracks the warmer than normal
heating seasons in calendar years 2015 and20l6 and slightly colder than normal heating season
in calendar year 2017. Energy efficiency impacts on use per customer are a small factor in
understanding overall change from the test year. Natural gas prices have been persistently low,
squeezing the cost effectiveness of natural gas efficiency programs. Other unidentified factors
were small in 2015 and20l7 but relatively high in20l6. One possible explanation is that the
2016 weather adjustment was understated by the weather normalization model.
Figure 2-5 shows a plot of total and each factor's influence on the percent change in use per
customer from test year assumptions for the natural gas non-residential rate group.
4.Vo
2.Wo
O.V"
-2.O%
-4.O%
-6.0%
-8.O%
-lO.Vo
-72.Mo
-74.Mo
-76.Vo
l;I
;11
| ---- 201s ----- | | ----- 2016 ----- |
lTotal Ctweather 6 Energy Efficiency EOther
l--- 2017 -----l
Figure 2-5. Percentage Change in Use per Custonter, Natural Gas Non-Residential
Except for weather in2017, all factors in each year have contributed toward lower use per
customer than test year assumptions. Unlike any of the other electric or natural gas rate groups,
other factors are an important influence on use per customer for the natural gas non-residential
group in each of the years examined. Other factors are by definition unquantified but could
include increased efficiency outside of Avista's energy efficiency programs, lower use of natural
gas due to fuel substitution (e.g. increased use of biomass in cogeneration) and cutbacks in
customer facility operations. Weather is also influential although less so than for natural gas
residential customers. Energy efficiency impacts on use per customer are a small factor in
understanding overall change from the test year. Again, this could be due in part to persistently
low natural gas prices putting pressure on the cost effectiveness of natural gas efficiency
programs.
Avista's electric and natural gas energy efficiency programs are discussed in detail in Section 3
and Section 6 of this report. An examination of actual weather experienced over the three
evaluation years is presented next.
Exhibit No. 1
Page 2-8 P. Ehrbar, Avista
Page 94 ol 224
6
Weather Compored to Normul
The impact of weather depends on the level of weather sensitive energy usage and the difference
between actual and normal weather.28 Weather that causes greater usage results in over
collection of allowed revenue (negative deferral balances) and vice versa. Residential is the
most weather sensitive customer group and natural gas customers are typically more weather
sensitive than electric customers because space conditioning makes up a greater percentage of
natural gas usage than electric. Given these relationships we would expect the residential natural
gas customer group to have the largest weather-related impacts on decoupling deferral balances
and rates.
Heating degree days are useful for describing atmospheric temperatures in units related to the
need for space heating. Figure 2-6 shows the difference between actual and normal heating
degree days (HDD) from January 2015 through December 2017. A negative value means
warner than normal weather (i.e., less than normal need for space heating).
Blue bars denote colder than normal
Orange bars denote warmer than normal
/O0
:m
200
100
.m ?iPi ?++?iq+?iFqi ?eTxF; ; eii?? ii|;i; i;! 3 i ig! : {.*8 g,i ! 3 i } g! : IsE E g E 3 } i g! : rss ! 3
Figure 2-6. Monthly Heating Degree Da.ys (61i.12,'ence.ft'otrt normal)
Actual weather was predominately warner than normal in 2015 and2016. In2017 actual HDDs
were much closer to but higher than normal, indicating a return to slightly greater but near
normal space heating loads. As shown earlier in this section, this weather pattern has the
expected impact on use per customer for nafural gas residential and non-residential groups.
Space heating is the predominant end-use for the natural gas residential group and a major end-
use in the natural gas non-residential group.
For both of Avista's electric customer rate groups, the need for space cooling is also an
important determination of use per customer. Cooling degree days are useful for describing
atmospheric temperatures in units related to the need for space cooling. Figure 2-7 shows the
difference between actual and normal cooling degree days (CDD) from January 2015 through
28 For this analysis, normal weather is defined as a thirty-year moving average.
Exhibit No. 1
6'tO_oE3E -CrOoze3oco'=o-uEttep<.!
i!o
I
r ll..f f
uase Nos. AVU-tr-t u-u_ ano AVU-U-] 9-u_
P. Ehrbar, Avista
Page 95 ot 224
Page 2-9
6
December 2017. Anegative value means cooler than normal weather (i.e., less than normal need
for space cooling).
Orange bars denote warmer than normal
Blue bars denote colder than normal
,rll
200
(o
-o(! B1s0trlt
Ooz>
3 Srooco'= o,
-EEt'"tsoo<.9ootJo
Source: DR40
I
I
I
-50 666nOOOO.O09O@O@Oi n tfi i r i n.r.'t 7 i n n a i r ? it! E iiiiists ! i! jigii iliE!i tiliiii r$E ig
Figure 2-7. Monthly Cooling Degree Days (dffirencefrom normal)
As shown by the monthly bars, significant warner than normal weather was experienced in the
summer months of 2015 and20l7 and somewhat warmer than normal in the summer of 2016.
This would have led to greater than normal levels of electric loads for space cooling in both
residential and non-residential rate groups in all deferral years, especially 20 1 5 and 2017 . The
increased usage for space cooling would put downward pressure on deferral balances, all else
held constant. This is especially true for the non-residential group where space cooling is likely
to be a larger percentage oftotal usage than the residential rate group.
Monthly heating and cooling degree data are summarized for each of the three calendar years in
Table 2-5.
Table 2-5. Comparison of Acrual and Normal Annual Heoting Degree Days
Heating Degree Days Cooling Degree Days
2015 2016 2017 2015 2016 2017
Actual 5,61I
6,629
-15.4%
s,610
6,547
-143%
6,72s
6,513
3.3%
828
4',77
73.6%
494
4'78
3.3%
794
490
62.0%
Normal
Percent Difference
Source: DR 40 and DR 76
Weather can be an important factor, along with energy efficiency achievements, contributing to
acfual use per customer variances from projected levels. It stands to reason that decoupling
deferral balances are related to weather patterns. Holding everything else constant and
considering just the variances from normal degree days shown in Table 2-5 it would be
reasonable to expect deferral balances for both natural gas rate groups to be positive in 2015 and
2016 and near zero or slightly negative in 2017 . This is in fact the pattern of deferral balances
observed in both groups. For electric rate groups the presence of cooling loads makes it more
Exhibit No. 'l
fase Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 96 of 224
Page 2-10
r I
Year
El:
Residential
E2A.
General
Services
E2B:.
Large General
Services
82C..
Pumping
E3A:
Excluded
Extra Large
General Services
E3B:
Excluded
Street &
Area Lishtine Total
2012 202.s41 28,868 2,440 2.416 22 357 236.644
2013 203.883 29,622 2,050 2,427 21 375 238.378
2014 205,621 30,570 2,011 2,435 21 381 241,039
201 5 209,419 3l,089 2.027 2,44s 23 400 245,403
2016 209,864 31.286 1.903 2,433 21 409 245,916
2017 212.495 31.666 1.896 2.432 22 413 248,924
6
difficult to explain deferral balances solely on the weather when heating and cooling differences
from normal cause use per customer to move in opposite directions. This is the case in 2015 and
2016 when less than normal HDDs put downward pressure on use per customer and gleater than
normal CDDs put upward pressure on use per customer.
Earlier in this section the balance of these offsetting weather impacts was quantified and
described along with energy efficiency and other factors impacting use per customer. Especially
for electric rate groups weather is only apart of the story for understanding usage and energy
efficiency achievements are an important factor in determining changes in use per customer.
Avista's energy efficiency achievements are described in detail in Section 3 and Section 6 of this
evaluation.
Task 2Part 1: Impact of Decoupling Tracker Adjustment by Customer Class
The objective for the first part of this task, as stated in the request for proposal, is shown below:
"An ossessment of the impacts of the Decoupling tariff tracker adjustments,
calculated in relation to energy sales (kWh/therms), as a percent of monthly bills,
and in total dollars for each rate category customarily used for purposes of Avista's
cost of service analyses."
Relating to this objective is the following evaluation question, also taken from the RFP:
"Were there any dffirences in Decoupling tracker adjustments between the rate
classes? "
We begin our analysis and reporting for this task with electric customer classes followed by
natural gas customer classes.
Electric
Six years of historical customer counts by customer class are shown in Table 2-6. Although,
Rate Group 3 is not subject to decoupling, Customer Classes E3A and E3B are included for
completeness and perspective.
Table 2-6. Annual Electric Custonter Counts by Customer Class
Avista serves approximately one quarter of a million electric customers in the state of
Washington. All but about 400 of these customers are subject to the decoupling tracker
adjustment. Customer growth has varied year to year consistent with economic conditions and
Exhibit No. 1
P. Ehrbar, Avista
Page97 ot224
Page 2-l I
0
construction activity, averaging about one percent annually for residential and slightly higher for
non-residential customers. As discussed in the previous section, although the decoupling
mechanism was effective January 1,2015, the decoupling hacker adjustment did not show up on
customer bills until late in 2016. Customer growth in20l7 was near the average of the 2012-
2017 period,l.3oA for residential (slightly above the average of 1.0%) and 1.0%o for non-
residential (slightly below the average of 1.3%).
Annual revenues by electric customer class over the 2012 through 2017 period are shown in
Table 2-7. For perspective and completeness Rate Group 3 customer classes are shown in the
table even though they are not subject to the decoupling mechanism.
Table 2-7. Annual Electric Revenue bv Customer Closs
Year
Residential
E2Al.
General
Services
E2B:
Large General
Services
E,2C:.
Pumpine
E3A:
Excluded
Extra Large
General Seruices
E3B:
Excluded
Street &
Area Lishtine Total
(thousands ofdollars)
2012 193,907 s9,984 129,863 10,068 58,697 6,772 459.290
2013 20s,149 67,922 126,981 10,431 6l,sl I 6,694 478.687
2014 208,603 70,884 I 28,958 11,576 64,35s 6.932 491.308
20t5 208,022 73,727 133,362 t2,5t6 70,931 7.201 505.758
20t6 207.40s 74.978 129.316 tt^26s 66.571 7.089 496.624
2017 237.119 78. I 86 130.454 1 1.396 68.445 6.776 532.376
Avista billed Washington electric customers $532 millionin2}l7, up over 7o/o from 2016 due
primarily to the effect on residential customers of a return to colder than normal weather. Like
most electric and natural gas utilities, Avista's billed revenue varies significantly with the
weather. Eighty six percent (86%) of revenue in20l7 was collected from Rate Groups 1 and 2,
and subject to the decoupling tarifftracker. Total revenue and Schedule 75 revenue are shown in
Table 2-8 for these four customer classes. Schedule 75 revenue is the revenue collected through
the decoupling adjustment mechanism.
Table 2-8. Annual Decoupling Tariff Revenue by Electric Customer Class
Electric Customer Class
2016 2017
Revenue
Schedule 75
Revenue
Percent
of BiIl Revenue
Schedule 75
Revenue
Percent
of Bill
El: Residential 207,405,033 82 1,1 87 0.4%237, I I 8,808 7. I 68.350 3.0%
E2A: Ceneral Services 74.978,073 - 106,490 -0.1%78, I 85,893 -777,980 -t.0%
E2Bl. Ls. Gen Services 129.31s.832 -236.728 -0.2%130.454.356 -1.723.065 -1.3%
E2C: Pumping 11,265,056 -6,223 -0.1%11,396,0'73 -188,410 -t.7%
In 2016 Schedule 75 revenue amounted to a small percent of the overall billed revenue for a
customer class. Schedule 75 adjustment to rates first took effect on November 1,2016, muting
the annual 2016 impact. The decoupling adjustment amounted to 0.4%o of 2016 residential bills.
The customer classes in the non-residential rate group (Group 2)had slightly lower bills in 2016
due to Schedule 75. The difference in the direction of Schedule 75 impact on billed revenue
between rate groups is due to deferral balance differences shown in the previous section.
Exhibit No. 1
Page 2-12 Case No-TV[I-F-19-0 and AVU-G-1 9-0
P. Ehrbar, Avista
Page 98 of 224
El:
0
Schedule 75 revenue is significantly higher in20l7, the first full calendar year with Schedule 75
in rates. Although still small in percentage of revenue terms, Schedule 75 accountedfor 3%o of
billed residential revenue in2017. The billed revenue impact was negative for Group 2
customers, ranging from -1.0% of revenue for General Services customers and -l .7o/o for
Pumping customers.
The pattern of monthly impacts, discussed next, provides insight on what to expect for 2018.
Summarizing impacts annually is useful at a high level but a monthly view is necessary to
examine the paffern of usage and impact on bills from the decoupling mechanism. Monthly
details are shown by electric customer class for 2016 and20l7 in Table 2-9 and Table 2-10,
respectively. These tables show total usage, revenue, meters (customers), average usage,
average revenue, and Schedule 75 revenue (total, average, and as a percent ofrevenue) for
customer classes subject to the decoupling mechanism.
Monthly revenue impacts follow the pattern of volumetric sales. As a result, customer classes
with high seasonality also show high seasonality in the average customer's monthly Schedule 75
charge. Due to weather induced seasonality in monthly usage, the surcharge paid per customer
varies significantly by month for the Residential Rate Group, ranging from a low of $0.86 per
customer in November 2016 as Schedule 75 began to be phased into customer bills, to a high of
$5.10 per customer in December 2017.
A review of the monthly data in Table 2-9 and Table 2-10 shows that the percentage impact of
Schedule 75 ontotal revenue tends to be relatively constant from month-to month. The months
of November and December can be exceptions and show significant differences in Schedule 75
revenue percentage from preceding months. This is due to the November 1 effective date of new
Schedule 75 rate adjustments. For example, the Schedule 75 percent for the General Services
class went from -l.lo/o in October 2017 to 0.3% in December 2017 as the new Schedule 75 rate
effective November 1,2017 became fully reflected in customer bills.2e
2e Although the effective date of revised Schedule 75 rates was November l, customer bills in November reflect
usage that is partially billed at the old Schedule 7 5 rate and part billed at the new Schedule 7 5 rate. The portion
billed under the old and new rates is determined by a simple prorating of usage based on the number of calendar
days in the billing period before November I and the number of days on or after November l.
Exhibit No. 1
Page 2-13 P. Ehrbar, Avista
Page 99 oi 224
l(5$O;N
cir '< o'tT<bO;oI(!0JOr)E ar:18co-[o
Io
o)
UJ
l
U;ozoooo
\
I\
bo
a,
c;z
=€
xIJJ
Fl
FoF
o
6INd
o
rad
I
q
N
N
6
F
N
6
atrlQ
E]q
J
riz
r-,i
)
de
r r
r6
F
a.l oi
r6d)N Io
ariQ
&
q
,.1
r.lzH
rd
,l
didr.I
Nr6
N
c)
6N
o 6
r r dr
N
d
.lo
z
eD
ONri
rrr N
N
o E Nd
I
N
\o
q)
I
rr
N
n
a
N
N
N
r 6 N
6€
N eN N
F
r
$.
NN€-
6
N .lo N
N N
6
N
F 6 d
II
6 N
ts
6\N
I
o
N
6
N
N
N q
\o
I
z
6rr
I
r
Nts
o9 N
N r r6.!aqo
oo q oI$.6
N
d N
o
6
6 6
r
r +di€
s r
r N
I oI o no
\o
o
Io
oieo
Fa ts9.
N
r o o eo
rN
r
*"
ts
n o o r
r
or
o
6
o
a
vl o qo o.oo
oi
N
o
N
o
d
6 d o oqo q
\o
oO
aoN
Ir
o
I
e N
6 rr o qo r
o
d
o
ts
I
tsoN aro
N
rar
6 oo
I
o o tsq
Nr.
N
6
Ir d
N rr o oqo qo
\o
IAI
ts
Nq
r
6N
N
aN
rr o o
o
N
I
Ir
Nrr N
I
doN
o oq .!ts
6 r r
I
N
6
o oo
o r
N
o
6
N
6
6
6r o o
\o
I
I
N+-
6 3.!6oN
r r o qo
r
r
rao6
I
N
6 d6 o Ia a
Irq
Io
6 6
6da
o o q 6
N
N
ts
r N
r Nr o o
o o
\o r
o6
N
r6
I
N
rr rI o oqo o o
ts
6
Nr
ror
N I o oeo qo
o r
o
N.l
I
Ir o oeo Io
r
N
r a r
o oe s
a
\o
a
N
I
6
6 o
N
or o aq Ic]
r
ro
I
o.1
o o q Nrr
d
N
r6 r
6
ro o o r
NrN
Nooo
+-
N
oqo o
\o I
N
r
$
Its
o
oal
r Ir o oqo qo
ao
rN
r
q
r o
I
6
Nr
€
o
6 6
6 o oo 6
II
rr 6
N
a
N d
o oo qo
\o
a
rov')
6
rN€
6
ar6oN
6
o o o\
o r
NN
rN a
o N
Nro
o
I
N6
6
r
€
F
N ao r
6 6{N
6 6 o oqo Io
\o
,!.otr o
N
II
oN
N
6 66 o oqo sq
N6
6aor
I
N
..1
rrr d
o oeo sq
IN
6o
o
N6 Nv.
66 o oeo sq n rI
N
N
N oqo
\o a
ts!l:
N
ts
I
N
ts
N
N
o ol o oea e
N
I
r
q Noo
N
o oq sq I
N
o
a'I
6 a
I
6r o oq sq
Na6
6
Fjts
N o o
t
g
D
F
a
&EoF
c
E
z
a
d
dg
)tr
d
r
!
o{
ot
I
^o
>
q
oF
o
!
E
z
a
g
ts
!
d
oq
A
)EaF
atr
o
p
Ez
I
&
E
c
r d
o
c
c
-I
st
D
oF
o
E
=z
g
tr
r
!
G
.a
d
o
a
c
r
=tr
d
r
tr
r
t
oF
-
d
r
!
a
d
a
b^
aa
U
L.
qJ
\Jta
\o
s\
oi
I\
q)
RF\
@
ll(EsD;NPii[9 c-La (50UOr
Err
l',Dbii;ut
tsIt6loz
lcrlal(E
f-)
!.}
I\q)b.
tr
c;z2!cxL!
ll
FoF
I
Fzt.]
or,
Ttrl
Ir
r
Na
N
a
r
N
6$"
N
rr
r
\6\q
ardo
&ria.]
r{zH
iNri
o
trts
6 r
aIE Nod
I
N
a
qrrr
rq
d
N
E
I
t
€6€-
N
r
ts Nr
oq
o6
ts
6N
r
F
6
N
N
rrt r
9-o RNr
r
F-
I
o
NFI
a
N
Nr
N
rr
N
o
N
6
o sI IFFtsts
r r
r 6r NN oN
o
N
oIN o6 6
I
Fd
o
on
N no cl
(jI
Fo r
N
66r aa .l
r-
I
z
N
6
NooI6.1
I
N
o6 r
I
q a
6oi
N
N.lI
aa
r6 N
o
N
N
aN
ro
Io
€
sri ..)6r
oe r I
o
NNr N
o
N
N6 ooI
ooq I
F-
I
5\
N
6r or rr q ?N N
I
N
N
o n IN
I
e ql
6
rNr
N
6
r
r I6r o
N 6 a
r
6NaN
N
r @
F
ts-
I
oa
r
a N
6
eN
N
r rrcJ N ?N r
N
r r II oN
r
nd
d
I
N
66a.
rq
o
6 N
Ir
Iq
N6
6o
N
N
F
oi
Nr I6
F-
Iot
6NIN
NaN
6@roo N
N
N
6
6 oI
o
d 09
N o o
a
@
I6r,r 6oN
rr
r N 6
o
N
N o6 Fo oo
6
ts
ts
oc
6
o o
o
o
Ir
N
N
N
F6N
d
6 or r
F-a
6
r
N
o
I
6 N r N
od
F
N oq
d N
6
I
tsNoI
Nr I 6
dor
dc]d N 6
r o
6
6@.N
6\r
6
n6
N
€-
N
r.'l
N
N$-N 6
Fr e
r-
I
a
6 6 N
Nr Iro
og
N
rrotsoI
r6N
r 6N
rI
N
N rr Nr
a6 o
o a
I
q
I
NaIoo NN
aN I
NN
N
Na:r
r-
I
z
6r
I
F
d
d
rr r .1 9N
s09
d
oI
ri
N 6N
a
ro
N
r
o9
c5
o 6
€.
F
F
6 .'jN v")
r
a
r
Ns-N
6
o IN 6
o
N11 a
t-
rl
rtsoir
o
I
6r
ool
N
o6 N c1N soq
d
6
o I
a6
r
I.lN N r Nr
r6 N r
N
6
c)
dN
N
Nr N I svt
F-.
2
N.i
rNd d
o Nro o N66
Noe
N ?N
ts$.
N
66r
I
a r N
N N
or
ov]
N
N
F
N
o
6 6Ieo€
N ro9-
dI
o6 €
r IN
r
N
6q r 9 6\v,l
F-
I!ol4
q
s..1
rd
rN
0N
6N N q
r
*N r
I
r6$-
Fj
o6 d6q I
r6
oq
N
cl d
3.
oq
d
3
E
--l
6 NN
a
6N r
N
r
o
6
o
d.
r
r
o
d
r
o9 ts o
ctI IN
r-.
I
6t
6
oN
N
NN
N
?€
rq o9
N
6r
6I
6€o
d
N r
N
N
oi6
N N
do rFts
o-ots
N
N
a
aor oo
I N c.!
ts
N
No\I
N
v1
f
oF
z
o
E
z d
r
t
oF
ac
d
r
o
c
>
:F
o
p
Ez
-
&
!9
r d
r
!
d
o
?
D
,o
E
E
z
tr
&
N
9
&r
a
oq
a
c D
-o
o
6
E
z
)
F
I
q
r
!
a
o
I
)
&
oF
&
oF
d
_oF
I
&
-ats
d
d
S
!\br
ca
.o
U\U
aJ
r-r\
F'\
\
I\\)
N
,9F,,\
@
0
To visualize and contrast the impacts on customer electric revenues between customer classes,
the percentage of monthly elechic revenues attributed to Schedule 75 from the time rates were
first impacted by the decoupling mechanism through December 2017 is shown in Figure 2-8.
5.096
't.0*
t.0*
2.0*
1'o%
oo*
t,096
2-&
3.&
.l
o DEC. JAI{.I'FEB.I7
"ARt6 17
AP Rt?I.IAY, JUII-I7JUI..I' AUGl7 17
SEP. OCI1t l16
Figure 2-8. Schedule 75 as a Percent of ll{onthly Customer Class Revenues
Figure 2-8 shows monthly Schedule 75 revenue as a percentage of total revenue for each
customer class subject to decoupling. The impact on revenue of the first decoupling tariff
adjustment effective November 1,2016 can be seen beginning with November 2016 billed
revenue. Residential customers saw the partial impact of Schedule 75 in November 2016 and the
full impact in December 2016 with Schedule 75 revenue accounting for 2.8oh of revenue.
Schedule 75 revenue as a percent ofclass revenue increased again with new rates effective
November 1,2017 . In incremental percentage terms, the 2017 increase was smaller than the
2016 increase in Schedule 75 revenue. In December 2017 when the full impact of the second
year decoupling adjustment is reflected in rates, Residential Schedule 75 revenues were 4.6Yo of
total revenue, 1.8 percentage points higher than the first rate adjustment revenue impact. As
indicated by the long straight line near 3o/o for the Residential group in Figure 2-8, Schedule 75
was limited by the 3o/o arcrtal cap in the first rate year but not in the second year, resulting in the
smaller incremental increase of about 2Yoinrateyear two (from around 3%to 5%).
For Group 2 (non-residential) customer classes, Schedule 75 had the impact of lowering
customer bills with the first rate year adjustment (effective November 1,2016). On a monthly
basis, the full impact of Schedule 75 as percentage of total revenue ranged from -l .lohto -l.8oA,
depending on the month and customer class. This effective rebate from decoupling was reversed
with the second rate year (effective November 1,2077), resulting in Schedule 75 as apercentage
ofrevenuesranging from}.2o/oto0.4Yo inDecember20lT. The3% ratecapdidnotimpact
electric Group 2 customer classes in either 2016 or 2017.
Natural Gus
Six years of historical customer counts by customer class are shown in Table 2-1 1. Although
Rate Group 3 is not subject to decoupling, Customer Classes G3A and G3B are included for
completeness and perspective.
Exhibit No. 1
Page 2-16 P. Ehrbar, Avista
Page 1O2 of 224
uase t\os. AVU-tr- tv-u ano AVU-b- tY-u
0
Table 2-11. Annual Natural Gas Customer Counts by Customer Class
Year
Gl:
Residential
G2A:
General
Services
G2B:
Large General
Services
G3A:
Excluded
Schedules I
G3B:
Excluded
Schedules 2 Total
20t2 146,776 2,476 25 5 46 149328
20r3 I 47,880 2,498 26 4 49 150,457
20t4 149,453 2.575 26 4 48 152,106
2015 152,182 2.648 26 4 43 154,903
2016 I 53,955 2.749 22 4 44 156,774
2017 157,563 2.896 22 4 45 160,530
Avista serves approximately 160,000 natural gas customers in the state of Washington. All but
about 50 of these customers are subject to the decoupling tracker adjustment. Customer growth
has varied year to year consistent with economic conditions and construction activity, averaging
l.4Yo anrnally for residential and 3.lyo for non-residential customers. As discussed in the
previous section, although the decoupling mechanism was effective January l,2015,the
decoupling tracker adjustment did not show up on customer bills until late in20l6. Customer
growth in2017 was higher than experienced over the 2012-2017 period, 2.3Yo for residential and
5 .3o/o for non-residential.
Arurual revenues by customer class over the 2012through2}l7 period are shown inTable2-12.
For perspective and completeness Rate Group 3 customer classes are shown in the table even
though they are not subject to the decoupling mechanism.
Table 2-12. Annual Natural Gas Revenue by Custonter Class
Year
Gl:
Residential
G2A:
General
Services
G2B:
Large General
Services
G3A:
Excluded
Schedules I
G3B:
Excluded
Schedules 2 Total
(thousands ofdollars)
2012 103.264 32,161 3,t76 1.s46 3.297 143.444
2013 1 08,1 36 32,719 3,255 1,184 3,s06 148.801
2014 114,968 36,439 3,520 1,060 3,597 I 59,584
2015 107,638 33,807 ? 11{t,027 3,686 149,493
2016 102,989 31,098 2.441 928 4,121 141,577
2017 123,005 3s.230 2.467 879 4.673 166.2s4
Avista billed Washington natural gas customers $166 million in2017,up 17% from 2016 due
primarily to a refurn to colder than normal weather and to a lesser extent rate changes between
the two periods. Like most electric and natural gas utilities, Avista's billed revenue varies
significantly with the weather. Ninety seven percent (97%) of revenue in2017 was collected
from Rate Groups I and2, and subject to the decoupling tariff tracker (Schedule 175). Total
revenue and Schedule 175 revenue are shown in Table 2-13 for these three customer classes.
Schedule 175 revenue is the revenue collected through the decoupling adjustment mechanism.
Exhibit No. 1
P. Ehrbar, Avista
Page 103 of 224
Page 2-17
o
Table 2-13. Annual DecouplingTariffRevenue by Natural Gas Customer Class
Natural Gas Customer Class
2016 2017
Revenue
Schedule 175
Revenue
Percent
of BiIl Revenue
Schedule 175
Revenue
Percent
of Bill
Gl: Residential 102,988,637 6t4,363 0.6%1 23,005,058 4,499,375 3.7%
G2A: General Services 31.098.227 162.110 05%35.230.221 1.253.729 3.6%
G2B: Large General Services 2,441,368 13,015 0.s%2,467,144 94,787 3.8%
In2016 Schedule 175 revenue amounted to a small percent of the overall billed revenue in each
customer class. Schedule 175 adjustment to rates first took effect on November 1,2016, muting
the annual 2016 impact. The decoupling adjustment amounted to 0.6oh of 2016 residential bills.
The customer classes in the non-residential rate group (Group 2) experienced a similar Schedule
175 impact, 0.5% of billed revenue. The 3Yo cap on Schedule 175 impact on rates was hit in
both the Residential and Non-residential groups in2016 (effective November 2016).
Schedule 175 revenue is significantly higher in20l7, the first full calendar year with Schedule
175 in rates. Although still small in percentage of revenue terms, Schedule 175 accounted for
3.7% ofbtlled residential revenue in20l7. The percentage of billed revenue for Group 2
customers was 3.60/o for General Services and 3.8o/o for Large General Services. The 3% cap on
Schedule 175 impact on rates was hit in the Residential group but not in the Non-residential
groups in20l7 (effective November 2017).
Summarizing impacts annually is useful at a high level but a monthly view is necessary to
examine the pattern of usage and impact on bills from the decoupling mechanism. Monthly
details are shown by natural gas customer class for 2016 and2017 in Table 2-14 and Table 2-15,
respectively. These tables show total usage, revenue, meters (customers), average usage,
average revenue, and Schedule 175 revenue (total, average, and as a percent ofrevenue) for
customer classes subject to the decoupling mechanism.
Monthly revenue impacts follow the pattern of volumetric sales. As a result, customer classes
with high seasonality also show high seasonality in the average customer's monthly Schedule
175 charge. Due to weather induced seasonality in monthly usage, the surcharge paid per
customer varies significantly by month for the Residential Rate Group, ranging from a low of
$0.36 per customer in August 2017 to a high of $6.35 per customer in December 2017.
A review of the monthly data in Table 2-14 and Table 2-15 shows that the percentage impact of
Schedule 175 on total revenue also varies with seasonal usage. Because space heating in natural
gas homes tends to be a much larger percentage of total annual usage than electrically space
heated homes, volumetric charges dominate billed revenue during space heating months and fall
off significantly during the summer. In summer months fixed charges make up a larger
percentage ofbilled revenue causing Schedule 175 revenue as a percentage oftotal revenue to be
lower in swing and summer months. In20l7, Schedule 175 revenue in the residential customer
class fell from 3 .3Yo of revenue during the winter months of January through March to 2.0Yo in
August 2017.
Exhibit No. 1
fase Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 104 ol 224
Page 2-18
l.=
HEil19 croLr ooD--E-E orEUI PlEtol(E
f,r
b\
F;1ll
B
k
lalo2
lorlalorJ
o\
Ic\
q)
boso_
ciz
=€tx]U
!
!
b0
a\
4Ir\
!L
=\o
Uc\
\
I\\)
*a
b\
@
,t
F
F
.)
FzrI]
U)rI]&
rN
6
r
€€6
N
@€E I 6 o\
af:lQ
>&
aJ
rI]zrd
td
€
N
rdN€a
6+r
N
r *
N€
@q€6\v')O
aHU
rrla
"l
rizri
ti
Fl
,iiN
I
r
@I-l
\f,\f,.
N
NN
r
*d
f;oI
o\
€
o
N€
ar
6
@\o
Nr €o o\6\6
a
@
s
@"
N
r
N
6 .l
$
o\ra
€N
€
$N \o
6
\f,
ar a O..1
Na
6\c1r
\e
tz €
r
€
d @N I 6\N
€
€*
so
r$N
6@r:
N N
r€N o\
O
@c)€
N
N €rN c.lI
6\R
\e @
v
€@a
N
<t@ N N N O o\rr
€
\tN
@€
6
r\N
ra O 6\O
O €N
a$N
6
N r
N
Nr..16
o o\
€
a
<f,r<t€N
d6
N $
r N Oq
O
o\€
@r
6rc.l
6d)
rr
N
{Oq sqO
N
r:
N
N
c]
r
$N €€6
Nr
O 6\oO
€
U
Ns
N
++\
di
<fr oN OO
6\@€Rt
n
6a
r
d
$N O 6\O
@NN
r@€
$N rN
a a
O o\IO
€arr
N
r9^r$<f
+
r NN o\O
N
r@
r$rN
rr
N
r N@ O
o\
@
N
6
\tr
N 6€r s6c]r
Oq 6\
I
r-
r
@N
€rr
*
*Ncl N N o\q €+@
<t
as\N
6I a s
l o
N
r
6
N rrNO
IOr
OO
6\
\o
2
@€
aN$
66N
No9
<f
6 @N O 6\q N+
@
at
6
6€r
c:N
a
@
N
I
O OqO
6\6
€N
N
N
N @
NNo"
@
O ?O
o\qo
€
L
6N6
6
€$IrNq€
o @ Oq
O O
N
a
$r
$@N
r\N
@ N 6\r
a
N
6N
N
sN O$
Nrqa
a 6\
O
€
t
z
!tr6
@
N
<t
N
oN
@ r O O
O $$
@
6
c.l€€q
\t
N
€N6c.l
O O
O
o\qO
6rs
@
N
N
N
!'N €6 6 O o o\
\o
E
o
<tr@\€
N€vl
rn
c.l o O oqo 6\
o r
@
r6\f,
@I
N
€
N
O o r
N
6N<f
r€
€N
N r
r
rNN
€
r-
6Nr
\t
6
N
N
6r
@
N
N
N Oq
O N
Nr
6c.t
€€-
\f
N9d
$t:
N
N$@
a O 6\qO
dOat
r
o"
@rN
NN rN
N
EN
O 6\
o
oo!
D
oF
o
o
o&
E,o
ao
o
o!
z
-to
ooI
D
bI
o
o
o&oddo
I
o
&
r
o
o
oa:
F
o
o
o
r
o
a
o(
o
bt
o9o
oId
D
d
F
o
a
o&
F
oo
o
o.otr
z
E
o
b{
o(
I
o
o&obx
EI
o
Io&
r
o3
o
oa
F
o
o
o
r
o)
o
achd
cq
b(
o
o
o
ooId
D
oF
o
o
od
=
F
oo
o,o
z
o
odd
OI
o
o
o&od!!o
o
o
o&
r
o3
o
aa
6oF
a
o
o
r
sad
o
bI
ooo
l._
EE*19 c<oLl ooUOrEE or
ETElc
f,r
b\
F;l-u
ts
l<
lalo2
lorlal(E[J
\Ic\
q)
boBq
c;z
=eExIU
a
bo
aa
.o
6ri
\
E
a--
\
a-;
Ic\
q)
*a
-!F\
@
j
F
.]
Fzf:)
U)L)
:
6r
€r
N
€
o
N
I
r
s@ @r rd)66$,
a
dN
o\\
r,
rdO
H(,
Fl
&
zri
tN
rNN
NN
N
€
N
6I
@ N
aN\
cl
609
N
6\
ariU
f.la,l
Hzri
ri
Fl
N
$
€€€
.+
r€nN
NN €
@
@
<f
r€\t6
I
-i6
I
o\09
r @€
d
@
€r
r
€r6
€o @r\f,,
O.
.1I
6\NNo"ron
Nto:
@o
+
Oq
N
N
N
N
6 @?6
N
r
€
o\
€
r@\
6
a66+
N
N 6@r6 r
N€N
@
9r
r
iz
a
q
$€6 N
6
$r a o
€r
oq o\
s
r
o.-+
a€N
N N
*r 6 €
6 o\n N
O N@N
N
s
€aN
aor:
N
t
€6\
$
F
.!
6€
rrN
€
6N
r6
@
€\t
rq
O 09
N
a€
€
N
r
$r
O
N
@€€
N
Nrnr
N
\f,
qr
N 6€€rv')
a
E @
d
N
6\N
r
oa
aCI
N
\o
N
r
r
ON aN
dtI
$nO
6\
6i r
@€N
s6@.
N
N *+
v')
rq
c.t
o\@oi
Nr$.
@rN
@r
\f,N E
N$<f.r
r€v]
$N
o\
r
U
6$
t6
@€
N
a@
r
N @
6I
€r)6\
N
*@dlN$c)
@N@
$Ic
N
oN
e EI r
N@N
Nr
6
6\\N
6
r
N
N N @@NO
$NNq E6
N
o\..1
r
-
€N
N
N
st
€N
o@o"r
$oN \€
Nn o\
ci
@@
@s
6@
*@
N
r6 N..1 aO
6\rN q
N
6
@
N
6
r
€
r
€
N
N
r 6
N
\t
N6
r
r
N rN 6o€O
69 6\
N
@6
N
€
d€
N6
N
r 6N..1
€n 6\qN €
N
r
o@
d s
@ €
r
r
N
@
a*N
o\
r
a.
N
@
r
i
@Nr
€
r N
NNN
Nn
6N
N
r
N
r@
N
I rO 6
r
n
N
o\+r
@
r+r
N
N rN 6
aa
N
r ri@N
o\c1
r
a r€
O
€r r 6
N
6\o.l
O
I
<f,
@
r @@.
N
Na
66
6\
-;
+rr
@
NN N
:t o"6
Iv1€
Nd)
@
N
6\
r
z
a
r
r
rEa6
6
€
r or
O
a
di
6\r+€6@N\t
+
@
N
rNc'l
N
N@ r@ o\q rt@
r
drdd
NN NOnI
aN
o
rOI
Ij
r
.+
o\
r
tr
r
N€
NN
r
N
N
N€
I
$€N €
@
rc..l o\c'.1
@N
rcl
@
O6
O@@"
N N
Nr o
Nr
r 6\N
rr
O
O
N
Nd €
N
@
€s $$n@
q
@ n
r
-
6@6r
@N
NN$"€o€"$N
N\t
o@ @ Ns
NN@
N 6\r
C\sqo
rrn$
€
ci
d I
N
N
OoN
@..)
r
\o
o'l
6N6 €
NN
NN <fNN
€€ct
O
r
cr.)
@
€
6
o\r
o
o
6
F
o
o
o&
d:F
o!
E
z
o
obId
bI
o
o
o
oOI
o
o
o
od
r
o
o
oU)
d
F
o
o&
r
o
o
ood
d
o
tro
dd
D
oF
=
&
F
oo
o
o!
z
o
obx6
D
bx
5
ugo
o
o
r
a
F
o
o
r
o
oail
o
d
o
oo
o
obI
D
d:F
o
o
od
oF
oo2
o.o
z
Eo
au(
$
o
Iodou6o
o
o&
r
o
o
a
F
o
o
o&
r
o
oa
oI
a
bI
o
o
o0.
o
The months of November and December can also show significant differences in Schedule 175
revenue percentage from preceding months. This is due to the November I effective date of new
Schedule 175 rate adjustments. For example, the Schedule 175 percent for the General Services
class went from3.0o/o in October 2017 to 6.1% in December 2017 as the new Schedule 175 rate
effective November 1,2017 became fully reflected in customer bills.3o In order to visualize and
contrast the impacts on customer natural gas revenues between customer classes, the percentage
of monthly natural gas revenues attributed to Schedule 175 from the time rates were first
impacted by the decoupling mechanism through December 2017 is shown in Figure 2-9.
a,096
7ffi
5_0$
5.016
4_096
1.096
2'016
1.096
0.096
.a
SE N OVt6 DEC. JAN-17TE8.!7 UARt6 t7 APRl7 trA?" Jufl-t7JuL,r7 AUG. SEF-17 oCTl7 lt t7 r,lov
17 DEClll6
Figure 2-9. Schedule 175 as a Percent of Monthly Custonter Class Revenues
Figure 2-9 shows monthly Schedule 175 revenue as a percentage of total revenue for each
customer class subject to decoupling. The impact on revenue of the first decoupling tariff
adjustment effective November 1,2016 can be seen beginning with November 2016 billed
revenue. Residential customers saw the partial impact of Schedule 175 in November 2016 and
the full impact in December 2016 with Schedule 175 revenue accounting for 3.3o/o of revenue.
Schedule 175 revenue as a percent of class revenue increased again with new rates effective
November 1,2017 . In incremental percentage terms, the 2017 increase was nearly the same as
the 2016 increase in Schedule 175 revenue. In December 2017 when the full impact of the
second year decoupling adjustment is reflected in rates, Residential Schedule 175 revenues were
6.50/o of total revenue, 3.2 percentage points higher than the first rate adjustment revenue impact
in December 2016. Schedule 175 was limited by the 3Yo annual cap in both the first rate year
and the second, resulting in the similar incremental increase in Schedule 175 revenue percentage
in both years.
For Group 2 (non-residential) customer classes, Schedule 175 was also about 3Yo of totalrevenue
with the first rate year adjustment (effective November 1,2016). On a monthly basis in2017,
the impact of Schedule 175 as percentage of total revenue averaged 3.60/o for General Services
with lower amounts in the summer months and a similar incremental increase in the second rate
30 Although the effective date of revised Schedule I 75 rates was November I , customer bills in November reflect
usage that is partially billed at the old Schedule 175 rate and part billed at the new Schedule 175 rate. The portion
billed under the old and new rates is determined by a simple prorating of usage based on the number of calendar
days in the billing period before November I and the number of days on or after November I .
Exhibit No. 1
Page 2-21 Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page'lO7 ol 224
6
year. The 3o/o rate cap limited the Non-Residential Schedule 175 rate the first rate year (effective
November 1,2016) but was not a factor in the second rate year (effective November I,2017).
The spike shown in Figure 2-9 for Large General Service customers in December 2016 and
December 2017 is due to retroactive bill adjustments that lowered the total revenue for these
customers, resulting in a jump in Schedule 175 revenue as a percent of total revenue. The small
number (around two dozen) of large customers in this group can lead to large changes overtime
when compared to other customer classes.
Task 2Part2z Are Allowed Revenues Recovering Cost of Service by Rate
Group?
The objective for the second part of Task 2 as stated in the request for proposal, is shown below:
"This assessment must include an analysis detailing if allowed revenuesfrom
the residential, non-residential, and customers not subject to decoupling rate
classes are recovering their respective costs ofservice."
Relating to this objective is the following evaluation question, also taken from the RFP:
"Are the allowed revenues from the residential class, non-residential class,
and customers not subject to decoupling recovering their cost of service? "
For this analysis it is necessary to show annual calendar revenues and cost ofservice for each of
the three rate groups; residential, non-residential and non-decoupled. Revenue details are fairly
straightforward and are determined by base rate revenue and revenues deferred through the
decoupling mechanism. Avista provided detailed electric and natural gas cost of service
workbooks showing revenue and cost calculations for the three rate groups.3l Acfual cost
allocations are based on allocation factors in cost ofservice studies provided in the general rate
case (GRC) proceedings adjusted for actual usage and customer counts in each calendar year.
GRC values for rates and cost of service changed between the first decoupled year (201 5) and
the last two years (2016 and20l7).32 This shift in assumptions may result in strange
relationships in the analysis of actual revenue and cost of service.
Results of this analysis are shown in Table 2-16 for electric and Table 2-17 for natural gas. Both
tables are structured the same and begin with lines showing base rate revenue (line 1) and
revenue from decoupling defenals (line 2) over the calendar year. Total revenue is the sum of
each of these revenue types. Cost of service is broken down by production and transmission
(electric)/underground-storage (natural gas), distribution and customer services, and
administrative and general expenses. Production and transmission/underground-storage
expenses are further broken out between fixed and variable costs. Variable production and
transmission costs for electric (Table 2-16,line 4) are defined as volumetric sales to each rate
group multiplied by the retail revenue credit (cost per kVfh). Variable production and
3l See Avista response to Data Request number 89.
32 See Table l-l for the electric and natural gas GRC in effect for a given year.
Exhibit No. 1
Page 2-22 Case Nos. AVU-E-19-0_ and AVU-G-I9-o_
P. Ehrbar, Avista
Page 108 ot 224
6
underground-storage costs for natural gas (Table 2-l7,line 4) are defined by the applicable
Weighted Average Cost of Gas (WACOG) rates from Schedule 150 multiplied by therm sales.
Net operating income is shown on line 11 and is derived by subtracting operating expenses (line
8) and income taxes (line 10) from total revenue (line 3). The earnings test rate of return (line
13) is calculated by dividing net operating income (line l l) by the rate base (line l2). The return
ratio (line 14) shows the rate of return for the rate group relative to the overall rate of return for
the calendar year. For comparison purposes, line 15 shows the return ratio from the applicable
GRC settlement.
The allowed return on rate base is shown as an expense on line 16 and is calculated at unity (i.e.
the allowed rate of return is achieved for each customer class). Other expenses related to
allowed refurn on rate base, taxes and revenue related expenses, are also included in line 16.
Total allowed cost at unity (line 17) is the sum of all expenses (lines 8, 10 and 16). The revenue
over (excess) or under (shortfall) allowed costs is shown on line l8 and is calculated by
subtracting total costs (line 17) from total revenue (line 3).
Various revenue-to-cost ratios are shown at the bottom of Table 2-16 and Table 2-17. Line 19
shows the actual revenue-to-cost ratio for each rate group and calendar year and is calculated by
dividing total revenue (line 3) by total cost (line 17). The corresponding relative revenue-to-cost
parity ratio (line 20) shows the revenue-to-cost ratio for the rate group relative to the overall
revenue-to-cost ratio for the calendar year. For comparison purposes, line 15 shows the allowed
revenue-to-cost ratio from the applicable GRC.
Readers can more easily understand the findings of this section by focusing attention on two
areas of results in Table 2-16 and Table 2-17. First, determine if revenues exceeded all costs
and, next, determine if the result was as planned given the structure of rates and costs in the
applicable GRC. First, we are able to quickly determine if revenues for the system and each rate
group were sufficient to cover all costs by looking at excess revenue (line 18). If excess revenue
is positive then revenue exceeded all costs, including the allowed ROR on rate base. If excess
revenues are negative then costs exceeded revenue. The revenue to cost ratio (line 19) shows the
same relationship and can also be used to determine if revenues exceeded costs (line 19 is greater
than 1.00) or fell short of costs (line 19 is less than I .00).
The other area of results we draw the reader's attention to provides understanding of whether or
not the observed excess or shortfall in revenue was expected (i.e. planned) given the rates and
costs in the applicable GRC. This can quickly be determined by comparing actual results of the
revenue to cost ratio (line 19) to the GRC allowed revenue to cost ratio (line 21). If line 19 is
equal to line 2l then actual results were as planned by the GRC. When line 19 exceeds line 2l
results were better than planned and, conversely, when line 19 is less than line 2l actual results
were worse than planned.
We begin our analysis and reporting for this task with electric rate groups followed by natural
gas rate groups. Within the elechic and natural gas sub-sections below, we organize our
discussion by rate group across the three years rather than by year across rate groups to highlight
any trends within rate groups.
Exhibit No. 1
Poge 2-23 P. Ehrbar, Avista
Page 109 of 224
6
Electric
An examination of the electric revenues and cost of service analysis summarized in Table 2-16
reveals that Avista's Washington electric system revenue exceeded total costs in all three years.
As reported elsewhere in this report, these excess earnings are shared with decoupled customer
groups. Overall the non-residential rate group subsidizes the residential rate group and, to a
much lesser extent, the non-decoupled rate group. These cross-subsidization results are
consistent with GRC expectations.
Electric residential customers have a revenue shortfall in each year and that shortfall (subsidy)
has increased since 2015. The subsidy to residential is an artifact of the GRC as is the increasing
level of subsidy. The GRC allowed revenue to cost ratio for electric residential was 0.89 in 2015
and 0.87 in20l6 and20l7. Although the actual revenue to cost ratio slightly exceeded these
values, the subsidy to residential customers was mostly as planned.
The electric non-residential rate group experienced increasingly higher levels ofexcess revenue
over the 2015 to 2017 period. Comparing the actual revenue to cost ratio with the GRC allowed
revenue to cost ratio shows that the excess revenue was expected at nearly the same levels as
experienced. The non-residential rate group has slightly exceeded GRC expectations in 2016
and20l7.
The electric non-decoupled rate group has received a slight subsidy (revenue shortfall). The
subsidy has decreased between 2015 and 2017. The subsidy and decline in subsidy were as
planned by the GRC with GRC allowed revenue to cost ratios moving from 0.96 in 2015 to 0.99
in20l6 and20l7.
Natural Gas
An examination of nafural gas revenues and cost of service analysis summarized inTable2-17
reveals that Avista's Washington natural gas system had a revenue shortfall in 2015 and a
surplus in20l6 and20l7. Unlike the electric system, excess revenue surpluses and shortfalls
have not been consistent across the three years or within rate groups. The change in GRC
assumptions between 2015 and 201612017 appears to have materially shifted actual and planned
earnings results for all rate groups. The difference between actual and planned performance
across each year and rate group has also been material. However, on a relative basis as measured
by the relative revenue to cost parity ratio (line 20) the performance between rate groups has
been as planned (comparing lines 20 and 21) except for the non-decoupled rate group.
After receiving a larger than plarured subsidy (revenue shortfall) in 2015, the natural gas
residential rate group experienced a small level of excess revenue in2016 and an even smaller
(in absolute value terms) level of revenue shortfall in2017. Combined excess revenue for 2016
and20l7 is only slightly greater than zero meaning that revenue from residential customers are
just covering all costs. This is slightly better than the expected subsidy to residential customers
based on the GRC allowed revenue to cost ratio of 0.97.
The non-residential natural gas rate group essentially broke even in 2015 with a small level of
excess revenue (revenue to cost ratio equal to 1.00). Excess revenue increased in20l6 and2017
Exhibit No. 1
Page 2-24 P. Ehrbar, Avista
Page 110 ol 224
6
to over 5 million dollars that when considered with allowed costs results in a revenue to cost
ratio of l.l7 for 2016 and l.16 2017 . The sharp increase in revenue to cost ratio from 1.00 in
2015 was largely although not totally planned. The GRC allowed revenue to cost ratio went
from 1.04 in2015tol.l2for2016and2077. Actualperformanceof 1.17and 1.16 in2016and
201 7, respectively, outpaced planned performance of I . 12 for these years.
Excess revenue in the non-decoupled natural gas rate group experienced a shortfall in 2015 and
2016 but was slightly positive in20l7. The 2015 shortfall corresponded to a revenue to cost
ratio of 0.89 and was largely unplanned. The GRC allowed revenue to cost ratio for 2015 of 0.99
was much higher than the actual value of 0.89. Actual performance, as measured by the revenue
to cost ratio, in 2016 and20l7 steadlly improved from 2015 levels. This improvement was
largely unplanned considering the GRC approved revenue to cost ratio in effect for 2016 and
2017 was 0.91 and actual results were 0.94 in20l6 and 1.02 in20l7.
Exhibit No. 1w-G-19-0_
P. Ehrbar, Avista
Page 111 ol 224
Page 2-25
llo<'PERI
BAbO cnrLl orUO-E-E or
ETE(E
fr
brI5Iao2.(,oGp
\oc\
I\q)
bo
tr
ciz
€-cxllJr
N
o
ZZo
Nr€
r
dr€
ts
€@6@€r$ars@6 F
Nia
€
6+
@
o $
€N
Nr
rr\o9@9:66
-OOr
€
6 €
qa6qqqOOO
6
=9>.=o
€ro@
N
€
g\
N
qdlqc\*rd$
eONrN
€r
Nr€
ah€
€
a
a
€a
N
6\o-o€-E -i -:
6rr$o
aor
o EtN"id-
r6aa
d
6$rNNffho- \ 1c'l6€6€
r)Nnr
r\\f,\f,
6r]€
r
N
I
r
6\o$NNo:aaqXoo
6$r
6
r€o
r€€oq
O
F
SS
-:6@v
hrh
I
n
O6-<tO€T€10q o- --:*NAN6O€T
6
dr)a 6
r d
Nd
6@d6r
F6aN
€
N
o
ZZon
N
r
q
6l
r
6-$orr€t€^ c.I .1 €-rr€a
6\o
h
N
\f,€
co
it)
6r..)
Nr
6
I
t
€\od@&n33
6v
r
€raqqq
6
=oJ:Ea'd;
&
N€$€
-6
N N
i6€€ioo€
!nn6d
*6q
N@
dEda
€
@
€@$
N
r\o€Orq-shc]^8 -j,j
hrrNo
€rr
N€€N60oiN
ar)
N
rh@a
€€r@
h
a\o
@
o
I
$
trh
N
N
N€
N\ofNaq'99;ooo
=.;r
$€N
6@roq oq oeooo
F
s€
6N
+
d66=
9@@-++Nhi-Q@
@NTN@o-F
rF
n
N
6
r
r<f,
h
=6
€
\f,
N
@\oOONo'oots- , .
Sra
$a
n
h
N
o 6s
r
a
ts
vf .\ .1 €"d€€6N-
Far€in
I
hF o6€r
$$r
+\o+6eil09?';NOOr6
66
€9€qqq
z.;o
r
ot
N
€
N
rr
N
€*+-hO6-€^ $^ @^ r:
h€6
ra
NNIr
ri
ita)
*Nn
@
6o\
N
6\oQ€
v}; -i --'c\l roe
hh
€d
-@
r
*
N
d
dr
N
hr@6*-h6r- dI A^ O-€*ar
a
6$F
Nr
@
€c}d$
daho
rEOtN
r
6\o6r a\a
=
d
66@
F
6r\f,
rr
r6t
Nr=N
m
@96$h-S+1o-@-q69f--6@f
6=€€
rr09oo =
aa
r
rEr
6ar
o
o
o
o
fEc.o9566F4u
oO
daoo
o
o
F
o
.-o9, 7ntr ul;B. nBx xorl.] .ix- arDo O-'a 7)Eo!o'= otrE EO6 l.^E E=r odE d8.:
Eo 6rq
5EE =.=: H 9€.E
O'L AE& Ei
o
o
r{
F!
a
dF.
Eo3Eco;ooX!El€xQ6
ootrtrooootrc
o
OI
Lo
z
ooxrI]
dod
Eoke
gD
oo- (_)tr-lao>&6E=9<
OQ
ao
o'E
36i- *E uZ'^ cd* Picra.=freX Eozoc< q.l
o
o
I&
L
O
,!1
ri
o
6
oUo
I
o
a&
o,o
U
o
o
6
d
o(J
o
o
o
o
o
o
o
o(J
o
o
o
o&
-N +6€r @ rI @ 6O-
-o.lN
4L
S\
\j
q
o
t)
q)
a<
q)U
Lq
qoU
U
Vaq)
q)q
(.)\
UU
tA
\cj
I\
q)
.aRF\
@
OrtENirf:f;rrr H,I(L
b.-\I\
\)bnR-a-
,l
,)
Il
c,Eo
,l
,)
!)(
U;oz
c)o(E)
oz
=€gxtu
sr
ox
o'*S
a
r.U
q)
Sq
q.)
U
r.q)q
o'
4oU
s
q
\,)
q)
q
!
P
sa
F.-
Ia\
\)
*a
N
@
F
N
ZZ
6 6
*\ON€Ntor F=
d
s
$
6 N@.€-\o_N60'66qGdo
f;
Fr N6-qqq
6
>.=o
€O6$N@
@
I
6c)
r6Nr
hr666 hN
hh6d
@d
o6
\oNr
NrN@6o'@9clG-j,j3:
totn
9Nd
e -d.i5
NC
a
h
N
-€rts6-6$€- \.1 q€--r
d€€ra
€€
€No\od
qh€F a
N
r6r
6\orr€^Iod$rd
orrqqq
F
PAt6vi '-i
@-
arh
=€
@6-6
\ON€O6 6N
Ih6d
*r
o696
Fn
EN
rr€
N
+\oOOI?-ice9*;dh
600
\o
d
I
ZZIH
Or r
=o
€oo@ h€d
N
NO
_N \o€+6O
+\ooNN1-r€€-:i o o9(j
Nd
Sotoo6
6
>.=o
O
N
$r
€
or6*hr-NsrhNr 6N
tI
\oN
€@a
inr €&
O
II6il -e:s:
€
h
rNN
€
o
o
r
r
o N
€or€ohNoqr:€-€.O-6r$ N-
a
a€
6€
-No€N
a\o
r
€
€\o@rc'lffoo
drrN
a€-rr
F
I€€
d
=rr
h
€ro@6Fr@
@N€O
d
r 6
*
*Nn
d
a
aI
@
r\oOO60loovl; .j _.i'3d,N
nm6n
+oo
n
N
o
ZZo
oIo-
a€
(.)
O:l€oN:f,Oh<f *:a=N
EO €
t @91R\q-.@OO6+
rc€
606ogqq
z.=o
rr+6Nr
66*r
-rO@h€€9
N
eo$@
r66
trr€€otor
N\o6h9S":cl
Sr
o\f,<'
r
=6
h-a€
r-6N+ N-
d6€
a
*rNrlj
I!
d
N€rNNNro
$\oES$o'66o:e o o
=-;d
6
6q6a
Iq
oF
r@
rOir
n€r
6=
66N@qc\o-o!6dhr€ 6-
mt€td
€
d}r
N
N
6r€
€€*N60
Nh
-\oOOTfrEq
s-dc.t
(o€
n
rOOqqq
O--
o
o
o
o9)
E,O9I
&aoE
60oo
o
F
o
.^od 7ntr r!;I -cx xot! .;xc lsl6 0-'= aEaLO'= OtrE tro2 60d ot)dF c)dE €.9
cu 6sd
'E -E r 5.:o.F oo c
E > ''i.E €d o<
o
o
rd
F
a
xdF
EO3Es8oX!qcYxo6
qo
o9
o
al
6Lo
z
o
oo.xrl.l
o
d
d
I
o.o9>
9D
caoo-(J
tvo-il6tq<
oa
o
o.9gt
g?ia=
"trdE.* I b c-.r:..= 6 4E EcZ(,d@
o
o&
e
ch
o
t{
d&
'Eo
;9oo€u
='doo
a3d.
.Y o!v>Q
e &.29 or: ><
in&&eo
&-N $h9r 6 a!r €6O-
-Nd
Summary - Task 2
Impacts of decoupling on customer bills have been small over the first three calendar years of
operation, partly due to the timing of billing impacts. The last year of the period, 2017, was the
only year with the decoupling rate in effect for all 12 months. The impact of the decoupling rate
on electric bills ranged from a reduction of 1.7 %o for the pumping customer class to an increase
of 3.lYo for the residential customer class. Monthly impacts in November and December of
2017 reflect the latest change to decoupling rates and show increases in the residential rate group
to 4.60/o of customer bills and around 0.3oh for the non-residential rate group.
The annual impact on natural gas customer bills followed a slightly higher path than electric due
to greater exposure to the impacts ofheating degree days on natural gas usage and deferral
balances. Still, the impact on annual natural gas bills was small and nearly the same for all
customer classes, around one half of one percent in20l6 and around 3.7% in20l7. The pattern
of monthly impacts shows that the greatest impact on customer bills occurred at the end of 2017
when new decoupling rates took effect November 1,2017. With the new decoupling rates, we
expect calendar year 2018 natural gas bill impacts to be around 60/o for both natural gas rate
groups, residential and non-residential.
An important characteristic of the Avista decoupling mechanism is that the possibility of ever-
increasing levels of carryover deferrals (snow-balling deferral balances) is greatly reduced by
allowing the decoupling rate to adjust incrementally higher each rate year, subject to the annual
3Yo cap. This feature limits rate shock while also allowing the decoupling rate to amortize higher
levels of requested recovery. At some point, even if weather or other conditions that caused
initially high deferral carryovers persist, the decoupling rate will eventually adjust to a level that
recovers 100 percent of requested recovery and carryover deferral balances will fall to zero.
An assessment to determine if allowed revenues from the residential, non-residential, and
customers not subject to decoupling rate classes are recovering their respective costs of service
shows significantly different results for electric and natural gas. Avista's Washington electric
system revenue exceeded total costs in all three years. Overall the non-residential rate group
subsidizes the residential rate group and, to a much lesser extent, the non-decoupled rate group.
These cross-subsidization results are consistent with GRC expectations. Avista's Washington
natural gas system had a revenue shortfall in20l5 and a surplus in 2016 and20l7. Unlike the
electric system, revenue surpluses and shortfalls have not been consistent across the three years
or within rate groups. The change in natural gas GRC assumptions between 2015 and 201612017
appears to have materially shifted actual and planned earnings results for all rate groups.
Exhibit No. 1
0
Page 2-28
Page 114 ol 224
6
This section provides an evaluation of trends in Low-Income Bill Assistance and the Low-
Income Weatherization services during the study period (2012-2014 and2015-2017). The
billing analysis compares data for the three-year period immediately preceding decoupling to the
three-year period following decoupling implementation to identiff any changes. Other analysis
covers time since the inception of the decoupling mechanism.
Task 3: An assessment ofthe impact of the Mechanisms specifically on Avista's low-
income customers. The known low-income population to Avista are those customers
who have received bill payment assistance through Avista's Low-Income Rate
Assistance Program ("LIRAP"), energy efficiency services funded by Avista's
electric and/or natural gas energy efficiency programs, or the Federal LIHEAP
program. Cognizant that a larger portion of the low-income population do not
participate in the three programs referenced above, the Consultant is encouraged to
use other available information, such as the information provided in Attachments G
and H to this RFP, to better determine the impact on all Avista's low-income
customers. The assessment should include: (3a-3e)
(3a) A summary of the annual deferrals and rate impacts of the Decoupling tariff
tracker adjustments (cents per kWh, cents per therm, total dollars, and percent of
monthly bills) on the group of customers receiving bill payment assistance through
the above-referenced low-income programs.
(3b) A summary of annual low-income conservation program savings, expenditures
and customers served compared with the rest of the residential class, where low-
income conservation programs are defined as the programs currently being run under
Electric Schedule 90 and Natural Gas Schedule 190.
(3c) A description of any modifications to conservation programs targeted to low-
income customers since the inception of the Mechanisms including changes to
funding levels as well as changes to specific measures.
(3d) A comparison of the effect of the Decoupling tariff tracker adjustment on the
average customer receiving bill payment assistance through the above-referenced
low-income programs relative to the impact on Avista's average residential customer.
(3e) To the extent data is available, Consultant should evaluate other factors such as
household size, housing stock (e.g. mobile home, multifamily) and heat source (e.g.,
electric space heat) and the effect of seasonality when comparing the impact of
decoupling on low-income customers versus other customer groups (such as average
residential customers).
Figure 3-1. The Parts of Task 3
Exhibit No. 1
P. Ehrbar, Avista
Page 115 ol 224
s and ContrastsSection 3. Low-Income Anal
Page 3-l
Low-Income Billing Impacts (includes Parts A and D)
In this section we examine the billing impacts of the decoupling tracker adjustment for low-
income customers. We also contrast those impacts with the residential customer class. To
facilitate communication, we report here on both Part A and Part D of Task 3.
The objective of Task 3 Part A, as stated in the Request for Proposal (RFP), is shown below:
"A summary of the annual deferrals and rate impacts of the Decoupling tariff
tracker adjustments (cents per kWh, cents per therm, total dollars, and percent of
monthly bills) on the group of customers receiving bill payment assistance through
the above-referenced low-income programs "
The "above-referenced programs" are addressed at the outset of this section. The objective of
Task 3 Part D, as stated in the request for proposal, is shown below:
"A comparison of the ffict of the Decoupling tariff tracker adjustment on the
cuerage customer receiving bill payment assistance through the above-referenced
low-income programs relative to the impact on Avista's average residential
cr.tstomer. "
Relating to these objectives is the following evaluation question, also taken from the RFP:
"On average, were there any dffirences in the annual Decoupling deferuals and
tariff tracker adjustment impacts between low-income customers and residential
customers? "
A good place to start the discussion is with the question of how to define Avista's low-income
customers. Because this section relies on customer billing records, it is important to have a
definition of low-income that can be applied to the customer information system. Avista refers
to this group in the RFP for this evaluation as the "known low-income population and includes
customers who have received bill payment assistance through Avista's Low-Income Rate
Assistance Program ("L[RAP"), energy efficiency services funded by Avista's electric and/or
natural gas energy efficiency programs, or the Federal LIHEAP program"33. These are the
programs referred to in the "above-referenced programs" quote from the RFP above.
For the purposes of this section, we use the known low-income population for analysis and
comparison to the residential customer class. Avista pulled account-specific billing records for
low-income customers. Customer usage and revenue information was included for billing
periods for which the customer participated in one or more low-income programs. Annual
average low-income customer counts summarized from the account level data provided are
shown in Table 3-l below. Total residential customer counts as reported in Section 2 are also
shown in the table.
33 It is understood that the low-income population is much larger than the participants in the referenced programs
See Section 8, the low-income appendix for discussion and analysis of broader definitions of low-income.
Exhibit No. 1
o
Page 3-2 P. Ehrbar, Avista
Page 116 of 224
uase t\os. t\vu-tr- I Y-u_ ana t\VU-\r- I Y-u_
6
Table 3-1. All Residentiol and Low-lncome Electric and Natural Gas Custonter Counts
Electric Natural Gas
Year Residential Low-Income Percent Residential Low-Income Percent
2012 202,541 31,539 t6%146,776 14,44t t0%
2013 203,883 31,343 ts%147,880 14.34t t0%
2014 205.621 3t.szs t5%149,453 14,104 9%
2015 209,419 32.793 r6%t52,182 14,208 9%
2016 209,864 33,088 t6%153,955 14,449 9%
2017 2t2,49s 31,782 t5%1s7.563 14.189 9%
The number of low-income customers on the electric system has varied narrowly between 31 and
33 thousand customers.34 This amounts to 15 percent to 16 percent of the total residential
customer class. Avista's natural gas system has served over 14 thousand customers annually
since 2012, about 9 percent of the residential customer class.
Our reporting and analysis of deferral balances and decoupling tariff tracker adjustments
(decoupling rates) for low-income customers, including a comparison to the residential customer
class on average, is organized by electric and nafural gas service.
Impact on Electric Low-Income Customers
Customer usage is an important driver in most utility operations and financial results, including
decoupling deferral balances and decoupling rates. Figure 3-2 shows electric use per customer
for all residential and low-income customers.
Figure 3-2. Annual Electric Use per Customer, Low-lncome and All Residential
Electric usage per low-income customer is distinctly higher than for the average residential
customer. This difference appears to have narrowed over time, most likely due to conservation
programs for low-income customers, including conversions to natural gas heat. Low-income use
per customer averaged about 10 percent higher than average residential usage between 2015 and
2017. This means that low-income customers will have a l0 percent greater exposure (higher
rebates and surcharges) to the decoupling rate (Schedule 75) than the average residential
3a References to the Avista system refer to operations in the state of Washington, the scope of this evaluation.
Exhibit No. 1
P. Ehrbar, Avista
Page117 of224
16,000
14,000
t2,oo0
10,000
8,000
6,000
4,000
2,O00
Lo
Eo
f(J
o,o.
=
=iz
20t2 20L3 201,4 20t5 20t6*Low-lncome .-fAlResidential 20L7
Page 3-3
0
customer. Possible explanations for higher use per customer in low-income residences are
explored in Section 3 Part E, below.
Energy conservation programs are most likely the driver behind the narrowing gap between use
per low-income customers and all residential customers shown in Figure 3-2. Arelatively
greater level of conservation savings in the low-income customer group relative to all residential
would lead to the declining difference observed in the historical data. Considering just 2017,
first year conservation savings for low-income customers amounted to 1.7 percent of usage while
first year conservation savings for all residential was 1.3 percent.3s The low-income
conservation effort is also using conversions from electric space and water heating to natural gas
at higher levels than all residential. In20l7 low-income conversions accounted for 73 percent of
first year savings compared to 31 percent for all residential.
Average customer revenue and decoupling revenue (Schedule 75) is shown in Table 3-2 below.
Table 3-2. Contparison of Average Annual Electric Reventte per Customer
Residential
Group
2016 2017
Revenue Schedule 75
Revenue
Percent
of Bill Revenue Schedule 75
Revenue
Percent
of Bill
Low-lncome $ I,l16 $ 4.33 0.4%s r,268 $ 37.02 2.9%
All Residential $ 988 $ 3.91 0.4%$ l.ll6 $ 33.73 3.0%
Difference $ 127 $ 0.41 0.0%$ 152 $ 3.28 -0.1%
As explained in Section 2, defenal rates first became effective November 1,2016. Decoupling
impacts on revenues in20l6 are small because the first decoupling tariff hacker adjustment did
not become effective in rates until November 1, 2016. In 2017 Schedule 75 accounted for about
3 percent of the revenue from each residential group. On a percentage of bill basis, there is no
meaningful difference between low-income and all residential.36 However, low-income
customers paid just over $37 in Schedule 75 charges in2077, $3.28 more after rounding than all
residential. This is consistent with higher use per customer of low-income customers. Electric
low-income customers will also receive a larger rebate than all residential when Schedule 75 is
negative.
Monthly usage and revenue details for the two residential groups are shown in Table 3-3 for
2016 and2017. The data for all residential is the same as reported in Section 2,repeated here for
ease of comparison to low-income customers.
Schedule 75 revenue varies with the prevailing rate and the pattern of monthly usage. Average
monthly payments are shown in Figure 3-3 for both residential groups.
In2017 the average low-income customer paid a low of $2.01 in June to a high of $5.76 in
December with higher winter usage and the new higher Schedule 75 rate effective November 1,
3s Conservation program information referenced here is taken from Section 6 of this report where the impact of
conservation programs is discussed in greater detail.
36 Electric low-income customers show Schedule 75 revenue to be a slightly smaller percentage of the total bill.
Exhibit No. 1
Page 3-4 Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 118 ol 224
0
2017. The impact of higher use per customer on Schedule 75 revenue is also evident in the chart
with payments from low-income customers averaging $0.27 a month higher than all residential.
Figure 3-3. Contporison of Average Monthly Electric Schedule 75 Revenue per Customer
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 119 of 224
s7.0
s6.0
ss.0
S4.o
S3.o
s2.0
S1.o
So.o ll|toF.T.tsNF\NNT\,\NT\Ffffffffff.tf.t+fEEEEEoagE=H3o8EE
I All Residential tr Low lncome
Page 3-5
losO;Noi'; lT<b0 co-I GNJOr1- a:uJ REo-ro
Io
o,
uJ
l
U;oz
c)o(oo
\o
Is)
q)
bos
Q.,
c;z
--€txul
j
F
F
,l
Fzri
ori&rlil
o
o
I
N
r
N
€
oN
N
o
r
N
6 o
-]
Fzri
aIi
f-l
taoUz
F
Fl
N
6
N
6o
o-
N
@,r .1 ao rl
F
F
.l
Fzrd
(nri
,l,
€.:
r
9!
N
r
N
N
N
FFF
I
ts
c:q
rl
Fzrd
aId
rd2
Qz
I,
Fl
@
a+
<t
No"
@@cl
$
N€\NI
N
€N N
r
Nqr 6N
\o
I
c)rr
d
t-
o
N
NIv)
N
ts N
oi
I
d
oq
N
o€o9!
di
r N
6o rN
N
Nn \F-
I
o
Nr
I
I
N
a
Nr:
N
rr
N
6
N
oq
r€€.
@.+N6
N
6
+
N
66N N €*r
€\+
\o
z
6rr
I
r
F
F
N
N r r6N o
N
r-6N
r
dN
N
6-
N
o I \qo F-
I
z
N
6
NIqI
N
ao
E 3r:
I
r@
@
N
t
O a
o O r6
o
n o\n
\o
I N
ts
E
rI
N
r I o
aJ N
N
r
N
6r r t-
I
ct 6
ts
N
6
F r rF q
N r6
$N
Errrd}
N
rr ar
rr E ON
o\\d
I
o(n
I
d
r I
d
tsr o aa Io
tsN
Nr
N
ci
dN
.i
o o oe F-
I
q)o
r
r
6
N
N
r rrN N 09
d
6rc?€
rN
-€
€
N
$
o-
*r@ r
r N
o\\N
\o
Iol
ts
6
r
qN
a
N
6
d
rr o o oI
r
FN
IN
N
r o o F-
I
EJ
N
N
N
N
6
ro
6
N
N
No 6
I
ov,]
N 09
N
ONqI
d)
N
Nr$.N
@"
N
€N 6 rr
tN
6\roi
\o
I
66 N r r o oo o
doNI
IN N
o
o6r r o o
o r-
I
o
6$.
F
d
I
6
N
F d r
N oqN clo€q
N
6
N
@
r
@
ar aNc)€I
N
6\roi
\o
I
r
o6
rI
r
a
N
ts
ts
r o oq
d
NI
di
N
FoNI.1
N Ir oqo qo t-€
o
o
N
dr 6o 3.ro
Nq o9
N rq6
N
€
I€
N
@
N r *r €s
I
d t:N
2
N
6*-
I
q
or I o o r€
N
N
6Nts6ir.1
N
Ncl r r o qo t-
z
6
F
es
r
N
N
rr F NN
q
N ?N
N6
N
N
Or
N
N
$I6 a
r
rc.lN
6\r6i
\o
.t
II
ts
r oN
r r o r
6N
r 6
r oqo qo F-I
rrr 6r
N
N
s6 N 11
N oe
N
r
$
N
N
N
N
@N 66 o
@
Oroi
6\r
N
\o
2
r
a
rNa
6r
r6
N
I
6
oqo o N
o
o-o o oqo qo F-
L
a
d
N
rNN N
N
d
Nr N
6
Noq
N
6
\f,
r
N
6
rN
6
N N Nr N 6\\N
\o
I
ofr
N
N
I
d
al
qq 6 o
o
Nr
cl
N
6\6q
o
N rN
o
o o t-
I
lL
q
E.1
rN
rN
IN
6
N
N
o..1 N 6
NF
?at r
N
6\f,
o{<f€\f,
@\N
Io @$O
E
N
I \d
\o o
rN
-i
N
rN
o
I
N
r
N
oN
o N o oq sc N
d
ts
Nc.l
a
6o I o o qo t-
t
o
N
N
NN
N
N
r
o
r 6\oq
d
N
$r
rIclNNo9
@N r rr €6 r o\\N
>
d
p
F
)
il
oF
a
I
E
z
>
d
D
!x
9
&
U
a
d
r
EoF
tr
r
!
6
d
o
c
L
H
oF
&EoF
o
!
E
z
d
u
I
d
d
r
E
@
oF
c
rIa!
ra
d
o
a
>
D
aF
d
oF
o
p
E
z
2
d
p
d
I
&{
a
&
r
?aF
r
9r
d
a d
p
F
dEoF
o
3
Ez
t
d
d
&
ds
tr
&r
!
oF
a
&
a
q
o
a
t4\,)
\l
!
R
q)
oU
Il
f,l
qi
q)
q)q
s
6
q)
\t60
bb
U
Us)t.l
.o
*;
Ioa
q)
.a
N
@
6
Impact on Natural Gos Low-Income Customers
As with electric, due to the influence of use per customer on decoupling deferrals, we begin our
discussion of natural gas with a comparison between low-income and all residential use per
customer. Figure 3-4 shows nafural gas use per customer for all residential and low-income
customers.
9m
8@
b7m
Eg6m
.3 sm
Eo*
Esmo€zm
1m
2072 2013 2014 2015 2015
+Low-lncome .+-AllResidential
2077
Figure 3-4. Annual Natural Gas (Jse per Cuslomer, Low-lncorne and Average Residenlial
Natural gas use per low-income customer is clearly lower than the average residential customer.
This is the opposite of the electric system where low-income use per customer is higher than the
residential class. Natural gas low-income use per customer averaged about 10 percent lower
than average residential usage between 2015 and 2017. This means that low-income natural gas
customers will have a 10 percent lower exposure (lower rebates and surcharges) to the
decoupling rate (Schedule 175) than the average residential customer. Possible explanations for
lower use per customer in low-income residences are explored in Section 3 Part E.
Average customer revenue and decoupling revenue (Schedule 175) is shown in Table 3-4, below.
As explained in Section 2, deferral rates first became effective November 1,2016. Decoupling
impacts on revenues in 2016 are small because the first decoupling tariff tracker adjustment did
not become effective in rates until November 1, 2016. In 2017 Schedule 175 accounted for 3.4
percent of low-income revenue and 3 .7 percent of all residential revenue. On a percentage of bill
basis, there is only a minor difference between low-income and all residential.3T However, low-
income customers paid just over $25 in Schedule 175 charges in20l7, $3.55 less than all
residential. This is consistent with lower use per customer of low-income customers. Natural
gas low-income customers will also receive a lower rebate than all residential when Schedule
175 is negative.
37 Natural gas low-income customers show Schedule 175 revenue to be a slightly smaller percentage of the total bill.
Exhibit No. 1
case Nos. AVU-E-19-U_ and AVU-G-I9-U_
P. Ehrbar, Avista
Page121 ot224
Page 3-7
6
Table 3-4. Comparison of Average Annual Natural Gas Revenue per Customer
Customer Group
2016 2017
Revenue
Schedule 175
Revenue
Percent
of Bill Revenue
Schedule 175
Revenue
Percent
of Bill
Low-lncome $ 629 $ 3.39 0.s%$ 731 $ 25.01 3.4%
All Residential $ 669 $ 3.99 0.6%$ 781 $ 28.s6 3.7%
Difference $ (40)$ (0.60)-0.1%$ (50)$ (3.ss)-0.2%
Monthly natural gas usage and revenue details for the two residential groups are shown in Table
3-5 for 2016 and2017. The data for all residential is the same as reported in Section 2,repeated
here for ease of comparison to low-income customers.
Schedule 175 revenue varies with the prevailing rate and the pattern of monthly usage. Average
monthly payments are show in Figure 3-5 for both residential groups.
Figure 3-5. Contparison of Average Monthly Natural Gas Schedule 175 Revenue per Customer
In2017 the average low-income customer Schedule 175 payments ranged from a low of $0.42 in
August to a high of $5.09 in December with higher winter usage and the new higher Schedule
175 rute effective November 1,2017. The impact of lower use per customer on Schedule 175
revenue is also evident in the chart with payments from low-income customers averaging $0.30 a
month lower than all residential.
Exhibit No. 1
P. Ehrbar, Avista
Page 122 of 224
S7.o
S5.o
Ss.o
54.o
s3.O
s2.O
sr.o
s0.o O6tsNtsFFtsFFFFtsFHHHddHdHioi;iiii+i.Ali0€Easi€griisdeE
I All Residential I Low lncome
Page 3-8 uase Nos. Avu-tr- I Y-u_ ano AVU-(,- t Y-u_
Ito<
BENF<b[9 ccoLl (5NUnrD-E or
ETFlotr
b5
F;IU
ts
K
t,;loz
lorlal(ErJ
o\
Icn
q)
boBtr
c;z
=p
!xuJ
rl
F
F
j
Fzf-I
af-l
,].]
rN
6
r
9-
c!
6 a
6€I 6q Io
Fl
FztrI
atrl&ri
a.
Qz
I
B
rl
N
6 6
I
6N 6 vto -]
F
F
Fl
Fzti
ati&j
rl
r
Ir
N N
I
F r r
N
c:
Fl
Fzri
aEI
li2o()z
B
,l
r
o
N
6 F
F ts o
N a
\o
I
q)
NI
Ir
oo
a
N6r
a
F
6
r N r
qo
IN t-
I
o
I
N
Ir
F
r6
6 r
-i
.')I nI
r
r
Nq
NN
oN6-o oq
r
c!I
\o
Iiz
a
r
d)
N
N
6
I N I
Ir
rcl6r
d
F r roots
N
lr-
I
z
I6
o
No
F oI 6
@r
o r
6
a Na o
a
aN ..)
\o
rl()
6
No
N N N o q o
r
FN
rr N oqo qo t'-
I
6aorrN
oN
r6I
o rqo eN
Iots t oN N N
N
09 IN
\o
I
q)
ah
r
I
N
o
o)aN
F N o oqa e 6
Fo
N
do
N
0
NN
r Iol o
o q r-
I
q)a
I
N
d
I
N
r
r
N I
I
o N N
a
N
N N 6Nr)
€n q
\o
IE(
ao
N
r r oN o oeo ts
F
oaq r I o oqo seo r-
Iqt
o 6
N
6
r
N I
q
I
€cl
N
I
I raaN
ro
N N Nno r
\o
I
oIr:r
v]N
N
r
di
6 r NN o oqo ea o
N
N
I
N
r N o qo F N
N
N
o0N
or
N
I
d
o N $-I 6
N
NN .1
\o
I
r
N
FF
I
dd N IN o oIo eo
€r
rN
od .'I N rd o qo t-
I
6
NI
Nq
r
r
N rN 6
ao
Io nd
o I tsr N rN ao r')
N
\o
I
G
2
Io6IN
6N
N
6 N o oqo o eor
o-N o
o
t-
2
6
6l
r Nr
a F
CJ
N
N
n
o€
N
I
3c):
N d 6
1
o(.!o9
N
\o
I
6
q
6
rN
6 aI Io oI I €N N o oqo sqo F.
I
ao
r
o
r-I
r a 6
N cl N
r6
a 6 r66rN
ooq
di
\o
IL
a
r
6
d
oN
o
N
qr o oqo qo
Fr o
tsIo
rI r €oqo qo ts-
.t
z
I
ro
r
rI
6
6
6I
6 r6 6ro
6 .1 o
NoN 6N
d€6 ro
I
N N
I
0fr
o
nr\a
NI
r*"
N o o r6
6
..1
r o o oqo F-
I
ofa
sN
r
N
N
aNI€
0 N
II
r.l 11 I
o
6.
aN
6I
r
I
ool o ...1
\o
I
o
ts
6
N
N
6r
N
6N
N o oqo eo
r 6
o
NI
rag.o o oqo sqo r-
I
6
o)r6o"
N
NN
N
N o€oNn
a.1 o
c]
..]
N
o09r
6i
r N
a
d.!...!
D
oF ,o
o
3
Ez
tr
q
r
g)
oF
r
!au
=o
o
e
-o
tr
d
oF aF
o
p
Ez
D
c
&
dF
tr
F
!
a
oF
r =inu
o
P
L
E
E
D
oF oF
a
!
E
z
trE
D
tr
r
!
oF
d
F
2
a
d
d
o
9
E
s
g
oF
9
&
oF
e
o
!
Ez
u
d
&
d
a
r
!
a
oF
a
r
a
a
ca
d
o
g
eJ
qq)
\t
q)
oU
I>
U{
\t
q)
a)q
u
q)
qJ
4
\./
L
N
R\
o\
-iI
\)ra
F\
@
6
Summary - Task 3, Parts A und D
The decoupling deferral tracker adjustment, Schedule 75 for electric and Schedule 175 for
natural gas, has had a relatively small impact on low-income customer bills. In20l7, the first-
year decoupling rates were effective the full calendar year, the average low-income customer
paid $37 in Schedule 75 charges and $25 in Schedule 175 charges. These charges amounted to
2.9 percent of the average low-income electric bill and 3.4 percent of the average low-income
natural gas bill. Looking forward to 20 1 8, both Schedule 75 and Schedule 17 5 are expected to
be negative effective November 1, 2018, resulting in a rebate from decoupling through October
2019.
On a percentage of bill basis there is no meaningful difference in decoupling charges between
low-income and all residential customers. However, low-income use per customer averaged
about l0 percent higher than average residential usage on the electric system and 10 percent
lower on the natural gas system. This means that low-income electric customers have a 10
percent greater exposure (higher rebates and surcharges) to the decoupling rate than the average
residential customer and low-income natural gas customers have a 10 percent lower exposure.
Possible explanations for higher electric and lower natural gas use per customer in low-income
residences are explored in Section 3 Part E.
Low-Income Savings, Expenditures and Customers Served
Task 3, Part B is defined as follows:
"3b) A summary of annual low-income conservation program savings,
expendifures and customers served compared with the rest of the residential class,
where low-income conservation programs are defined as the programs currently
being run under Electric Schedule 90 and Natural Gas Schedule 190."
C ons ervotion Program S ovings
Residential and low-income electric energy savings are shown in Table 3-6 and these results are
partitioned into conservation (Table 3-7) and conversion of electric heat and hot water to natural
gas Table 3-8.38
Table 3-6. Total Electric Energy Savings - Conservation and Conversions (kllrh)
Sector 2014 2015 2016 2017
Residential 25.397.486 t6.082,204 43.063,551 33.376.237
Low-Income 400.247 829.091 546.066 710.204
Percent Low-Income 1.6%5.2%1.3%2.l%o
38 The source of information for the energy savings tables is the set of Washington DSM Annual Conservation
Report & Cost-Effectiveness Analysis for each year from 2014 throughZ}l7.
Exhibit No. 1
P. Ehrbar, Avista
Page 124 ot 224
Page 3-10 \,ase t\os. Av u-tr- I Y-u_ ano Avu-\r- I Y-u_
Table 3-7. I-937 Electric Conservation (kWh)
Table 3-8. Electric Conversion to Natural Gas Savings (kwh)
Sector 2014 2015 2016 2017
Residential 1.810.904 5,365,595 9.766.855 t0.237.036
Low-Income 201 ,855 619,s84 273,628 518,748
Percent Low-Income tt.t%11.5%2.8%5.1%
The percentage of electric energy savings due to conversions is shown in Table 3-9. In 2017 this
was about 3lo/o for residential and about 73Yo for low-income.
Table 3-9. Percentage Electric Savings Due to Conversionsfront Electric to Nalural Gas
Sector 2014 2015 2016 2017
Residential 7.1%33.4%22.7%30.7%
Low-Income s0.4%74.7%50.1%73.0%
Residential and low-income natural gas energy savings are shown in Table 3-10.
Table 3-10. Total NaturalGas Conservation Savings (therms)
Sector 2014 2015 2016 2017
Residential 355.443 343,395 367,891 773,030
Low-Income 14,944 13,154 18,490 3,034
Low-Income as a Percentage of Other Residential 4.2%3.8%5.0%0.4%
Before turning to expenditures and customers served, we first provide a discussion of the low-
income payment assistance and energy savings programs.
Avista service to low-income customers includes both bill assistance and low-income
weatherization programs. Bill assistance programs are analyzed first, followed by low-income
weatherization.
Exhibit No. 1
Sector 2014 2015 2016 2017
Residential 23.586.582 10.716.609 33.316.699 23.139,201
Low-lncome 198,392 209.567 272,438 191,4s7
Percent Low-Income 0.8%2.0%0.8o/o 0.8%
6
Page 3-l I Case Nos. AVU-E-19-0_ and AVU-G-I9-o_
P. Ehrbar, Avista
Page 125 ol 224
Low-Income Bill Assistance
To assess the impact of the decoupling mechanism on Avista's low-income customers we
evaluated the trends in bill assistance before and after decoupling implementation in January
2015. We analyzed each of the bill assistance programs that are available to assist Avista low-
income customers including bill assistance funded by outside organizations. The purpose of
these programs is to alleviate the home energy burden for low-income customers and to provide
emergency assistance as required, while keeping service connected.
o
Low-Income Rute Assistance Program (LIRAP)
LIRAP provides energy assistance grants to low-income customers in Washington, Idaho, and
Oregon. LIRAP grants are used to help with paying off a portion of a past due energy bill to ease
the energy burden on limited income customers below one-hundred and twenty-six percent
(126%) of the Federal Poverty Level (FPL). Benefits for limited income households are based on
eligibility and a percentage of the customers' utility bill.
LIRAP services are delivered by the Washington State Department of Commerce (DOC) in
collaboration with a network of Community Actions Agencies (CAA) throughout the Avista
service area in Washington State. The CAA's provide the client intake and eligibility
determination services required to distribute LIRAP benefits.
The program is funded by rate payers through the LIRAP Tariff Rider applied to energy usage on
both electric and natural gas customers. The LIRAP tariff rate for electric service is established
through the rate setting process and decided by the Washington State Utilities and Transportation
Commission. The level of LIRAP funding is determined by the Schedule 92 and Schedule 192
rate applied to the volumes of electric and natural gas sales, respectively. Table 3-11 presents
the electric service LIRAP tariff rates in each of the listed rate schedules used to determine the
available funding.3e
To provide a simple combined view of the overall trends in the LIRAP electric service tariff rate
since 2015 we calculated a weighted average $/kwh LIRAP rate. The calculation included all
schedules listed in Table 3-l I except Schedules 4l-48 since these schedules are not billed on a
$/kwh basis.
The weights are based on the projected dollar sales of electricity in each of the affected rate
schedules listed in Table 3-l 1.40 Figure 3-6 illustrates that the weighted average LIRAP rate
increased steadily since 2012 and has continued that trend after 2015 and through 2017. This
trend is projected to continue with a proposed increase planned for October 1, 2018.
3e DR Response: 073, Attach. A and DR Response: 073 Attach. A, Revised
a0 DR Response: 073, Attach. A and DR Response: 073 Attach. A, Revised
Exhibit No. 1
Page 3-12 P. Ehrbar, Avista
Page 126 ot 224
o<-ENirfF-oc!
€F<c)UJB
o- 0-
.O
I.A
q)
u0
a_
ciz
=€ExIIJ
Uh.u\
AJ
$
a,)
h.
\q)
h\
E-t
a_\n
o<
*l
a)Q
L
-\,)q
(.)
L
Q!Jr{
\c;
Ioa
\J\
b.
tL
\)
9a<t\
\-
-!F\
q.\
!q)
Q
q
U\
U
AJ
hl
*i
I
\).a
Ft
€
o
a
on
TO
oi oo+rco
:Uo
o
c
o
Xo
Eci
PoI
8qoEo
q,l rlls
ooeo
oE
o
CE
o
.o
rd
€
I
IoI
o
6l
o
e.l\o r-r e.l 6
q)
6t
o
c!
ooLq)
o\@a
F-
I
oI
+
o
o\6 o\
\o
o.cl
\o
I
I
F-+al \f\o rc ta c]
\o
I
6
c-l o\\o
o
oo
o\
ia
I
I
I
6 ol o\oo
-i
oor
o
o\
o\$
rr)
I
I
@ t-6 <.F-1A6
o\
t
I r-o +r \o+o $\o
o
a{tr
o\€6
?a
!
6t
I
6\o €atF-s al\o r-
o\66
N
I
6t
I
o\
o
t-s+
ci
t--\o
\o
rc
o
()
o
ol al al(\
a.l
C.l
c.l
6+
$
o
o
oo
Loa
o
oo
oo
oU)
!so
0)
obI
dJ
oo
oU)
(s
o
ooo6I
d-l
xrll
AI
oAIqE
o
o
bT b{
-]
o
"tsoo
U)
@
6
Table 3-12 presents the natural gas service LIRAP tariff rate which is applied to therms of
natural gas sales for each of the rate schedules listed to determine available funding. The LIRAP
natural gas rate is established through the rate setting process and decided by the Washington
State Utilities and Transportation Commission.
To provide a simple combined view of the overall trends in the LIRAP natural gas service tariff
rate since 2012,we calculated a weighted average LIRAP rate for all affected rate schedules. The
weights are based on the projected dollar sales of natural gas in each of the affected rate
schedules listed in Table 3-12.
Figure 3-7 shows that the weighted average natural gas LIRAP tariff rate increased steadily since
2012 tfuough2Dl7. The positive trend is expected to continue with a proposed increase in the
natural gas service LIRAP tariff rate planned for October 1, 2018.
Exhibit No. 1
P. Ehrbar, Avista
Page 128 ot 224
Page 3-14 uase t\os, i\vu-tr- [r-u_ an(J Avu-u- rY-u_
l(E$
tsENF;<E[, t'o,L]. ONUOF
Z,EotIrrt PEo-[6
Io
o)
IJJ
D
uiozoooo
!a
Ien
q)b.
tr
oz
=_-oEx
UJ
h^B-:..
a,)
\
\)
b.
\q')
Sh-\.
s.
F,\
a_\a<
{q)
U
\q)q
t
U
L.
tE
t--
Ioo
q)s.
b.q
@
o
N
o
e
c?ooo oo
6=
cO!E.:ro
o
o
oE
d
q
ci
aqoIo
u.uoqVS
o66<Ndqqcqoooo
q)
\
L-
t\
o_\
{
U
\q)q
q
r\
h\
N
I.a
qJ
'a
-9t\
@
o
o
o
^o
f-I
o
@
6
I
oI
++
OI
at
r
.+al @ tatao\
t-
g()oIo
o\\o
Ns <f
$q
€lr)
a
\o
I at€a1+F-?at'
\o
I c.l
t-
o
$
Oi
e'l
F-o\F-
\o
in
I
oI
6r-$
ol
al
o
FT6 \t€\oo\(.)
ro
I
I
+
6.t6 o\r-
o
F-ii(a
t
I
cl
+\oo.roa
N\t€tao\
(a)
I
I
ol
o
6\o6
+
00
N€
a.l
I
CB
+OiO
o
r
Oi:
F-
@
-{
+
6
rOt
o
o
oa
C.l el a.la.l
al
ol
\o+
oo
oa
do
o(,
oo
oa
o
o
o6n
6
-.1
oo
oa
o
(.)
ob0
drl
xrI.]
O
-o
Fo
d
o.
d
F
ou!
Lo
o
o0
q)
6
Rate Discount Pilot Programfor Seniors
Avista has an experimental pilot program that offers a rate discount to fixed-income seniors and
customers with disabilities whose household income is between one-hundred and twenty-six
percent (126%) and two-hundred percent (200%) of the FPL. This program began October 1,
2015 and will end September 30,2019, though it continues for those customers who are
currently enrolled. The rate discount is limited to 800 customers (700 in Spokane County and
100 in Stevens, Lincoln and Ferry counties. The pilot program was only available through SNAP
and Rural Resources for customers in Spokane, Stevens, Lincoln and Ferry counties. This
program is an irmovative approach for Avista and was implemented in the year that decoupling
started.al
Low-fncome Home Energt Assistance Program (LIHEAP)
LIHEAP is funded by the US Department of Health and Human Services (HHS). It operates in
every state and the District of Columbia, as well as on most tribal reservations and U.S.
territories. The purpose of LIHEAP is to assist low-income households, particularly those with
the lowest incomes who pay a high proportion of household income for home energy, primarily
in meeting their immediate home heating and cooling needs. The primary factor determining
eligibility is the household income level which must be at or below the LIHEAP State Poverty
Guideline (Table 3- 1 3).
Table 3-13. LIHEAP Poverty Guidelines (2017)
Number of Persons
in Household
State Poverty Guideline
for LIHEAP
Number of Persons
in Household
State Poverty Guideline
for LIHEAP
I $12,060 5 $28,780
2 $16,240 6 $32,960
J $20.420 7 $37, l 40
4 $24.600 8 $41,320
The LIHEAP statute defines home energy as a source of heating or cooling in residential
dwellings. The LIHEAP block grant serving Avista customers is administered by the
Washington State Department of Commerce (DOC) in collaboration with a network of CAAs'
across the state.
Because LIHEAP is a Block Grant progmm, states are authorized to add additional criteria to
determine the level of benefit provided to each eligible household such as hypothermia risk,
crisis interventions, and high energy burden.
Project Share
Project Sharea2 is a donation-based program that helps keep homes wann through crisis
situations like a sudden loss of income, expensive medical costs, malfunctioning heating
equipment and other unforeseen circumstances that deplete available funds and make it difficult
to pay household energy costs. The program is a partnership between utilities, fuel vendors and
al Comwell, John, Avista Low-income Rate Assistance Program Rate Discount Pilot Impact and Process Evaluation,
Primary Report Update. Evergreen Economics: July I l, 2017.
a2 Response to DR 045
Exhibit No. 1
Page 3-16 Case No6.AVU.E:ag.-0 anaIAVU-G-1 9-0
P. Ehrbar, Avista
Page 130 of 224
6
community action agencies that provide emergency energy assistance to qualified households
that have exhausted all other energy assistance resources.
The goal of Project Share is to help stabilize households-in-crisis for 30 days. People do not need
to meet federal poverty guidelines to qualiff, but they must contact their energy provider to make
payment arangements to avoid fufure emergencies.
o Project Share funds can help cover utility bills, deposits, deliverables - oil, wood, coal or
propane - and furnace repairs.o Project Share decisions are made on a case-by-case basis in accordance with the Project
Share Administration and Distribution of Funds Agreement.
Project Share currently receives donations froma3:o The Avista Corporationo Avista employeeso Avista customerso Ferry County PUD customerso Inland Power &Light Corporationo Inland Power & Light customers
o Modern Electric customers
o The Spokane AdFed Golf Tournament
Miscelloneous Bill Assistunce
The MISCaa Assistance Category consists of several dozen organizations that provide energy
assistance grants to Avista customers. These organizations include churches, social service and
government agencies: such as the Salvation Army, Catholic Charities, the Department of Health
and Human Services or the local Housing Authority. Energy Assistance is not the primary way
that these organizations help individuals (it is not their core mission or function); however,
during their service they may help individuals with their utility bill. Additionally, many of these
organizations do not have an established source for funding to help with energy assistance. In
receiving these assistance payments Avista customers are categorized as MISC within the
Customer Care and Billing System.
a3 Funds raised from Utilities other than Avista provide bill assistance benefits to customers of those utilities and not
to Avista customers.
aa Response to DR 046.
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 131 ol 224
Page 3-17
o
Bill Assistance Funding Trends
The bill assistance funding study period (2012-2017) provides three years of pre-decoupling data
and three years of post-decoupling data. Figure 3-8 illushates that between 2012 and2014,
combined bill assistance funding levels from all sources were stable ranging between $8.26 and
$8.83 million per year.as In 2015 a significant decrease in funding was reported, from $8.7 in
2014to $6.8 million in 2015.
s10.@0.000
se,5@,@o
99,0@,mo
s8,sm,mo
s8,0@,mo
s7,5m,m0
57,0m,mo
95,s@,mo
s6,0@,mo
5s,s@,mo
-TOTA!
1012
s8,833,026
)o71
s8,263,195
2014
s8,776,156
2015
s6,884,652
2015
s7,704,536
20t7
59,672,6v
Figure 3-8. Value of All Bill Assistance Grants
Figure 3-9 illustrates that funding levels for all of the four bill assistance programs decreased in
2015. The largest declines were the LIHEAP and MISC sources. Overall funding levels
recovered in both 2016 and2017; however, the recovery was not uniform for each funding
source.
s4,5@,Om -
s4,O@,mO
$,s@.mo
53,Om,mo
s2,sm,60
s2,o@,co()
sl,5m,m0
91,0m,@o
9sm,00
-UHEAP
-Proiect
share
-^ *" M isce llaneous
2072.
52,530,905
53,974,285
st,21t,oo7
51,OS,830
2013
s3,206,011
52,729,031
91,289,148
91,03s,0os
2014
93,290,187
53,068,301
s1,332,286
sr,o8s,3eo
2015
53,026,172
s2,737,6A
s1.001,978
9718,538
2016
s2,9S.873
52,487,023
s1,618,6s8
9@7,9A2
2017
s3,709,447
$2,08s,741
93,196,733
5680,7r3
Figure 3-9. Value of Bill Assistance by Funding Source
Since 2015 the LIHEAP funding trend has been level while LIHEAP and MISC funding
continued to slightly decline. However, LIRAP and Project Share both show significant increases
a5 DR Responses: 026 Attach. A,027 Attach. A, 028 Attach. A, 048 Attach. A, 048 Attach. B
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-o_
P. Ehrbar, Avista
Page 132 ol 224
Page 3-18
6
in funding levels, particularly Project Share which increased funding by $2,194,755 between
2015 and 2017. Project Share and LIRAP funding have made up for losses from LIHEAP and
MISC funding reductions over the study period. Figure 3-9 reflects a shift in bill assistance
funding with increased dependence on LIRAP and Project Share and reduced dependence on
LIHEAP and MISC. This reflects a shift toward increased local and utility funding. LIHEAP
funding like other federal block grants is subject to significant changes depending on
Congressional Appropriations.
Number of Bill Assistunce Grants
Figure 3-10 shows a significant decrease in the combined number of grants from all funding
sources provided in 2015, reflecting the decreased funding levels for each funding source in
20t5.46 This is followed by a recovery in the number of grants in2016 and2077.
25,000
24,0@
23,0@
22,O@
21,000
20 000
19,0m
18,000
17,000
15 000
15,000 2012 2016 2077
-TOTAL
23,232
2013
23,369
2014
23,247
2015
18,2t2 20,863 24355
Figtu'e 3- 10. Number of Bill Assistance Grants Provided
Figure 3-11 illustrates the trend in the number of bill assistance grants for each funding source.47
The data reflects a continuing downward trend in the number of MISC bill assistance grants and
a leveling-off of the number on LIHEAP bill assistance grants. Consistency with funding levels,
the numbers of LIRAP and Project Share grants has increased annually since 2015 and has made
up for the decreases in the LIHEAP and MISC bill assistance grants provided to Avista
customers.
a6 Responses to DRs: 021 Attach. A, 021 Attach.B,02l Attach. C,022 Attach. A,023 Attach. A,024 Attach. A,
047 Attach. A.
a7 Response to DR's: 021 Attach. A, 021 Attach. B, 021 Attach. C, 022 Attach. A,023 Attach. A, 024 Attach. A, 047
Attach. A,026 Attach. A,027 Attach. A,028 Attach. A, 048 Attach. A, 048 Attach. B.
Exhibit No. 1
Page 3-19 P. Ehrbar, Avista
Page 133 of 224
6
9p00
8,0@
7,0m
6,0m
5,0@
4,000
3,000
2p00
1,0m
2012
5,@l
4,275
5,245
2013
7303
5,136
sE27
4,103
70t4
7,159
6,314
5,745
4.069
2015
562r
4.545
3842
3,104
2015
6,718
5.669
6.1 34
2,342
2017
8,r30
4,993
8,904
232e
Figure 3-l l. Number of BillAssistance Grants by Funding Source
Average Bill Assistance Grant
Figure 3-12 presents the average grant levels of the bill assistance grants for each of the funding
sources.48 The average grant levels have remained relatively stable over the (2012-2017)
evaluation period with a modest increase from $378 in 2015 to $397 in2017.
aE Response to DR's: 021 Attach. A, 021 Attach. B, 021 Attach. C,022 Attach. A,023 Attach. A,024 Attach. A,
047 Attach. A,026 Attach. A,027 Attach. A, 028 Attach. A, 048 Attach. A, 048 Attach. B.
Exhibit No. 'l
ffi0
P. Ehrbar, Avista
Page 134 of 224
Page 3-20
-Ut€at,
-
Proic€t share
-Mirdl.mru
o
9soo
5400
s3m
s2m
slm
S
-LltuqP
-LIHEAP
2012
s4s2
s480
s235
s256
s380
2013
s439
s44s
5221
52s3
S354
2014
5460
s486
s232
5267
5377
2015
s4s7
s470
52s4
s232
S378
2016
544s
s439
5254
s260
s36e
2077
s4s6
s418
s3s9
s292
S397
-
Pro.iect Share
Miscellaneous
-Total
Average
Figure 3-12. Average Bill Assistance Gront by Funding Source
L ow-Inc ome We ath erizatio n S ervic es
Avista provides low-income customers with weatheizationrebates to reduce costs of energy
with the following qualiffing conditions.ae
o Primary fuel used for space heating must be Avista electric or natural gas service.
o Rebates must be submitted within ayear of completion of energy efficiency measure,
. Only new equipment qualifies.
o All improvements must be agency or contractor installed.
. The rebates are available for primary residential single family up to a fourplex, including
manufacfured and modular homes.
o Rebates are not available for seasonal or recreational homes or condos.
Low-income weatherization rebates fund such measures as air sealing, attic insulation, wall
insulation, duct sealing, and conversion from electric space heating and hot water to natural gas
space heating and hot water. The community action agencies select the clients and determine the
optimal measures for each home.
LIHEAP weatherization dollars and US Department of Energy Weatherization Assistance
Program (WAP) also fund weatherization services in homes of Avista low-income customers.
However, the Company only tracks low-income weatherization work that is funded through the
Avista Demand Side Management (DSM) Tariff Rider.
Exhibit No. 1
aeAvista Website: Rebates: Washington - Avista
Page 3-21 P. Ehrbar, Avista
Page 135 ot 224
-
O
Avista Low-Income Weatherization F unding
Avista's low-income home weatherization program is funded strictly through the company's
DSM Tariff Rider. The DSM tariff rate for electric service is established through the rate setting
process and decided by the Washington State Utilities and Transportation Commission. Table
3-14 presents the electric service DSM tariff rates which are applied to kWh sales in each of the
listed rate schedules to determine the available funding.so
Table 3-14. Electric Service DSMTarilf
To provide a simple combined view of the overall trends in the electric service DSM tariff rate
since 2015, we calculated a weighted average electric service DSM tariff rate for all affected rate
schedules. The weights are based on the projected dollar sales of natural gas in each of the
affected rate schedules listed in Table 3-14. Figure 3-13 illustrates the weighted average electric
service DSM tariff rate from 2012 to 2018.sr After increasing in 2013 the weighted average
decreased until an increase in August 2016. It increased again in 2017 and is projected to
increase in September of 2018.
3:
0.o050
0.0045
0.o040
0.o035
0.o030
0.o025
0.0020
0.o015
0.o010
0.o@5
o.omo
ol-Aug-12 01-4l{.13 01-ALg-14 ot-A'A-15 01-A(.€-16
Effectfue Date
ot-Arg-17 O1-Ar{-
Figure 3-13. Electric Service DSM Tariff (Weighted Average)
Table 3-15 presents the effective DSM natural gas tariffs for each customer class from August 1,
2012 to September I , 2}l8.s2 The DSM natural gas tariff rates are applied to Therm of natural
50 Response to DR 074 Attach. Revised
51 Response to DR 074 Attach. Revised
s2Ibid.
Exhibit No. 1
Effective Dates
($/kwh)
Schedules 01-Aug-12 01-Aug-13 01-Aus-15 08-Apr-16 0l-Aug-16 0l-Aug-17 0l-Sep-18
Residential 1,2 0.00168 0.00268 0.00215 0.00201 0.00262 0.00344 0.00433
General Service 11, 12 0.00235 0.00365 0.00289 0.00272 0.00362 0.00463 0.00597
Large General Service 21.22 0.00176 0.00276 0.00220 0.00208 0.00273 0.00366 0.00460
Extra Large General Service 25 0.001I I 0.00176 0.00137 0.00129 0.00172 0.00232 0.00297
Pumping 30,3t,32 0.00l ss 0.00245 0.00198 0.00190 0.00261 0.00341 0.00433
Street & Area Lishtins 41-48 0.02030 0.03 130 0.02400 0.02360 0.00862 0.01 21 5 0.02017
Weighted Average 0.00197 0.0031I 0.00247 0.00234 0.00276 0.00364 0.00469
Page 3-22 P. Ehrbar, Avista
Page 136 ot 224
6
gas sales in each of the listed rate schedules determine the available funding for DSM services.
The DSM tariff rate for natural gas service is established through the rate setting process and
decided by the Washington State Utilities and Transportation Commission.
Table 3-15. lrlatural Gas Service DSM Tariff
Schedules
Effective Dates
($/therm)
01-Aus-12 01-Nov-15 0l-Jut-16 01-Jun-17 01-Sep-18
General Service l0l. 102 0.02310 0.02'750 0.03472 0.02229 0.03028
Large General Service 1ll, ll2 0.01824 0.0209s 0.02475 0.01s81 0.01626
Extra Large General Service 121,122 0.01630 0.0196s 0.02t76 0.01614 0.01276
Intem.rptible l3l,132 0.0t476 0.02384 0.02300 0.01521 0.01132
Weiehted Average 0.02177 0.02578 0.03220 0.02071 0.0274s
To provide a simple combined view of the overall trends in the DSM natural gas service tariff
rate since 2015, we calculated a weighted average for all affected rate schedules. The weights are
based on the projected dollar sales of natural gas in each of the affected rate schedules listed in
Table 3-14.
Figure 3-14 illustrates the hend in the weighted average DSM natural gas tariff from August 1,
2012to September 1,2018.s3 The weighted average tariff increased from August 2012 through
July 20 1 6. It decreased in July 2017 , and it is projected to be increased in September 201 8.
However, the weighted average projected DSM natural gas tariff rate, effective September 2018,
is lower than the July 2016 rate.
E
o4
0.035
o.o1)
0.025
oo20
0015
0.o10
o.06
o.0@Ol-A({'l2 01-Ar{.11 01-Al{.14 0l-Ar{.15 0l-Ar{.16 01.Are-17 01'Ar{-18
Eff.ctlv. O.tG
Figure 3-14. Natural Gas Service DSM Tariff (Weighted Average)
Exhibit No. 1
C-Se nIoS.TVU:FTS-U andAVU-G- 1 9-0_
P. Ehrbar, Avista
Page137 ot224
s3Ibid.
Page 3-23
6
Low-income weatherization is frrnded as follows:
"Avista is ordered through General Rate Case settlements to spend tariff rider funds on
low-income weatherization. Since 2012, $2 million is set aside for Washington
customers who meet the income qualification requirements. This has been allocated to
six network agencies and since 2015 also includes a tribal housing authority. The
division of $2 million is done by determining the meter count in each county the
agencies serve. The percentage of meters is then applied to the $2 million to create an
allocation by agency for weatherization and other energy efficiency improvements for
the income qualified home." 54
Figure 3-15 presents overall funding trends and separates funding levels for electric and natural
gas customers.ss's6 Since 2015 the weatherization allocation to electric customers decreased
from23Yo to l6Yo while the allocation to natural gas customers increased from 77Yo to 84% of
the total allocation. Overall funding allocations have remained stable.
s2,@O,0@
s1,soo,0@
s1,@0,0@
ssoo,mo
s-2012
s1,989,8s2
sx3,833
sr,646019
2013
s1,974,533
s429,520
sr,14t013
2014
s1,$9,936
s370,495
sr,539,440
2015
s1,939,&i5
s446208
5r,493,627
2016
s1,983,215
s474,472
s1,so8,743
2077
s1,93Z08S
s312,431
sr,624254
:Total
-Electric
-Natural
Gas
Figure 3-15. Avista Low-lncome Weatherization Funding Trends
sa Response to DR 075.
ss For the purpose of this analysis the electric category includes electric-only customers while the natural gas
category includes both natural gas-only customers and dual-service (natural gas and electric) customers.
s6 Responses to DR's: 033 Attach. A, 034 Attach. A, 035 Attach. A, 036 Attach. A
Exhibit No. 1
Page j-24 P. Ehrbar, Avista
Page 138 ol 224
6
Number of Low-Income Weatherization Grants
Figure 3-16 illustrates that the number of Avista low-income weatherization jobs increased from
2012to 2013 when it reached its highest level during the study period.57 The trend in the
number of low-income weatherizationgrants then began a downward trend from 2013 through
2017 . The decrease in the number of grants reflects the increasing average cost of the
weatherization jobs.
Figure 3-l6. Nuntber of Low-Income Weatherization Grants
Average Weatherization Job Costs
Figure 3-17 presents the average cost of low-income weatherization jobs for electric and natural
gas customers.s8 While the average cost of both electric customer and natural gas customer jobs
have increased consistently since 2013,the cost ofnatural gasjobs has increased at a faster rate.
s10,0m
59,soo
59,@o
s8,soo
s8,cDo
57,soo
s7,mo
s6soo
56,Gro
Electric
Natural Gas
Average
20r2
57,475
S2283
s2316
2013
s6,608
s6,28r
s634e
-Electric
2014 2015 2016
s7.26s S7,s63 szs31sz62l s7,n9 s8,671
57,s9 57,728 58,368
-Nalural6as, -
Average
2017
s7,630
59 7?6
s9,313
Figure 3- I 7. Average Cost of Weatherization Jobs
57 lbid.
s8 Responses to DR's: 029 Attach. A,030 Attach. A,031,032 Attach. A, 033 Attach. A, 034 Attach. A, 035 Attach.
A, 036 Attach. A,049 Attach. A
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 139 of 224
350
3{E
2SO
2@
150
1@
50
0
E:=lToral
-NaturalGas
20r2
272
46
226
2013
311
69
246
2014
253
51
202
2015
251
59
r92
2015
237
63
174
2017
208
4l
167
Page 3-25
6
Inflation Adjasted Fanding Levels
To account for the cost of living increases since 2012, we calculated inflation adjusted funding
levels for both low-income bill assistance and low-income weatherization programs.
Figure 3-18 presents inflation adjusted bill assistance funding levels for Avista customers from
all funding sources including LIRAP, LIHEAP, Project Share and MISC using the Bureau of
Labor Statistics Inflation Adjustment Calculator.se The inflation adjusted data reflects the
buying power of the funding based on the 2012-dollar value. The inflation adjusted curve in
Figure 15 represents the trend in the buying power value of bill assistance over the evaluation
period. Inflation adjusted bill assistance funding has increased since 2015 and 2017 levels and
are above 2012 funding.
9lq(H,00o
s9s6,m
590m,0@
s6.500,06
$8,000,(m
s7,tm,00o
sr,o@.mo
55,s@,OCX)
s5,0@,0(D
-loTAt
]0tl
sEa33.0:6
s&t13.0,5
2013
$8,?$,r95
s8,l 20,722
rcl{
t8,776,156
s&449.73'
,0ts
56,84a,652
s6,61o,3E6
2015
57,r04.5*
57,335,6{2
tol7
59,612,6Y
59,061,{88
Figure 3-18. lnflation Adjusted BillAssistance (All Sources)
Figure 3-19 presents inflation adjusted Avista low-income weatherization funding levels, using
the Bureau of Labor Statistics Inflation Adjustment Calculator. Inflation adjusted Avista
weatherization funding has decreased from 2012 to 2015 and continued to decrease through
20t7.
s2,6qoo
s2"s0,0@
91,9s0,0@
s1"90,@
5r,850,0{D
s1,800,0(}1
5r,r5q0@
51'76'0@ 2ol2 zorr 2014 2ors 2ol. 2o1t
-rd.t
Sl,9gt8s2 $1"9r4,st3 $LS9.9r6 9r,9i98t5 St,9g3,2ts SL93?,69
-rdt{ionadirskd
st,s9.852 st,9,r0,489 t!,83q697 s1,t653r5 S1,88E258 S!"8r4,504
Figure 3-19. Inflation Adjusted Avista Weatherization Funding
5e https ://www.bls. gov/datalinfl ation_calculator.htm
Exhibit No. 'l
Page 3-26 Case Nos. AVU-E-19-0_ and AVU-G-19-o_
P. Ehrbar, Avista
Page 14O of 224
-
Inlhtkf adiult.d
6
Summary - Task 3, Part B
Avista low-income customers are provided with bill assistance and weatherization services
funded by Avista and several other Federal, State, and community-based organizations. These
services are provided in cooperation with the Washington State Department of Commerce, the
Community Action Agency network across the State, the LIHEAP program, Project Share and
directly through community-based groups.
We have provided an overview of all the bill assistance programs available to Avista customer
from all funding sources. Since Decoupling was implemented in 2015 the level of bill assistance
funding has increased. The increase in funding was driven by the Avista LIRAP program and
the Project Share program each of which showed significant increases while LIHEAP funding
remained level and MISC funding declined. Because of the increases in LIRAP and Project
Share funding the number of customers receiving bill assistance increased from 18,212 to 24,355
households. During the same period average bill assistance benefits increased from $378 to
$397 per grant.
While Avista customers receive weatherization services from several sources, only Avista
weatherization is tracked by the Company and is analyzed in this evaluation. Avista
weatherization funding remained level at approximately $2 million per year between 2012 and
2017. lnflation adjusted Avista weatherization funding decreased from2012 to 2015 and
continued to decrease through 2017.
During the 2015 to 2017 period the average Avista weatherization costs increased from $7,728 to
$9,313 per customer. Because of increasing costs and level funding, the number of
weatherization rebates decreased. Since decoupling was implemented in 2015, the number of
weatherization rebates decreased from 251 to 208.
This analysis did not evaluate whether the low-income energy assistance programs reviewed in
this report are adequate to meet the need. The RFP No. R -41321provided two Attachments that
address this question: Attachment G - An Estimate of the Number of Households in Poverty
Served by Avista Utilities in Washington Statu60 and Attachment H - The Self-Sfficiency
Standardfor Washington State 2014.61 We have analyzed these reports and have provided our
findings in Section 8 (Low-Income Appendix) of this evaluation.
60 An Estimate of the Number of Households in Poverty Served by Avista Utilities in Washington State, Brian
Kennedy, MS and D. Patrick Jones, Ph.D., Institute for Public Policy and Economic Analysis, May 2015.
6t The Sey-Sufficiency Standardfor Washington State 2014, Diana M. Pearce, PhD, Center for Women's Welfare
and the School of Social Work at the University of Washington, Revised August 2015.
Exhibit No. 1
Page 3-27 P. Ehrbar, Avista
Page 141 of 224
Modifications to Low-Income Programs
Task 3, Part C is defined as follows:
"(3c) A description of any modifications to conservation programs targeted to low-
income customers since the inception of the Mechanisms including changes to
funding levels as well as changes to specific measures."
The funding level for conservation programs targeted to low-income customers since the
inception of the Mechanisms in 2015 is best reflected in Figure 3-19, for which the relevant
portion is from 2015 onwards. As shown in this figure, Avista inflation-adjusted Weatherization
funding increased from 2015 to 2016 and then dropped in2017 . The unadjusted amounts were
$1,939,835 in 2015, $1,983,215 in2016 and $1,937,085 in 2017, or essentially, about
$2,000,000 per year. The adjusted amounts were $1,865,375 in20l5, $1,888,258 in 2016 and
$1,814,694 in20l7, or roughly from about $1,900,000 to $1,800,000 per year in real dollars.
From an administrative perspective, funding was essentially constant at $2,000,000 per year. In
real terms, funding dropped to about $1,800,000 in2017. This suggests that Avista might want
to take inflation into account in carrying out the 'ocarve out" for low-income in each year.
In 2015, the Company continued to reimburse Community Action Agencies for 100% of the cost
of installation for a select group of pre-approved energy-efficiency measures (Table 3-16). The
Company continued to offer an additional "Rebate List" of other energy efficiency measures
(Table 3-17). Payment for measures on the "Rebate List" covers only the energy value of the
measures. In this way, the CAAs are able to reliably secure funding for pre-approved measures
and to leverage utility funds for partial funding of other measures that improve functionality of
weatherization retrofits. Agencies can apply funds to electric or natural gas homes at their
discretion and to charge a fifteen percent (15%) administration fee.
For 2016, the same system was continued, but with some changes in the measure tables. The
2016 group of pre-approved energy-efficiency measures is shown in Table 3-18. Partial rebate
measures are shown in Table 3-19.
For 2017, the basic system was continued with changes in the measure tables. The2017 group
of pre-approved energy-efficiency measures is shown in Table 3-20. Partial rebate measures are
shown in Table 3-21. There was also a clarification that measures found in Washington's
Weatherization Manual priority list are deemed to be cost-effective and are paid at l00yo,
regardless of whether their computed Total Resource Cost (TRC) test value is below 1.0. Also,
Health and Safety dollars may be used to fully fund measures on the partial rebate list, at the
discretion of the CAAs.62
62 Low-Income program changes are sourced from the Washington DSM Annual Conservation Report & Cost
Effectiveness Analysis studies for 2015, 2016 and20l7.
Exhibit No. 1
6
Page 3-28 Case Nos. AVU-E-19-0- and AVU-G-19-0_
P. Ehrbar, Avista
Page 142 ol 224
Electric Measures
6
Table 3-16. Low-Income 100% Approved Measures (2015)
Air infiltration
lnsulation (floor, ceiling, wall)
Duct sealing
ENERGY STAR doors
Electric to Natural Gas Conversion
(Space and Water Heat)
ENERGY STAR Refrigerators
o lnsulation (Wall, Ceiling, and Floor)
. Air infiltration
. Duct sealing. ENERGY STAR doors
. ENERGY STAR windows
o
Table 3-17. Lotry-Income Partiol Rebate Measures (2015)
o Duct insulation
r ENERGY STAR refrigerators (for
replacement of a refrigerator that is not
fully operational)
. High efficient water heater
r Electric to air source heat pump
o Electric to natural gas water heater
. ENERGY STAR windows
o Duct insulation. High efficiency furnace
. High efficiency water heater
Table 3-18. Low-Income 100% Approved Measures (2016)
. Air infiltration
o Duct sealing. ENERGY STAR doors
o ENERGY STAR windows. High efliciency air source heat pump
(8 HSPF)
. Electric to air source heat pump
o lnsulation for attic, walls, floors. and
ducts
. Air infiltration
. Duct sealingo ENERGY STAR doorso ENERGY STAR windowsr High efficiency fumace (90% AFUE). lnsulation for attic. walls, floors, and ducts
r Electric to natural gas furnace. Electric to natural gas furnace and water heat
Exhibit No. 1
Case Nos. AV[I-E1 9-0_ ahdAW-G-1 9-0_
P. Ehrbar, Avista
Page 143 ol 224
Electric Measures Natural Gas Measures
Electric Measures Natural Gas Measures
Page 3-29
Natural Gas Measures
Fuel Conversion Measures
Electric Measures NaturalGas Measures
6
Table 3-19. Lotu-lncome Partial Rebate Measures (2016)
r High efficiency water heaters (0.93 EF)
r ENERGY STAR refrigerators
r Ductless Heat Pumps
a High efficiency water heaters (0.62 EF)
a Electric to natural gas water heater
Table 3-20. Low-lncome 100% Approved Measures (2017)
a Air infiltration
Duct sealing
lnsulation for attic, walls, floors,
and ducts
LED lighting
r Air infiltration
o Duct sealing
. ENERGY STAR doorsa
a
. ENERGY STAR windows
. High efficiency furnace (90o/o AFUE)
. High efficiency gas water heater
o lnsulation for attic, walls, floors, and ducls
. Electric to natural gas furnace
. Electric to natural gas water heat
r Electric to duclless heat pump
Table 3-21. Low-Income Partial Rebate Measures (2017)
a Heat pump water heaters
ENERGY STAR refrigerators
ENERGY STAR doors
ENERGY STAR windows
Electric to air source heat pump
a
a
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-I9-O_
P. Ehrbar, Avista
Page 144 ol 224
Fuel Conversion Measures
Electric Measures Natural Gas Measures
Electric Measures Natural Gas Measures
Page 3-30
0
Effect on Low-Income vs. Average Residential
Task 3, Parts A and D are combined and are presented in Part A, above.
Other tr'actors
In this section we examine additional contrast between low-income customers and other
residential customers using premise specific data for nearly 130 thousand Avista residential
customers in Spokane County.
The objective of Task 3 Part E, as stated in the request for proposal, is shown below:
"To the extent data is available, Consultant should evaluate other foctors such as
household size, housing stock (e.g. mobile home, multifamily) and heat source (e.9.,
electric spoce heat) and the effect of seasonality when comparing the impact of
decoupling on low-income customers versus other customer groups (such as
average residential customers). "
There were no specific evaluation questions related to this objective in the RFP.
Our team approached this task by first exploring the possibility of obtaining housing attribute
data such as size and vintage of construction directly from Avista or from secondary sources
such as the US Census. Avista does not maintain housing attribute data within their customer
information system. We also explored using the American Community Survey (Census) and
American Housing Survey (HUD) but found the data details did not provide the ability to drill
down and compare households by income levels, energy usage and housing attributes at the same
time.
We turned next to the possibility of acquiring detailed housing attribute data directly from the
Spokane County Assessor office and merging the data with Avista's customer information.
After some initial testing to see what data could be acquired and a subjective assessment of data
quality, we decided to pursue the development of a site-specific data base combining Avista's
billing data and low-income status information with Spokane County's assessor data. The
resulting data base of nearly 130 thousand Avista customers in Spokane County provides the
ability to drill down in ways that would not otherwise be possible to compare housing size, type,
vintage and energy intensity between low-income and other residential customers.
Overview of Approach
The approach of combining Avista residential billing records with assessor data was selected to
overcome the lack of housing attribute data. Our team has had extensive experience combining
county assessor data with utility data and we understand the rich analytical database that results
from this effort. The resulting database is expected to provide a level of understanding and
insights into contrasts between low-income and other residential customers that would not
otherwise be possible in this evaluation.
Because of the time requirements involved with processing county assessor data and the fact that
data structure, format and processes vary greatly between counties, we focused our effort
exclusively on Spokane county which accounts for about 75 percent of all households within
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-I9-0_
P. Ehrbar, Avista
Page 145 ot 224
Page 3-31
O
o Owner occupancy was assigned by comparing the physical address of the parcel with the
mailing address of the owner. The overall results compare favorably with Census
estimates for the County.
. When possible, heating fuel was assigned based on the heating method.
Accuracy of assessor data tends to be highest for variables such as square footage of the
structure, number of bedrooms and year built. Variables related to heating and cooling
equipment tend to be less accurate and are often unavailable for a parcel.
To combine the county data with Avista data, we first summarized Avista's billing records to an
individual premise level using standardized addresses. Low-income premises are flagged and
the type of Avista service assigned as electric only, natural gas only or both. A low-income
premise flag is assigned based on the existence of the premise in customer data of participants in
one of Avista's low-income programs63. Site address is the information in common between the
Avista records and assessor records. In order to increase the quality of the join, we first address
standardized the two datasets using AccuMail software.6a The datasets are address standardized,
so an address-component-based match key can be used to join the Avista billing records with the
Spokane county assessor data. There are limitations to joining utility records with assessor data
in this manner, but the approach is highly accurate for single family housing. It tends to break
down in instances where there is not a one-to-one correspondence between a utility premise
record and a tax parcel record such as multifamily housing (one parcel and many utility
customers).
A match key must be present in both the Avista data and the county assessor data for a premise
to be retained for this analysis. Table 3-22 shows premise counts by residential group and
service type that passed the match criteria.
63 See Section 3a for more information.
6a AccuMail is certified by the US Postal Service for address standardization and processing. Address
standardization helps to improve match results. If addresses are incorrectly spelled, or components (eg, zip plus 4,
pre- and post-directionals and/or city) are missing, or unit numbers are in the wrong position, the match routine will
be less reliable.
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 146 of 224
Page 3-32
Avista's Washington service territory. From the assessor data we compiled parcel-level housing
attributes, including square footage, year built, number of bedrooms, heating and cooling
method, housing type, and market value. From this data, we also inferred certain variables as
follows:
6
Table 3-22. Avista Customer Counts by Residential Group and Service Type
Avista Service Type Residential Low-Income Total
Low-Income
Percent
Service Type
Percent
Electric Only 14,373 4,1 50 18,523 22%t4%
Gas Only 14.527 892 15.4t9 6%12%
Electric and Gas 8l.l I 5 14.876 95.991 l6Yo 74%
Total 110.015 r9,918 129.933 l5o/"IOOYI
The merge results in nearly 130,000 Avista premises matched to Spokane assessor data, of which
15 percent are classified as low-income. Nearly three-fourths of the premises receive their
electric and natural gas service from Avista. The remainder of this section compares housing
attributes and energy usage between the two residential groups; low-income and other
residential. For ease of discussion in the remainder of this section, we use the term "residential"
to mean all other residential customers not identified as low-income.
Energt Usage
Annual energy usage for 2017 is shown in Figure 3-20
12,608 317 900
8m
7m
11,911
700
q
Eoc
oe
3
o
6m'Eo
5mLoamf
3mE
F200
1m
0
Low-lncome oo Residenlial
Figure 3-20. Annual 2017 Unad.justed Billed Energ,t Usage per Premise
Annual kWh usage for low-income premises was about 6 percent higher than residential
premises in20l7. For natural gas the opposite is true with low-income premises using about 16
percent less therms over the year than residential. As will be shown below, low-income
premises are smaller on average than residential. Figure 3-21 shows energy usage per square
foot for both kWh and therms between the two groups.
Exhibit No. '1
Case Nos. AVU-E-19-0_ and AVU-G-I9-0_
P. Ehrbar, Avista
Page 147 ol 224
Page 3-33
14,mO
12,mO
10,mo
8,0@
6,0m
4,00
2,0@
12.0
1
0.6
0.5 oo40.4 g
(9
=E03toCL
0.2 Eo
F0.1
o
0.532
Residential Lorv-lncome
EK\irh tsIherms
10.38
Lorv-lncome
0.0
6.0
4.0
2.0
0.0
Residential
o
q
o
0o
x
7.24
Figure 3-21 . Annuol 2017 Unadjusted Billed Energy per Square Foot
With smaller homes using more electricity, the low-income group's kwh per square foot
averaged over 40 percent higher than residential premises. Possible explanations for this
difference are explored below. Therm usage per square foot is also higher for low-income
premises, averaging 16 percent more in 2017 thanresidential.
Ho using Churacteristics
Housing characteristics obtained from Spokane County Assessor records are shown in the table
below. Mean values and differences between the two residential groups are shown for each of
the characteristics listed. The last column shows the directional energy use impact of low-
income relative to residential. For example, an upward affow on a characteristic means that
considering that attribute alone, low-income energy usage would be expected to be higher than
residential usage. A listing of "electric" with a directional indicator means that the relative fuel
usage impact only applies to electric and not natural gas usage.
Toble 3-23. Comparison of Housing Characteristics
Characteristic Low-
Income Residential Difference Percent
Difference
Relative Energy
Use Impact
Year Built 195 I 1968 08)1
Finished Square Feet 1,403 I,916 (sl3)-27%J
Market Value $117,810 $ l9l,966 -$74,1 s5 -39%1
Bedrooms 2.8 3.2 (0.44)-14%J
Owner Occupied 62%85%-23o/o e
Avista Natural Gas Service 79%87%-8%J (Electric)
Air Conditioning 2t%47%-26%J (Electric)
Low-income homes are 18 years older than residential homes on average. Older homes are more
likely to have less thermally efficient building shells than newer homes. The impact of this
characteristic is to increase low-income energy usage relative to the residential group. Low-
income homes are about 500 square feet smaller on average compared to residential, a
substantial twenty-seven percent (27%) percent difference. Market value and market value per
Exhibit No. 1
0
Page 3-34 uase Nos. AVU-E-I9-U_ ano AVU-G-I9-U_
P. Ehrbar, Avista
Page 148 of 224
0.458
6
square foot are indicators of current quality of construction and building shell efficiency and
suggest that low-income homes will use more energy than residential, all other things equal.
The number of bedrooms is not only another measure of size of home, it is a beffer correlate to
size of household and baseload energy usage than is square feet. Fewer bedrooms in low-income
housing suggest lower energy usage than residential. Average size of households may also vary
between the two groups. Owner occupancy is lower in low-income housing than it is in
residential. This variable says more about the occupant's ability to make energy efficiency
improvement decisions than it does about relative energy usage.
The percent of the group with natural gas service from Avista is an indication of the
predominance of natural gas heating. A lower percentage of low-income homes with natural gas
service means a greater reliance on electricity and other fuels for space and water heating in low-
income homes than found in the residential group. This characteristic coupled with the age and
quality differences of the building shell are likely to explain alarge proportion of the greater
electric usage per square foot in low-income homes.
Assessor data regarding heating, ventilation and air conditioning (HVAC) equipment is generally
less reliable than square footage and year built. Still the data can be useful for comparing
relative values between groups. Air conditioning is far less prevalent in low-income homes than
it is in residential. This characteristic taken alone suggests less electric usage in low-income
homes compared to residential.
Housing Type snd HVAC Equipment
It is important to keep in mind that the approach of combing utility records with assessor records
results in a data set that is single family construction centric. Utility customers living in
multifamily housing are largely omitted from the combined data base of 130 thousand premises.
The percentage of the 130 thousand homes by type of housing and residential group is shown in
Table 3-24.
Table 3-24. Distribution of Housing Types
Housing Type Low-[ncome Residential
Condos and Townhomes 1.2 2.2
Mobile Homes l0.l 3.9
Plexes 7.0 2.9
Single Family 81.6 91.0
Total 100.0 r00.0
The nearly eighty-two percent (82%) of low-income customers in single family homes is nearly
10 percentage points lower than residential. That difference is made up by a higher percentage
of low-income customers in mobile homes and plexes (duplexes, tri and quad).
The distribution of heating equipment is shown in the table below.
Exhibit No. 1
Page 3-35 P. Ehrbar, Avista
Page 149 oI 224
UaSe NOS. AVU-tr- lV-U ano AVU-\r- ly-U
Table 3-25. Distribution of Heating Equipment
Heating Equipment Low-Income Residential
Forced Air Fumace 81.4 85.7
Zonal 14.0 7.9
Heat Pump l.l 3.6
Other 3.4 2.9
Total 100.0 100.0
The majority of heating equipment is some form of forced air system. These include wall and
floor systems as well as ducted systems. Zonal is more prevalent in low-income housing, not
surprising given the smaller and less expensive housing stock of low-income customers.
Cooling equipment distribution is shown in the table below.
Table 3-26. Distribution of Cooling Equipment
Cooling Equipment Low-Income Residential
Central Air Conditioning 18.3 40.9
Heat Pump l.l 3.6
Other 1.9 2.0
None 18.6 s3.5
Total 100.0 100.0
Central air conditioning is far more prevalent in residential than low-income homes. Assessor
data likely understates the prevalence of window units and these relatively inexpensive and
inefficient systems are more likely to be found in low-income homes than residential.
Summary - Task 3e
In this section housing attributes and energy usage of low-income and other residential homes
are compared using a data set developed for this evaluation of nearly 130,000 premises with
Avista residential customer records combined with Spokane County Assessor data. The
resulting data is single family centric, with multifamily underrepresented in the results. Data on
heating and cooling equipment may also be incomplete or out of date for what is currently used
at the premise. Notwithstanding these limitations, the data provide a rich set of information for
insights between the differences of low-income and other residential premises.
The average low-income customers used six percent (6%) more electricity per premise in2017
than other residential customers. Low-income homes were also substantially smaller. With
higher use in smaller homes, electric use per square foot in low-income homes was about forty
percent (40%) higher than for other residential customers. Analysis to determine why this is the
case is beyond the scope of this evaluation but older less efficient homes and greater reliance on
electric space heating in low-income homes are at least part of the explanation.
The average low-income customer used 16% less natural gas per premise than other residential
customers. On a per square foot basis, natural gas use was sixteen percent (16%) higher in low-
income homes than other residential. Much of this difference is likely due to older less efficient
building shells in low-income housing units.
Exhibit No. 1
0
Page 3-36 P. Ehrbar, Avista
Page 15O ol 224
uase t\os. AVU-tr- tv-u_ ano AVU-(J- rv-u_
Section 4. Ana is of Revenue Effects
In this section we examine the effects of the decoupling mechanisms on Avista's revenue. The
objective of Task 4, as stated in the request for proposal, is shown below:
"Analysis of the Mechanism's impact on Company revenues (i.e., whether there has
been a stabilizing effect)."
Relating to this objective are the following evaluation questions, also taken from the RFP:
"V[hat impact did the Mechanisms have on the Company's revenues (i.e., whether
there has been a stabilizing effect)? "
What were the causes of the deviation of actual revenue-per-customerfrom
autltori z ed r evenue-p er - cus to m er? "
"Pleose provide analysis and trends on whether the rate cap wcts reached and the
results of the earnings test."
"Whatfactors impacted the deferral and rate changes, and what was the magnitude
of that impact? (e.g., weather, customer counts, conservation, economy, etc.)"
"Wat was the impact of the Decoupling deferual on Avista's revenues and rates? "
"V[hat was the ffict of updates to the decoupling baseline and resulting effects on
deferrals under the mechanisms? "
Our discussion in this section is organized by each of the evaluation questions listed above.
Much of the data used to address these questions has been presented in earlier sections of this
report and repeated here for ease ofdiscussion and the convenience ofthe reader.
Has Decoupling Stabilized Revenue
The question as stated in the RFP is:
"V[rhat impact did the Mechanisms have on the Company's revenues (i.e., whether
there has been a stabilizing effect)? "
This is a straightforward question and easy to answer by comparing actual revenue with actual
revenue plus deferred revenue. Here the limiting factor is the relatively short three-year period
that the mechanism has been in place. In order to answer this question, we calculated the annual
variation in revenue over the 2015 to 2017 period with and without the revenue from decoupling
deferrals. We used the coefficient of variation, calculated as the standard deviation divided by
the mean, as our measure of variability. Figure 4-l shows the results of our calculations for
electric revenue.
Exhibit No. 1
6
Page 4-l Case Nos. AVU-E-19-O_ and AVU-G-I9-U_
P. Ehrbar, Avista
Page '151 o1224
o
5.O%
4.5%
4.0%
3.5%
3.0%
2.5%
2.Wo
1.5%
1.096
0.5%
0.0%
4.4%
.9
.g
(o
o
c
.c,
(uoU
2.4%
1.AYoni o.9%
O.4o/on
Res Non-Res Total
IWithoutDecoupling trWithDecoupling
Figttre 1- I . Electric Revenue Variability (201 5-2017)
The bars labeled "Without Decoupling" refer to base rate revenue only and does not include
deferred revenue through the decoupling mechanism. Bars labeled "With Decoupling" include
base rate and decoupling deferral revenue. Results are shown for both decoupled rate groups and
their total. It is clear from the results shown in Figure 4-l that there has been a stabilizing effect
on revenue as a result of decoupling. For both rate groups, variability is roughly one third of the
level without decoupling deferrals.
Variation in natural gas revenue is shown in Figure 4-2.
Figure 4-2. lVatwal Gas Revenue Variabilint (2015-2017)
Decoupling has also helped to stabilize natural gas revenues. Although the stabilizing effect is
not as large for the natural gas rate groups as it is for electric rate groups, revenues from nafural
gas residential customers are about four percentage points less variable with decoupling than
Exhibit No. 1
Page 4-2
Page 152 ol 224
12.Wo
r0.096
8.0%
6.Wo
4.O%
2.O%
0.0%
10.1%
Res Non.Rest Without Decoupling tr With Decoupling
9.0%
Total
o
r!
(E
o
.g
o,o(J
6.70h5.9%
5.3%
I
6
without, a drop in variability of roughly 40Yo. Yariability in the non-residential rate group is
nearly two percentage points lower with decoupling, a roughly 30o/o drop in variability.
Revenue Deviations from Planning Assumptions and Causes
Some of the revenue related evaluation questions have to do with the magnitude and causes for
deviations from planning assumptions. These questions as stated in the RFP are:
"V[/hat were the causes of the devtation of actual revenue-per-customer from
author ize d r ev enue-p er- cus t omer? "
"V[hat factors impacted the deferral and rate changes, ond what was the mognitude of
that impact? (e.g., weather, customer counts, conservation, economy, etc.)"
Actual and authorized revenue-per-customer is shown for electric rate groups in Table 4- 1.
Table 4-1. Authorized and ActualElectric Decoupled Revenue per Customer
Year
Residential Non-Residential ---------
Authorized Received
Percent
Difference Authorized Received
Percent
Difference
2015 $709 $673 -5.1%$4,209 $4,279 1.7%
2016 $73s s685 -6.80/,$4,4s3 $4,396 -1.3%
2017 $738 $748 1.4%$4,4ss $4,405 -t.t%
Avista received less decoupled revenue per customer from the residential group than was
authorized in 2015 and20l6. This pattern was reversedin2}lT when Avistareceivedl.4Yo
more revenue per customer than authorized. Decoupled revenue per customer for the non-
residential rate group exceeded the authorized level in 2015 but fell short in 2016 and20l7. The
percent difference shown for residential customers in Table 4-l closely follows the difference
between actual and planned use per customer examined in Section 2. Test year and actual
electric usage, customer counts and use per customer are shown for each deferral year in Table
4-2.
Table 4-2. Test Year and Achtal Electric Usage, Customers and Use per Customer
2015 2016 2017
Usage
(MWh)
Use per
Customer(kwh)Usage
(MWh)Customers
Use per
Customer(kwh)Usage
(MWh)Customers
Use per
Customer(kwh)
Residential
Test Year 2.43'1,508 207,8s0 tr,727 2,378,478 205.172 1 I,593 2.378,478 205.172 l 1.s93
Actual 2,323300 207,371 11.204 2,288,227 209,864 10,903 2,492,293 212,49s I I ,729
Change from Test Year (l 14,208)(479)(s24)(90.251)4,692 (68e)I13,8t5 7,323 136
Percent Change -4.7%-0.2%-45%-3.8%2.30/o -s.9%4.8%3.6%1.2yo
Non-Residential
Test Year 2. I 50.843 35,277 60,970 2.144.857 34.823 61 .593 2.144.857 34.823 61.593
Actual 2,t79,747 35,265 6l,810 2, I s8,998 35.617 60.618 2.1 84.830 35.994 60.700
Change from Test Year 28,904 (12)840 14,142 794 (97s\39,974 I,l7l (8e3)
Percent Change 1.3%0.0%1.4%0.7%2.3%-1.6%1.9%3A%-1.5%
Because Avista's decoupling mechanism is structured to allow a certain level of revenue per
customer, more or less customers than planned does not lead to greater deferral balances, all
Exhibit No. 1
P. Ehrbar, Avista
Page 153 ot 224
Page 4-3
6
other things equal. Avista relies on volumetric charges to recover a portion of fixed costs for all
rate groups and fuels. This causes use per customer to be an important factor in determining
defenal balances and decoupling rates through the decoupling mechanism. More specifically,
changes in use per customer from levels used in the test year to set decoupled revenue will lead
to positive or negative deferral balances depending on the direction of change, all other things
equal. Higher use per customer will cause negative deferrals and lower use per customer will
result in higher deferrals, again all other things equal.
Considering electric residential as an example, actual decoupled revenue per customer was 6.802
lower than authorized in 2016 (Table 4-1). During the same period customer counts were 2.3
percent higher than the test year and use per customer was 5.8% lower (Table a-D- Higher than
planned customer counts did not drive authorized revenue higher. Rather, as designed and
expected, use per customer explains nearly all of the lower than authorized revenue per
customer. A comparison of the values in Table 4-1 and Table 4-2 show that almost all of the
variance in revenue per customer can be explained by differences in use per customer.
Two important factors causing use per customer to vary from test year are actual weather
deviations from normal weather and acquired energy efficiency savings through Avista
programs.65 There are other factors of course but these two are either known in the case of
energy efficiency or readily measurable in the case of weather. Changes due to weather are
shaightforward calculations. Avista provided the weather impacts and supporting monthly
details by rate schedule showing the deviation in heating and cooling degree days from normal
and the corresponding model coefficient on each weather term. Energy efficiency impacts are
calculated as cumulative savings from Avista programs since the test year.
The results of these calculations are shown in Figure 4-3 for the electric residential rate group.
5.096
3.Cfro
1.096
-t.M
.3.0%
-5.0%
-7.wo
r"u
l---- 201s -----l l---- 2016 -----l
lTotal BWeather EEnergy Effkiency trOther
| ----- 2017 ------ |
Figure 4-3. Percerttage Change in Use per Custonter, Electric Residential
Considering 2017 results, use per customer was l.2o/, higher than test year assumptions.
Weather impacts alone are estimated to have pushed electric residential use per customer 4.602
higher. The 2017 weather impact was largely offset by a 2.5% drop in use per customer due to
6s For this analysis, normal weather is defined as a thirty-year average.
Exhibit No. 1
Page 4-4 uase Nos. AVU-ts-]9-U_ and AVU-G-I9-U_
P. Ehrbar, Avista
Page 154 of 224
6
Avista's energy efficiency achievements. The "Other" category is simply the difference between
the total and the readily quantifiable factors of weather and energy efficiency. Other unidentified
factors have pushed use per customer lower and have been lessening in influence over time.
For electric residential customers it is clear that weather impacts on use per customer can be
large and work in either direction. It is also true that energy efficiency impacts always push use
per customer lower and that downward influence becomes more pronounced the further in time
an evaluation year is from the test year. Cumulative energy efficiency savings will reset with a
new rate case and test year.
Figure 4-4 shows a plot of total and each factor's influence on the percent change in use per
customer from the test year for the electric non-residentialrate group.
1.096
O.gr
-1.O%
-2.096
-3.0,6
-4.AYo
I*;c 0"996
l--- 2o1s ----l l--- 2016 -----l
lTotal !Weather @Energy Effkiency EIOther
| ---- 2017 ----- |
-1.5r
Figure 4-4. Percentage Change in Use per Customer, Electric Non-Residential
Avista's energy efficiency achievements have been the primary factor influencing changing use
per customer in the electric non-residential group. From having no influence in 2015 because
they were implicitly included in test year assumptions, energy efficiency impacts more than
offset weather and other factors in20l7 causing an overall drop in use per customer of l.5o/o.
Weather appears to be far less influential in electric non-residential customer usage than it is for
the electric residential group. Other unidentified factors have pushed use per customer higher at
a small but fairly consistent percentage over time. Actual and authorized revenue-per-customer
is shown for natural gas rate groups in Table 4-3.
Table 4-3. Authorized and Actuol NaturalGas Decotrpled Revenue per Customer
Residential Non-Residential -----
Year Authorized Received
Percent
Difference Authorized Received
Percent
Difference
2015 $280 $24s -12.5%$4,509 $3,835 -14.9o/o
2016 $347 $299 -13.8%$s,097 $4,338 -14.9%
2017 $3s I $364 3.7%$5.128 $4.828 -5.9Yo
For reasons discussed above for electric, the percent difference between authorized and actual
revenue per customer shown in Table 4-3 closely follows the difference between actual and
Exhibit No. 1
UASE NOS. AVU-E-I9-U ANO AVU.U-]9-U
P. Ehrbar, Avista
Page 155 of 224
Page 4-5
6
planned use per customer. Test year and acfual nafural gas usage, customer counts and use per
customer are shown for each deferral year in Table 4-4.
Table 4-4. Test Year and Actual Natural Gas tJsage, Customers and Use per Customer
2015 2016 20t7
Usage(MWh)Customem
Use per
Customer(kwh)Usage(MWh)Custom€re
Use per
Customer(kwh)Usage(MWh)Customere
Us€ per
Customer
akwh)
Residential ----
Test Year lt7,011,207 I 50,1 86 779 120,721,607 148,99s 810 120,721,607 148,995 810
Actual 103.436.220 t5t-254 684 108,796,187 I s3,99s 706 131,782,922 r57,563 836
Change from Test Year ir3.s74.987\1.068 (9s)/L1.925.420\s.000 004)I 1.061.3 l5 8.568 26
Percent Change -11.6%0.7%-t2.2%-9.9%3.4%-12.8o/o 9.2%5.8%3.2%
Non-Residential
Test Year 51.764.097 2-548 20-316 52.606.812 2,s84 20,358 s2,606,812 2,584 20,358
Actual 4s,886,s68 2,65t 17.309 48.208.894 2.770 17.404 55.684.308 2.918 I 9.083
Chanse from Test Year (5.877.529\103 (3,006)(4,397.918'l 186 (2,954'l 3,077,496 334 0,27s\
Percent Change -11.4%4.0o/o -14.8%-8.4o/o 7.2%-14.5o/o s.8%t29%-6.3%
Considering natural gas non-residential as an example, actual decoupled revenue per customer
was l4.9Yo lower than authorized in 2015 (Table 4-3). During the same period customer counts
were 4.0 percent higher than the test year and use per customer was I4.8% lower (Table 4-4).
Higher than planned customer counts did not drive authorized revenue higher. Rather, as
designed and expected, use per customer explains nearly all of the lower than authorized revenue
per customer. A review and comparison of the values in Table 4-3 and Table 4-4 also show that
almost all of the variance in revenue per customer can be explained by differences in use per
customer.
Two important factors causing use per customer to vary from test year are actual weather
deviations from normal weather and acquired energy efficiency savings through Avista
programs. There are other factors of course but these two are either known in the case of energy
efficiency or readily measurable in the case of weather. Changes due to weather are also
straightforward calculations. Avista provided the weather impacts and supporting monthly
details by rate schedule showing the deviation in heating and cooling degree days from normal
and the corresponding model coefficient on each weather term. Energy efficiency impacts are
calculated as cumulative savings from Avista programs since the test year.
The results of these calculations are shown in Figure 4-5 for the natural gas residential rate
group.
Exhibit No. 1
Page 4-6 P. Ehrbar, Avista
Page 156 ot 224
\,ase l\OS. AVU-tr- lY-U_ anO AVU-tr- I V-U_
6
6.@6
4.Uo
2.@6
o.m6
-2.Wo
-4.OYo
4.AYo
-8.07o
-ro.M
-12.Vo
-14.Mo
tu
PE\
0.016 EI*"
\d:,#
| ---- 201s ----- | | --*-- 2015 ----- |
lTotal EWeather trEnergyEfficiency OOther
l---- 2017 -----l
Figure 4-5. Percentage Change in Use per Customer, Natural Gas Residential
Weather is clearly the predominant factor in understanding changes in residential therm use per
customer from the test year. The total change in use per customer tracks the warmer than normal
heating seasons in calendar years 201 5 and 2016 and slightly colder than normal heating season
in calendar year 2017. Energy efficiency impacts on use per customer usage are a small factor in
understanding overall change from the test year. Natural gas prices have been persistently low,
squeezing the cost effectiveness of natural gas efficiency programs. Other unidentified factors
were small in 2015 and20l7 but relatively high in20l6. One possible explanation is that the
2016 weather adjustment was understated by the weather normalization model.
Figure 4-6 shows a plot of total and each factor's influence on the percent change in use per
customer from test year assumptions for the natural gas non-residential rate group.
4.go
2.Vo
O.go
-2.O%
4.A96
-6.096
€.o%
-10.096
-L2.Vo
-r4.M
-16.Wr
l;;I;-il
| ---- 201s ----- | | ---- 2016 ----- |
ITotal trweather BEnergyEffkiency OOther
| ------ 2017 ----- I
Figure 4-6. Percentage Change in Use per Customer, Natural Gas Non-Residentiol
Except for weather in20l7, all factors in each year have contributed toward lower use per
customer than test year assumptions. Unlike any of the other electric or natural gas rate groups,
Exhibit No. 1
UASE NOS. AVU-E-]Y-U ANd AVU-G.I9.U
P. Ehrbar, Avista
Page 157 o1224
Page 4-7
0
other factors are an important influence in each of the years examined. Other factors are by
definition unquantified but could include increased efficiency outside of Avista's energy
efficiency programs, lower use of natural gas due to fuel substitution (e.g. increased use of
biomass in cogeneration) and cutbacks in customer facility operations. Weather is also
influential although less so than natural gas residential customers. Energy efficiency impacts on
use per customer usage are a small factor in understanding overall change from the test year.
Again, this could be due in part to persistently low natural gas prices putting pressure on the cost
effectiveness of natural gas efficiency programs.
Avista's electric and natural gas energy efficiency programs are discussed in detail in Section 3
and Section 6 of this report. An examination of actual weather experienced over the three
evaluation years is presented in Section 2.
Review of Rate Cap and Earnings Test
The question as stated in the RFP is:
"Please provide analysis and trends on whether the rate cap wcts reached and the results of the
earnings test?
The earnings test is calculated to determine the amount of excess earnings, earnings over the
allowed rate of return. [f excess earnings exist, Avista shares 50 percent of the excess earnings
with the residential and non-residential rate groups. Table 4-5 shows the level of shared revenue
(50% of excess revenue) in each year for the electric system and natural gas system.
Table 4-5. Earning Test Shared Revenue
Year Electric Natural Gas
(lhousands of dollars)
2015 $899 $0
2016 s2,597 $2927
2017 $ I ,493 $2,600
Normalized revenue for the applicable year is used to determine the split of shared revenue
between the two rate groups. Shared earnings are paid by Avista to each customer rate group
through the decoupling rate established with each annual filing.
The decoupling sefflement stipulates that the change in the decoupling rate cannot add more than
3 percent to expected revenue before the change. Ifnecessary, decoupling rates are capped to a
level that limits the expected change in revenue to 3 percent and the amount of revenue that was
not allowed to be amortized in the new decoupling rate is carried forward. Table 4-6 shows the
annual history of rate cap results for each fuel and rate group.
Exhibit No. 1
Page 4-8 P. Ehrbar, Avista
Page 158 of 224
Table 4-6. History of Rate Cap Results - Was Rate Cap Reached?
Electric Natural Gas
Deferral Year Residential Non-Residential Residential Non-Residential
2015 Yes No Yes Yes
2016 No No Yes No
2017 No No No No
On the electric side, the 3o/o cap on annual rate increases from the decoupling rate was only
reached one out of six possible times. After reaching the rate cap based on 2015 results, the
electric residential rate group did not reach the rate cap in20l6 and20l7. For natural gas, the
rate cap was reached 3 of 6 times, twice for residential customers and once for non-residential.
Electric non-residential is the only rate group that has not reached the rate cap. None of the four
rate groups were subject to the decoupling rate cap in20l7, meaning there were no unamortized
revenue balances to carry forward to 2018.
Review of Deferrals
The question as stated in the RFP is:
"V[rhat was the impact of the Decoupling deferuol on Avista's revenues and rates? "
"Wat was the ffict of updates to the decoupling baseline and resulting fficts on
deferrals under the mechanisms? "
As reported earlier in this section, deferrals have had the effect of lowering the variability of
annual revenue. This is true for all rate groups. Allowed electric revenue (revenue with
deferrals), base rate revenue and decoupling defenals are shown in Table 4-7.
Table 4-7. Electric Revenue.from Decoupled Rate Groups
Decoupled Year
Revenue with Deferrals Base Rate Revenue Decoupling Deferrals
Residential
Non-
Residential Total Residential
Non-
Residential Total Residential
Non-
Residential Total
(millions of doilari (millions of dollars)(millions of dollars)
2015 217.2 2t3.8 431.0 210.0 216.2 426.2 7.2 Q.4\4.8
2016 213.9 213.1 427.0 203.6 2tt.t 414.8 10.3 2.0 12.3
20t7 220.0 214.9 434.9 222.1 213.2 435.3 (2.1)1.7 (0.4)
Mean 217.0 213.9 431.0 211.9 213.5 425.4 5.1 0.4 5.6
Std Dev 3.0 0.9 3.9 9.4 2.5 10.3 6.4 2.4 6.3
Coefficient of
Variation 0.014 0.004 0.009 0.044 0.012 0.024 NA NA NA
The calculations for the coefficient of variation, a measure of variability, are also shown in Table
4-7. Allowed natural gas revenue (revenue with deferrals), base rate revenue and decoupling
deferrals are shown in Table 4-8.
Exhibit No. 1
6
Page 4-9 ffi_0_
P. Ehrbar, Avista
Page 159 ot 224
Table 4-8. Natural Gas Revenuefrom Decoupled Rate Groups
Decoupled Year
Revenue with Deferrals Base Rate Revenue Decoupling Deferrals
Residential
Non-
Residential Total Residential
Non-
Residential Total Residential
Non-
Residential Total
(millions of dollan)fuillions of dollars)(millions of dollars)
2015 108.s 36.9 r45.4 103.2 35.2 138.3 5.3 t.7 7.0
2016 112.3 35.7 148.0 105.1 JJ. /138.8 7.2 2.0 9.2
2017 121.5 38.8 160.3 123.5 38.0 161.5 (2.0)0.8 0.1)
Mean 1 l4.l 37.r 151.2 I10.6 35.6 146.2 3.5 1.5 5.0
Std Dev 6.7 1.6 8.0 11.2 2.2 13.2 4.8 0.6 5.4
Coefficient of
Variation 0.059 0.043 0.053 0.101 0.061 0.090 NA NA NA
Because deferred revenue has averaged above zero for all rate groups, deferrals have worked to
increase revenue from base rates. As has been discussed, much of the increase has been due to
lower use per customer due to weather, especially in electric residential, natural gas residential
and natural gas non-residential. Avista's energy efficiency programs have also worked to lower
use per customer, especially for electric rate groups. Going forward, weather could just as easily
have the opposite effect causing negative deferrals and higher base rate revenue than revenue
with deferrals. The same is not true for Avista's energy eff,rciency savings, which always work
in the direction of lower use per customer and increasing deferred revenues. The impact of
energy efficiency has been especially significant in explaining changes from test year
assumptions in the electric non-residential group.
Deferral balances and decoupling rates are shown in Table 4-9.
Table 4-9. Summary of Deferral Balances ond Decoupling Recovery Rotes
Electric
Residential Group Non-Residential
Notes 2015 2016 2017 2015 2016 2017
Deferred Revenue ($)7,167 .7 48 10.288.205 -2.092.790 -2.373.472 1.967.777 1.73s.91I
Requested Recovery ($)A 7,360,678 10,91 3,9s0 -2.76s.635 -3,081,249 864,012 1,170,966
Customer Surcharse (Rebate) Revenue ($)6.48s.021 10.913.950 -2.765.635 -3,081,249 864,012 1,170,966
Carryover Deferred Revenue ($)875.657 0 0 0 0 0
Decouoline Rate (Schedule 75) (S/kWhI B 0.00263 0.0044s -0.001l6 -0.00143 0.00040 0.000s4
Incremental Revenue (Percent)3.00%2.00%-5.78%-1.40%0.40%0.14%
Limited by 3o/o Cap?Yes No No No No No
Natural Gas
Residential Group Non-Residential Group
Notes 20ls 20t6 2017 2015 2016 2017
Deferred Revenue ($)5.3 I 7.198 7.ts2.977 -t.972.082 1.736.736 2_002.6s4 840.286
Requested Recovery ($)A 5,750.096 7.652.369 -3.44 I.586 1.879.152 2.212.881 407.7 t9
Customer Surcharge (Rebate) Revenue ($)3,488,984 6,951,431 -3,441,586 I, I 08,839 2,212,88t 407,719
Carryover Deferred Revenue ($)2.261.t12 700,938 0 770,313 0 0
Decoupling Rate (Schedule 175) (S/therm)B 0.02927 0.05580 -0.02720 0.021 08 0.03904 0.0069 I
Incremental Revenue (Percent)3.00%3.00%-10.08%3.00%2.95%-6.13%
Limited by 3%o Cao'l Yes Yes No Yes No No
A: Requested recovery is equal to deferred revenue after adjusting for shared excess eamings (if applicable), deferral balance carryover
from prior year (ifany). interest, and revenue related expenses.
B: DecouplingratesScheduleT5(electric)andSchedulelT5(naturalgas)takeeffectonNovemberlstofthefollowingyear. Forexample,
rates shown in the 2016 column have an effective date ofNovember 1,2017
Exhibit No. '1
0
Page 4-10 UASE NOS. AVU-ts-]9.U ANO AVU.U-I9-U
P. Ehrbar, Avista
Page 160 ot 224
6
Comparing deferred revenue with the requested recovery shows the importance of deferral
balances in determining decoupling rates. They are not the only factor, however, and in some
instances other factors are actually larger than the deferral balance. This was the case for electric
non-residential in 2016, for example, when the requested recovery was only 44 percent of
deferred revenue ($ 864,012 I $ 1,967,777), due mainly to shared excess eamings.
The decoupling baseline or test year is another factor that comes into play when analyzing
deferral balances and the impacts from various factors, such as energy efficiency. The test year
used for 2015 deferral calculations was a projection of 2015. The test year for 2016 and 2017 is
a 12-month period ending September 2014. The practical implication of this change in baseline
for acfual weather compared to normal weather are insignificant. However, Avista's energy
efficiency programs have a gteater impact the further in time the actual calendar year is from the
test year. So, moving the baseline from 2015 to 12 months ending September 2014 resulted in a
larger variance in use per customer due to Avista's energy efficiency programs. The same is true
for the "other" category of factors impacting use per customer which would include efficiency
gains outside of Avista's programs.
Removing the influence of weather from deferred revenue provides another way to view the
impacts of Avista's energy efficiency achievements and "other" unexplained influences on
deferral balances. Table 4-10 shows actual deferred revenue and deferred revenue estimated at
normal weather.
Table 4- 10. Deferred Revenue at llormal Ll/eatlter
Electric
Residential Group Non-Residential Groun
20ts 2016 2017 201s 20r6 2017
Deferred Revenue 7,t67,748 70,288,20s -2,O92,790 -2,373,472 t,967.777 1.735.91I
Weather Impact on Deferrals 2,4L6,743 5,547,227 -8,618,230 -451.215 465.250 -1.646.26s
Deferred Revenue at Normal Weather 4,75L,O05 4,700,978 6,525,440 -1,922,257 1,502,527 3,382,176
Natural Gas
Residential Group Non-Residential Group
2015 2016 2017 2015 2016 2017
Deferred Revenue 5,317,198 7,1s2,977 -1,972,082 1,736,736 2.002.654 840,286
Weather Impact on Deferrals s.739.128 4,'720,02t -t,961,267 1,262,997 967,162 -380,599
Deferred Revenue at Normal Weather -421.930 2.432.956 -10,815 473,739 t,035,492 1,220,885
Deferred revenue at normal weather is calculated by subtracting the weather impacts on deferrals
from actual deferred revenue. The weather impact is estimated using Avista's weather
adjustment coefficients as reported in weather adjustment calculations workbooks.66 Deferred
revenue at normal weather shows the same patterns of influence of Avista's energy efficiency
programs and other unidentified factors on deferred revenue. Consider, for example, the electric
non-residential rate group. Deferred revenue estimated at normal weather was negative in 2015
and increasingly positive in 2016 and20l7. This is the same pattern shown in Figure 4-4 where
the net influence of Avista's energy efficiency programs and other factors excluding weather
Exhibit No. 1
66 See Data Request number 76.
Page 4-l I uase Nos. AVU-ts-l9-u and AVU-G-I9-U
P. Ehrbar, Avista
Page161 ot224
o
lead to higher use per customer in 2015 (and negative deferrals) and progressively lower use per
customer (and positive deferrals) in20l6 and2017.
Summary - Task 4
Avista's decoupling mechanism has had a stabilizing effect on revenue, reducing variability to
between 30 and 70 percent of variability without decoupling. On the electric side, the 3o/o cap on
annual rate increases from the decoupling rate was only reached one out of six possible times
when it came into effect for electric residential in 2015. For natural gas, the rate cap was reached
3 of 6 times, twice for residential customers and once for non-residential. Electric non-
residential is the only rate group that has not reached the rate cap. None of thefour rate groups
were subject to the decoupling rate cap in 2017.
Because deferred revenue has averaged above zero for all rate groups, deferrals have worked to
increase revenue from base rates. Much of the increase has been due to lower use per customer
due to weather, especially in electric residential, natural gas residential and natural gas non-
residential. Avista's energy efficiency programs have also worked to lower use per customer,
especially for electric rate groups. The impact of energy efficiency has been especially
significant in explaining changes from test year assumptions in the electric non-residential group.
Exhibit No. 1
P. Ehrbar, Avista
Page 162 of 224
Page 4-12
Section 5. Fixed Costs and Cha s, Non-Decou Ied Customers
In this section we examine fixed costs and fixed charges for electric and natural gas customer
classes.
The objective of Task 5, as stated in the request for proposal, is shown below:
"Analysis of the extent to which fixed costs are recovered in fixed charges for the
cus tomer clas s es, excluded from the Mechanisms. "
Relating to this objective is the following evaluation question, also taken from the RFP:
"How much of the Company's fixed costs recovered from non-decoupling customer
classes are recovered infixed charges?"
The scope of this section was expanded to include decoupled electric and natural gas customer
classes to facilitate comparison to customer classes excluded from the decoupling mechanisms.
To address the evaluation objective, it is necessary to compare revenues from fixed charges to
fixed costs for these customer classes. Fixed cost and revenue collected from fixed charges was
provided by Avista in response to data request (DR) 89. Beginning with electric customer
classes, we examine the recovery of fixed cost through fixed charges and the relationships
presented in the data. Throughout the discussion it is useful to keep in mind that the basis for
cost allocation changed between 2015 and 201612017.67
Electric Customers
Annual revenue from fixed charges and fixed costs are shown for elechic customer classes in
Table 5-1.
Table 5-1. Electric Reverutefrom Fixed Charges and Fixed Cost (thousands o/'dollars)
Total
Decoupled Non-Decoupled
Residential
General
Service
Large
General
Service
Pumping
Service
Extra Large
General
Service
Street & Area
Lishtins
Schedules
1,2 tl, 12 21,22 31,32 ,(4t-49
Revenue from Fixed Charses 52.730 21.450 6.728 12.06t 527 5.292 6.672
Fixed Cost 382,17 191,696 43.845 86.254 9.376 43.585 7.360
Percent Recovered from Fixed Charges t3.8%tr.2%153%14.0%5.6%12.1%90.7%
Revenue from Fixed Charges 52,944 21,969 6,883 11,447 546 5,271 6,828
Fixed Cost 400,66t 202,356 47,747 87,775 9,116 45,439 8,23s
Percent Recovered from Fixed Charges 13.201 10.9%14.4%13.0%6.0%ll.60/o 82.9%
2017
Revenue from Fixed Charges 53.013 22-226 6,95s I 1,396 s33 s,426 6,475
Fixed Cost 408.126 210.268 48,363 86,777 9,144 45,287 8,286
Percent Recovered from Fixed Charges 13.0%10.6%t4.4%13.1%5.8%12.0%78.1o/o
67 For 201 5 the cost of service study used for the General Rate Case (GRC) for electric (UE- 140 I 88) and natural gas
(UG- 140 I 89) was the basis for cost allocation factors. The cost of service study used for the GRC for electric (UE-
150204) and natural gas (UG-150205) was the basis for cost allocation factors used for 2016 and 2017. These cost
allocation factors were adjusted for actual customer counts and usage levels for the analysis reported in this section.
Exhibit No. 1
Page 5-l Case Nos. AVU-E-19-0 and AVU-G-19-0
P. Ehrbar, Avista
Page 163 of 224
O
6
Over the 2015-2017 period fixed charges for total electric have averaged slightly higher than l3
percent of fixed cost. The percentage has fallen slightly since 2015. The customer class that
covers the highest percentage offixed costs through fixed charges is street and area lighting,just
over 90 percent in 2015 and falling to 78 percent in 2017 . The customer class collecting the
smallest percentage of fixed costs through fixed charges is pumping services. Pumping services
have averaged a little less than 6 percent recovery offixed cost through fixed charges. About I 1
percent ofresidential fixed costs are recovered through fixed charges. The percentage has fallen
from I I .2 percent in 2015 to 1 0.6 percent in 2017 . The percentage of fixed cost recovered
through fixed charges from general services and large general services have each fallen about
one percentage point between 2015 and the two-year period 2016 and2017.
Natural Gas Customers
Annual revenue from fixed charges and fixed costs are shown for nafural gas customer classes in
Table 5-2.68
Table 5-2. Fixed Cost and Fixed Charges, Non-Decoupled Natural Gas Customer Classes
Decoupled Non-Decoupled
Total
Residential
Large
General
Service
High Load
Factor Large
General Service
Interrupt-
ible
Service
Transportation
Service
Schedules
101. 102 ttt. tt2 tzl. t22 13l. 132 146
Revenue from Fixed Charges 19.5 l9 16.471 2.748 70 0 229
Fixed Cost 81.40s 61 5q1 13,6s2 1,166 173 2,822
Percent Recovered from Fixed Charges 24.0%25.9%20.1%6.0o/o 0.0%8.1%
Revenue from Fixed Charges 20,544 16,896 3,324 78 0 245
Fixed Cost 84,923 69,266 11,542 818 184 3.1 l3
Percent Recovered from Fixed Charges 24.2%24.4%28.8%9.6%0.0o/o 7.9%
20 t7 ------- ----- --- - ---
Revenue from Fixed Charses 21.184 17.287 3.s36 108 0 253
Fixed Cost 89.681 72.938 12.464 793 179 3.308
Percent Recovered from Fixed Charses 23.6%23.7%28.4%13.7%0.0%7.6%
Over the 2015-2017 period fixed charges for total natural gas have averaged around 24 percent
of fixed cost. Residential customers cover the highest percentage of fixed costs through fixed
charges, ranging from a high of 25.9 percent in 2015 to a low of 23.7 in20l7. The two non-
residential decoupled customer classes have both seen a marked increase in the percentage of
fixed costs recovered through fixed charges since 201 5. This sort of change is likely due to rate
restructuring between the UG-140189 and UG-150205.
Non-decoupled customer classes recover the smallest percentage of fixed cost through fixed
charges. Fixed charges revenue as a percentage of fixed cost is zero for intemrptible services.
68 Avista's natural gas cost of service studies use different customer groupings than the decoupling mechanism. The
cost ofservice roll-up does not differentiate Schedules ll2,122, or 132 which were excluded from the decoupling
mechanism. Consequently the I I l/l 12 and l2lll22 columns overstate the decoupled amounts and the Schedule
l3l/132 column understates non-decoupled sales service. The difference is not considered material for the cost of
service portion of this evaluation.
Exhibit No. 1
Page 5-2 CaseNo-TVU:E:ag-U_ and AW-G-1 9-0_
P. Ehrbar, Avista
Page 164 of 224
6
Fixed costs are a very small level of the total costs for this customer class. The percentage of
fixed cost recovered through fixed charges from transportation service has averaged a little less
than 8 percent and has been falling between 2015 and20l7.
Summary - Task 5
Avista recovers about l3 percent of total electric fixed cost through fixed customer charges,
trending only slightly lower over the 2015-2017 period. On the natural gas side, fixed charges
have averaged nearly 24 percent of fixed costs between2015 and2017. The percentage has
moved higher for decoupled natural gas non-residential customer classes and lower for
residential.
Exhibit No. 1
uase Nos. Avu-E-l9-u_ and AVU-G-I9-u_
P. Ehrbar, Avista
Page 165 of 224
Page 5-3
6
Exhibit No. 1
case Nos. AVU-E-19-0_ and AVU-G-I9-0_
P. Ehrbar, Avista
Page 166 of 224
Page 5-4
6
In this section of the evaluation, we use results of DSM Annual Conservation Report & Cost
Effectiveness Analyses and the Annual Conservation Plans. There are three questions (Figure
6-1):
o First, what is the impact of conversions from electric to natural gas on decoupling
revenue?
. Second, has decoupling had an impact on nafural gas conservation savings?
o Third, has decoupling had an impact on electric conservation savings (leaving out the
commitment to an additional five-percent (5%) energy saving)?
Conservation achievement through regional market transformation (which Avista co-funds
through the Northwest Energy Efficiency Alliance) is left out of all analysis in this section of the
report.
First, we examine the impact of fuel conversions on decoupling revenue. Then we examine
whether decoupling has had an impact on energy savings.
Task 6: An analysis of each Mechanism's impact on conservation achievement, in total and
by sector (residential, low-income, non-residential), and identification of conclusive or
meaningful trends in the performance of the Company's electric and natural gas conservation
programs since the inception of the Mechanisms (did the Company achieve a higher level of
savings with the mechanisms in effect). This analysis should be based on information
already available as part of the Company's biennial conservation achievement evaluations
filed with the Commission including changes to program delivery strategies as reported in
annual evaluations, significant changes in program budgets, or reported savings levels.
6a For the electric and gas conservation programs, what impact has the shift in customers
(electric to natural gas) due to fuel conversions had on decoupling revenue?
6b Have the Mechanisms had an impact on natural gas conservation savings?
6c Have the Mechanisms had an impact on electric conservation savings (not including the
decoupling commitment to an additional5o/o savings)?
Figure 6-1. Task 6 - Conservotion Achievement
Exhibit No. 1
uase Nos. AVU-ts-] 9-U_ and AVU-G-I 9-U_
P. Ehrbar, Avista
Page167 of224
Section 6. Ana is of Conservation Achievement
Page 6-1
6
What is the Impact of Fuel Conversion on Decoupling Revenue?
Evaluation question 6a (Figure 6-1) asks, "For the electric and gas conservation programs, what
impact has the shift in customers (electric to natural gas) due to fuel conversions had on
decoupling revenue?" The goal is to decrease electric usage by increasing sales ofnatural gas.
First, three observations to set the context:
o For 2015, there was no decoupling revenue, so there was no fuel conversion effect on
decoupling revenue for 2015.
o For 2016, decoupling revenue was limited to a partial collection (or rebate) of revenue
from decoupling in November, phased in using billing cycles and full collection in
December. For January through October of 2016 there was no decoupling revenue
recovery or rebate. This means any effect demonstrated for calendar 2016 is quite small.
o For 2017,there is a full year of application of the decoupling adjustment to customer
bills. This means calendar 2017 is the first year to show the full effect of decoupling
revenue recovery and/or rebate.
For the fuel conversion program, change is directional. Fuel conversion operates in only one
way, from electric to natural gas. Conversion begins by disconnection of electric end-use
equipment, so analysis begins on the electric side. From an electric perspective, yearly kWh
conversion savings as a percentage of overall savings achievement by group is shown for
Electric Residential in Table 6-1 and for Electric Nonresidential in Table 6-2:6e From an electric
perspective,
o the residential percentage converted for2016 is about 23 percent ofresidential overall
savings achievement; for 2017 it is just under 31 percent (Table 6-1).
. and for low-income, conversion is about 50 percent of overall savings achievement for
2016;for2017 it is 73 percent (Table 6-1).
o low-income converted kWh is just under 3 percent of residential converted kWh in 2016;
for 2017 it is 5 percent (calculated from "Converted kWh" columns in Table 6-1).
o the non-residential percentage of overall conservation achievement due to conversions is
2.1 percent for 2016; for 2017 it is 2.6 percent (Table 6-2).
6e 2016 Washington DSM Annual Conservation Report & Cost-Effective Analysis, June l, 2017,Table ES-l; 2017
Washington DSM Annual Conservation Report & Cost-Effective Analysis, June 1,2018, Table ES-1.
Exhibit No. 1
Page 6-2 Case Iloe AVU:F19-0 and AVU-G-19-0
P. Ehrbar, Avista
Page 168 ot 224
0
Table 6- I . Electric Residential Conversions as Percentage of Conservation Achievement (kwh)
Electric Residential kWh (Including Low Income)
Decoupled Group
2016 2017
kwh Converted kWh Percentage
Converted kwh Converted kWh Percentage
Converted
Residential 43,083,55 I 9,766,qss 33,376,237 t0,237,036 30.7%
Low Income s46,066 273,628 50.lYo 710,204 518,748 73.0%
Total 43,629,617 10,040,483 23.0%34,086,441 10,755,784 3t.6%
Table 6-2. Electric Non-Residential Conversion as Percentage of Conservation Achievement (kwh)
Electric Nonresidential kWh
Decoupled Group
2016 2017
kwh Converted kWh Percentage
Converted kwh Converted kWh Percentage
Converted
Nonresidential 38,226,357 805,779 2.1%4l,93 0,099 1,070,262 2.60/o
70 Electric usage (kWh per month) is shown in Section 2 of this evaluation. No allocation is perfect and other
allocations could also be used. Reporting of conserved kWh is typically on a first-year projected basis for projects
completed during a calendar year. Converted kWh is treated on the same basis for allocations to table columns and
estimation of the total.
Exhibit No. '1
Page 6-3 P. Ehrbar, Avista
Page 169 ol 224
Residential Electric: The Schedule 75 Residential electric decoupling rate (from Task 2) is
$0.00263 per kwh for the first rate-year and $0.00445 for the second rate-year (in the case of the
Residential Electric group, for both years, these are surcharges to customers). Since the specially
defined year for application of rates runs for the twelve months from November through
October, the electric decoupling rate for the 2016 cannot be used for the full calendar 2016 (the
value is zero for January through October of 20t6, then $0.00263 per kWh for November and
December 2016).
This value also applies for January through October 2017. The value for the second rate-year
applies to the last two months (November and December) of 2017. Converted kWh is taken
from the Washington 2017 DSM Annual Conservation Report & Cost-Effectiveness Analysis
(see note in Table 6-3). The ratio of Residential kWh usage per time block as shown in the
columns of Table 6-3 is developed from monthly energy use, summed over each time block and
then divided by the total Residential energy use. This ratio is used to spread the application of
the Converted kwh.70 As shown in the last row and final column of Table 6-3, conversion from
electric to natural gas is estimated to cause $29,389 of fixed cost Electric Decoupling Rate
Adjustment over November 2016 through December 2017.
6
Dollar values in the columns result from the application of electric decoupling rate values. The
value shown,$29,389, is an estimate. This estimate is determined in part by the allocation of full
year Converted kWh savings by time blocks to which the different values of Electric Decoupling
Rate Adjustment apply.Tr
The values in the row next to the bottom row of the table are the Electric Decoupling Rate
Adjustment surcharges whichwould have applied if there had been no conversion and equipment
had remained in place connected to the electric system. Since these devices were disconnected
from the system, customers retained a net value of $29,389. From a Company perspective this
represents a net loss of $29,389 of Electric Residential Decoupling Revenue.
Table 6-3. Residential Fuel Conversion Progrant Savings
Allocation of Residential Electric Decoupling Revenue Based on Gross Verffied Savings ( kWh)
Converted kWh
t6-17 Jan-Oct 16 Nov+Dec l6 Jan-Oct 17 Nov+Dec 17 Total
Converted kWh 16-17 16,541,961
Residential kWh 1.852.650.051 416,441,453 2.062.581.745 439.889.023 4.771.562272
Weights 0.3883 0.0873 0.4323 0.0922 1.0000
Allocated Conve rted kWh 16.541.961 6.422.732 rA43,711 7.1 50.5 l9 1.524.999 16,541,961
Decoupling Rate 0 0.00263 0.00263 0.00445
Decouplinq Revenue 93.797 $18.806 $6.786 $29,389
Surcharge Surcharge Surcharse Surcharse
The Converted kWh 16,l7 is from Page 2,Tabb.2: Washington Electric Portfolio Evahration Resuhs, Appendix C (201G2017 Electric Impact
Evaluations) of the Washington 2017 DSM Annual Conservalion Report & Cost Effectiveness Analysb, June 1,2018. Electric Residenthland
Electric Residenthl Low Income conversiors have been combined.
For Residential Electric, each time block in which the decoupling rate is applied represents a
surcharge to the customers and would have been paid by the customers if they had not
disconnected equipment from electric service. However, here we examine the loss of kWh sales
due to conversion away from Electric Residential service, so the surcharges that would have
been paid by customers are not paid, representing a gain for the customers and a loss of Electric
Residential decoupling revenue to the Company.T2 The Company would eventually recover this
revenue through future decoupling rate adjustments and surcharges so that the net effect is a
transfer of income from all customers within the rate group to those customers that convert from
electric to natural gas.
Nonresidential Electric: The Schedule 75 Nonresidential Group electric decoupling rate is
negative $0.00144 for the first rate-year (a rebate to customers) and positive $0.00040 (a
surcharge to customers) for the second rate year. Since the specially defined year for application
of rates runs from November through October, the decoupling rate for the 2016 (negative
7l Results are not directly metered, they are modelled using assumptions.
72 We treat conversion here the same way that conservation is treated. If conservation occurs during a calendar year,
its value for that year is counted ("first year energy savings"). Generally, this is a modeled value based on a
combination of empirical measurement and engineering analysis and assumptions. It is not the actual metered value
for that year, except in special projects with quasi-experimental or experimental designs.
Exhibit No. 1
Page 6-4 P. Ehrbar, Avista
Page 170 ot 224
6
$0.00144) can be used for the two months to which it applies in calendar 2016. This is also the
rate for January through October 2017. The positive value of $0.00040 then applies for
November and December of 2017.
Results for Nonresidential Electric conversions are shown in Table 6-4. The values in the table
represent what would have happened if the electric equipment was not disconnected. In this
instance, a negative decoupling rate for the first rate-year (a rebate) has a much larger effect than
the customer surcharge Electric Decoupling Rate Adjustment for the two months to which it
applies in the second rate-year. The net result is a rebate of $ I 1,807. However, in fact, the
customers did disconnect the electric equipment. So, on the electric side their net rebate was
foregone and can be treated as a cost of $l1,807. This means the Company gained $11,807 by
not having to pay out this amount in rebates to customers.
Table 6-4. Allocation of Nonresidential Revenue based on Gross Verified Savings (kwh)
Allocation ofNonresidential Electric Decoupling Revenue Based on Gross Verified Savings ( kWh)
CoDverted kwh
t6-17 Jan-Oct l6 Nov+D€c I 6 Jan-Oct l7 Nov+Dec l7 Tolal
Conveded kwh l6-17 t.8t0.107
1.791.155.8U 345379.13(I -820.084. I 74 259943241 4223.162.961
We ishts 0.425i 0.0818 0.43tc 0.0616 I O0fit
Allocated Converted kwh t.8r 0.r07 '1.M1 73t tJ52,839 7,129,191 l0l8,r 87 t6,54t,961
Decoupling Rste (-0 0014/-0.00t44 0.00040
-$1,948 -$r0266 $407 -$l t,80?
Rebab Rebate Smharge Rebate
The Convefred kwh l&17 is fM Page 2, Tabb 2: Washhgton Electra Podolb Evahatbn Resuhs, Appendix C (201G2017 Elecric Impdct
Evaluatbns) of the Washington 2017 DSM AmulCoNeryatbn Repon & C6t Effectiveress Amlysb, Jw 1,2018. NotresllentblCroups E2A
(CEneml Senices)), E2B (Large C*real Seruices Included in Decoiupling) and E2C (Pmping) conversions have been combined.
Residential Natural Gas: By means of similar calculations, the new sales of therms effect on
Residential Natural Gas decoupling revenue from increased gas sales is shown in the boffom two
rows of Table 6-5. Here the magnitude of the change is only $1,079. Since natural gas sales
were increased and the Natural Gas Residential Group Decoupling Rate Adjustment is on a per
therm sold basis, the Natural Gas Residential Group received an additional cost of $1,079. From
a Company perspective this is $1,079 in additional Residential Natural Gas Decoupling Revenue.
Table 6-5. Residential Gos Decoupling Revenue Based on Gross Verified Savings (tlterms)
Allocation ofResidential Natural Gas Decoupling Revenue (Gross Verified Savings - Therms)
CotrveEiotr
IncrcNed Sales
(2016-20r7)
fTh€ ms)
Jan-Oct l6 Nov+Dec l6 Jen-Oct l7 Nov+Dec 17 Total
Cotrverted Thems l6-17 1,116.582
Residential Thems 19954 265 25 g6t m6 1n,72t248 30,034J33 328,671,052
Weiphts 0.2433 0.0790 0.5864 0.09t 4 I ffno
Allocated Converted Thems l.t 36-582 276.491 2 l -8:19 12.806 1.t70 3 12.306
DecouDlitrg Rate 0 0.02927 0.02927 0.05580
Decouplinq Revenue $639 $375 $65 $1.079
SurcharEe SurcharEe Surcharse SurcharRe
The C@verted Thems 1617 is frcm Page 20, Table 2-15: ResHentbl Reported Panicimtbn and Savhgs, Appendix D (20162017 NamlCas
lmpact Euhtsth) of the Washingtm 2017 DSM AnnulCmeilatbn Repon & C6t Effectiveress Analysb, Jme 1,2018. NatuBl Cas
Resilential and Natual CBs Low Ircw cqversixs have been cmbined.
Exhibit No. 1
P. Ehrbar, Avista
Page 171 ol 224
Page 6-5
lnrcsi.lential kWh
Dec^nnlino R.vcnn.
0
Nonresidential Natural Gas: By means of similar calculations, the converted portions of
decoupling revenue for residential natural gas is shown in the bottom two rows of Table 6-6.
Here the magnitude of the change is $1,384. Since there is a per therm surcharge for additional
natural gas sales for the Nonresidential Natural Gas Group, this results in an additional
Nonresidential Decoupling Revenue Charge of $1,384. This is a cost to customers of $1,384 and
from a Company perspective an increase in Nonresidential Natural Gas Decoupling Revenue of
$1,384.
Table 6-6. Nonresidential Gas Decoupling Revenue (Gross Verified Savings - thernts)
Allocation ofNonresidential \atural Ges Decoupliog Revenue (Gross \:erifed Sarings - Therns)
Cmr erled Ttems 16-17
.{ bcrl.d CoDr.rl.d Tt.ru
ReraDua
lhe Convened Therru 16t? is from PaBe 19. Table 2-l4rAvista Nonrcsidential Reported Parnopatiqr ad Savin8s, Appeodir 0 (201620U
NatuclG3s lmprlEvduaion)of theWashin8too20lTDSMAnnualConservationneport&CostEff?divef,e$Anallsi',.,une1,20E. N.tuEl
Gas G.orral Seilic. and Natucl Gas LrrB? C*n.ral SeME locluded (Deoupled Groupr G2A & GzB) haw b..n ombin?d
Summary - Impact of Fuel Conversion on Decoupling Revenue
The impact of fuel conversion on decoupling revenue is small.
o For residential customers, there was a decoupling tariff adjustment of (cost to the
customer) $29,389 for disconnecting electric service to equipment. Adding natural gas
equipment as replacements, additional natural gas sales caused a decoupling tariff
adjustment (cost to the customer) of $ 1,079, for a net cost to customers of $29,868. At
the same time, net residential effect on Residential (combined) Electric and Natural Gas
Decoupling Revenue was a gain of $29,868 to the Company.
o For Nonresidential customers, there was a decoupling tariff adjustment (cost to the
customer) of $11,807 which would have been rebated had equipment remained in place.
In addition, there was a cost to the customer of $ 1 ,3 84 for the decoupling cost adder per
therm for additional therm sales, for a net cost of $13,191. From the Company's
perspective, this is a gain of $ I 3, I 9 I in Nonresidential (combined) Electric and Natural
Gas Decoupling Revenue benefit.
Exhibit No. 1
Corv?Eloo
hcErscd Srli
(2016-2017)
aTt.rsl
J.uocl l6 \6 +Dcc 15 JmOd 17 Sor+D€c l7 Totd
ss.08s
i6i9l.05t lo5t0.l6r {.t:77:80 il.E8{.119 l0l lt86n
6 i<:{o J:gi 0 11<:l a00c0. r0:5
ss oss -il 089 90$3is:0 l0,l5t
!p Rrta 0 0 o]os 0 0llos 0.0i90{
si9(sl_i8lsl90s:9r
Srcharqe Smhaee Smhaee slch!F
Page 6-6 -P. Ehrbar, Avista
Page '172 ot 224
Tbrrilt
0
Have the Mechanisms had an Impact on Natural Gas or Electric
Conservation?
This question combines evaluation analysis questions 6b and 6c in Figure 6-1. For electric
conservation savings, the decoupling commitment to an additional f,rve-percent(5%) savings is
excluded from analysis: the question concerns conservation beyond the five-percent decoupling
commitment. We first look at conservation savings totals. The look at totals (electric and
natural gas separately) is followed by examination of conservation savings for each of the three
electric decoupling groups (residential, low-income, non-residential). Then, we examine each of
the three natural gas decoupling groups (residential, low-income and non-residential). In each of
these analyses, we conclude there is no evidence that the decoupling mechanisms hod an impact
as a driver (either positive or negative) on Conservation Achievement. However, wefind that
decoupling is important in removing baruiers to Conservation Achievement.T3
D eco upling snd C ons erv ation Achievement (Totals) : Perspective
Electric conservation is primarily influenced by the I-937 Energy Independence Act74, rather
than by decoupling, but decoupling does have an important role in removing barriers to
Conservation Achievement. The role of the decoupling factor is to eliminate a financial
disincentive so that other factors may operate as drivers; but it does not drive conservation
programs.
Beyond the current I-937 Energy Independence Act conservation effort, Avista is a national
leader in Smart CitiesTs, Distributed Energy Resources (DER) and microgrid developmentT6.
These are major efforts that go beyond decoupling. The future is likely a combination (yet
unnamed) of DSM, DERs, Smart Cities, an ecology of microgrids and nanogrids, and likely also
integrates elements of climate adaptation.
Business Plannine: Electric and Natural Gas 2012-2017
For perspective, we drop back in time, prior to the current decoupling because Avista has a deep
history in DSM planning. For example, the Business Plan for 201277 notes that "Avista has
73 In response to DR 94, Avista states: "With or without Decoupling, Avista will make any necessary investments
required in order to ensure high quality service for our customers. That said, decoupling positively effects how
Avista now looks upon proliferation of distributed generation (net metering) in our system. Without decoupling, it is
entirely reasonable to think from a regulatory and policy position, Avista would seek to minimize the amount of net
metering on our system. With decoupling, that is not the case, similar to the goal of decoupling to remove any
disincentive towards promoting energy conservation/efficiency." We do not disagree with this statement; however,
we think that decoupling is important in removing barriers to Conservation Achievement.
7a Washington,l-937:- Utilities must pursue all conservation that is cost-effective, reliable and feasible. They need to
identify the conservation potential over a 10-year period and set two-year targets. See:
httD://rvu'r.v.coulnerce.wa.gov/grou'ing-the-ecortomv/energ-v/ener&v-independence-acti. In response to DR 093,
Avista states: "Avista does not feel that decoupling is a driver nor a barrier removal mechanism on conservation
achievement. Given the requirements under the Energy Independence Act (EIA/I-937) to pursue all cost-effective,
reliable, and feasible savings, that is the primary driver of conservation achievement."
75 Data Response 043 (University District Smart City Accelerator Initiative).
76 Data Response 044 (Micro-Transactive Grid). Avista has also done earlier work with microgrids and is viewed in
the industry as a leader.
77 Response to Data Request 016 (Annual DSM Plans), 2012 DSM "Revised" Business Plan, Avista Utilities,
Revised December 7 ,2011 .
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-I9-o_
P. Ehrbar, Avista
Page 173 of 224
Page 6-7
0
continually been providing energy efficiency programs, unintemrpted, since November I't,
1978." That is forty (a0) years of DSM planning and program implementation if we count the
current year (2018).The2012 Business Plan goes on to say: "[t]he Company's planning process
builds on previous years experiences and addresses a number ofchallenges in regard to
achieving energy acquisition targets, meeting cost-effectiveness criteria and satisfuing regulatory
reporting requirements."T8 Avista has substantial depth in DSM planning and program operation,
as well as experience with evolving legislative and Commission targets that orient and drive the
DSM planning function. Against this deep background, decoupling affects the current context in
which conservation takes place. However, decoupling does not drive Conservation
Achievement. Rather, the annual DSM plans are technical documents informed by technical
concerns and directives to develop energy savings targets. The2012 through 2014 plans are not
influenced by decoupling (which began in 2015).
As part of the pending General Rate Case Settlement Agreement in Docket Nos.
UE-140188 and UG-140189, the Company agreed, in consideration for receiving a
full electric decoupling mechanism, to increase its electric energy conservation
achievement by 5% over the conservation target approved by the Commission,
beginning with the 2014-2015 biennial target. The scope of the DSM Business Plan
covers the majority of the acquisition eligible to achieve this target but does not
include efficiencies achieved through distribution or generation facilities. Since the
planning process has led to the expectation that the acquisition target will be
achieved, the Company has not designed, and is not currently considering any
contingency programs to increase acquisition to meet the target.
Figure 6-2. Planning.for Decoupling 5?(,
Beginning with the 2012 Business Plan (completed in2011) and moving forward, the first
mention of decoupling occurs in the Business Plan for 2015 (Figure 6-2)." Here, the electric
planning targets contain the five percent (5%) addu to DSM energy savings which is a part of
the decoupling order. Since the electric adder was already covered within the flexibility of the
planning process, no action was required to specifically further consider or address decoupling.
There is no indication of any other influence of decoupling on planning for conservation
achievement in the 2015 plan.
There are similar mentions of decoupling in the 2016 plan8o and the 2Ol7 electric plansr. While
each plan is a comprehensive document, usually of 150 or more pages, there are no further
78 Ibid., Executive Summary, P. 2.
7e Response to Data Request 016 (Annual DSM Plans), Avista Utilities Washington/Idaho 2015 Demand-Side
Management Business Plan, October 31,2014, P. 9. Of course, the addition of the five-percent (5%) itself is an
effect of decoupling. It was added and agreed to part of the decoupling agreement. At the policy/management
levels decoupling had this influence on the DSM Plan that drives conservation. However, Task 6 directs that this
addition to Conservation Achievement planning and accomplishment not be included in the analysis in this Section
of the evaluation. We note it here for completeness.
80 Response to Data Request 016 (Annual DSM Plans), Avista Utilities Washington/Idaho 2016 Demand-Side
Management Business Plan, October 26,2015, P7 & P, 8.
81 Response to Data Request 016 (Annual DSM Plans), Avista Utilities Washington/Idaho2017 Electric Demand-
Side Management Annual Conservation Plan, November 15,2016,P.6 &P.23.
Exhibit No. 1
Page 6-8 P. Ehrbar, Avista
Page 174 ol 224
6
substantive considerations of decoupling in any of the plans for 2015 throughz0l7. Similarly,
there are no mentions of decoupling in the plans from20l2 through 2017 for natural gas.
We concludefrom the anolysis of Business Plans and Evaluationsfor 2012 through 2017 that
decoupling had no independent effect on electric or natural gas planning beyond the 50% adder.
Next, we examine Conservation Achievement directly in the series of Avista evaluations.
Totol Conservation Achievement: Electric and Natural Gas: 2012-2017
To assess the role of decoupling in Conservation Achievement, we examine the Annual
Conservation Reports & Cost Effectiveness Analyses for Washington for 2012 throtgh2}I7.82
The Annual Conservation Reports & Cost Effectiveness Analyses report electric and natural gas
conservation achievement against planning target goals. The Biennial Conservation Plan (BCP)
for Washington's Energy Independence Act (Initiative 937 or I-937) provided energy savings
targets for 2014 through 201 5.
o In the2014-2015 Biennium, Avista acquired 70,959 MWh (verified gross savings) in
Washington or l04Yo of its two-year electric target of 68,204 MWh.83 The five-percent
(5%) decoupling adder did not apply in this Biennium.o In 2016, Avista acquired 71,572 MWh (I-937 total adjusted reported gross savings) in
Washington, or l3Ooh percent of its I-937 target of 54,978 MWh.84 The five-percent (5%)
decoupling adder is included.o In 2016-17, Avista acquired 139,450 MWh (total verified gross savings) in Washington,
or 183% percent of its I-937 target of 141,331 MWh. The five-percent(5Yo) decoupling
adder is included.8s
With exceptionally high achievement levels for 2015-2017,the five percent (5%) conservation
achievement for decoupling was easily surpassed.86 The Annual Conservation Reports & Cost
Effectiveness Analyses for 2012 through 2017 contain no further mention or analysis of
decoupling. There are no mentions of decoupling from 2012 throughz}l7 for natural gas. We
conclude from the analysis of the Annual Conservation & Cost Effectiveness Reports for 2012
through 2017 that at the level of total achievement, decoupling had no independent effect on
driving overall electric conservation achievement. The substantial increase in performance for
residential, low-income and non-residential from 2015 to 2016 is attributed "...to the increasing
popularity of LED light, TLED lighting and Fuel Conversions."sT This finding was repeated in
the 2017 evaluation.ss
82 Responses to Data Requests 0 I 7, 0 I 8 and 070 (Annual Conservation Reports and Cost Effectiveness for 2012
through 2017).
83 Washington 2015 Annual Conservation Report (ACR) & Cost-Effectiveness Analysis, May 31, 2016,P.4.
8a Washington 2016 DSM Annual Conservation Report & Cost Effectiveness Analysis, June l, 2017,P. 18.
8s These results have been updated to comect an error using numbers provided verbally by Avista during the report
presentation/review meeting. Numbers reported here are slightly less than those in the Washington20l7 Annual
Conservation Report & Cost-Effectiveness Analysis, June l, 2018, Executive Summary, P. l.
86 Washington 2017 DSM Annual Conservation Report & Cost-Effectiveness Analysis, June l, 2018,P.17.
87 Washington 2016 DSM Annual Conservation Report & Cost-Effectiveness Analysis, June 1, 2017 ,P . 7 ,
88 Washington 2017 DSM Annual Conservation Report & Cost-Effectiveness Analysis, June l, 2018, P. 6.
Exhibit No. 'l
Page 6-9 P. Ehrbar, Avista
Page 175 ol 224
o
Also, "[a]t the start of 2017, the Washington electric tariff rider was underfunded by
$8,283,048."8e "The primary driver for the underfunded balance was the unanticipated high
participation in the non-residential lighting program in2017."e0 Similarly, for natural gas the
tariff rider balance was underfunded by $1,410,964 at the start of 2017 and there was an
underfunded balance of $626,653 at year-end.er These budget figures illustrate the positive
operation of decoupling. Decoupling is not a driver for energy conservation. But itfacilitates
pursuit of all cost-effective energl conservation in accord with Commission direction. Anyone
who has been present in a non-decoupled utility when a planned program budget cap is reached
has heard staff telling customers that the budget cap has been reached, so they should consider
tracking when the program will reopen in the next year and get their application in immediately.
From experience, we have seen major programs (elsewhere) that are open for applications for
one or two days ayear. With decoupling, that barrier is removed; so, programs can follow the
direction ofl-937 to pursue all cost-effective conservation.
Residentiul Electric Group
As shown in the accompanying graph (Figure 6-3), residential electric conservation achievement
dips in 2015 (as decoupling starts, but before decoupling has any effect on customer bills), jumps
in2016 (which has negligible bill effect from decoupling) and dips back to the 2014 pre-
decoupling level in 2017 (the first full year subject to the decoupling adder each month).
However, the reasons for these changes have little or nothing to do with decoupling.
For 2013, a major concern in planning was how to deal with the Washington I-937 Standards for
the 2014-2015 Biennium.e2 For example, an agreement was reached holding that the unit energy
savings used by the third-party completing Avista's CPA (used to establish thel-937 target) will
remain fixed for the duration of that biennium, and there was a resolution of the problem of
different market forecasting methods used by NEEA, reducing uncertainty for the Company.
There were no major changes to residential electric programs. Decoupling was not mentioned in
analysis or presentation.
In the 2015 Business Plan, Avista noted that "...falling avoided costs permeate throughout all
phases of DSM operations and will require considerable innovation and flexibility in order to
continue to deliver value to the customer."e3
8e Ibid., P. 4.
eo Ibid., P. 4.
er Ibid., P. 4.
e2 Avista Utilities Washington/Idaho2014 Demand-Side Management Business Plan, November 1,2013, Pp. l4-18.
e3 Avista Utilities Washington/Idaho 2015 Demand-Side Management Business Plan, October 31,2014,P.4.
Exhibit No. 1
Page 6-10 P. Ehrbar, Avista
Page 176 oI 224
6
Rcsidcntial Elechic Conseryation Achievement (MWh)
i
=
Figure 6-3. Conservation Acltieventent - Residential Electric
Further, the bundling of measures into programs was creatively optimized as follows. "The
Company provides the highest possible value for the cost-effectiveness metric applicable to each
program, maximizing the residual benefits (benefits less costs) of the applicable metric.
This choice plays an important role in the Company's planning process and the development of
the final portfolio in three ways:ea
1. By maximizing the portfolio residual benefits the Company will seek to add measures
and programs to the extent that the incremental benefits of that resource option exceed
the incremental cost. This approach precludes the rejection of measures or programs that
favorably contribute to the cost-effectiveness of the portfolio but are not able to bear the
non-incremental infrastructure cost that would be assigned to the program.
2. By only burdening measures and programs with the costs that are incremental to them at
each level of aggregation, the potential for a 'death spiral' is reduced. If each measure
were required to bear their fully allocated (including non-incremental) costs,
incrementally cost-effective measures would potentially fail and, by being excluded from
the portfolio, increase the non-incremental cost allocation to be bome by other measures.
3. In comparison to simply establishing a benefit-to-cost ratio in excess of 1.00 as a target,
Avista's chosen approach leads to a larger portfolio as well as one which has higher
residual benefits. It does this by providing a means for accepting cost-effective but
marginal measures and programs that favorably contribute to the portfolio's residual
benefits but may reduce the overall portfolio benefit-to-cost ratio."
Residential program and approaches were continued from the prior year. All analysis and
discussion for 2015 was based on policy approaches and technical considerations. Decoupling
ea Avista Utilities Washington/Idaho 2015 Demand-Side Management Business Plan, October 31,2014,P.7.
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 177 ol 224
Page 6-1 I
6
was not mentioned in analysis or presentation.es Similarly, discussion followed technical and
policy approaches. Decoupling was not discussed in the 2016 plan,e6 or the 2017 plan.eT
Low-Income Electric Group
As shown in Figure 6-4, low-income electric conservation achievement rises slightly from 2014
to 2015 (as decoupling starts, but before decoupling has any effect on customer bills), jumps in
2016 (which has negligible bill effect from decoupling) and dips back to slightly below the 2014
pre-decoupling level in20l7 (the first full year subject to the decoupling adder each month).
However, the reasons for these changes have to do with program realities rather than with
decoupling.
Avista uses a system of pre-approved measures to facilitate low-income weatherization work by
the implementation agencies. Avista also notes that "CAP agencies individually prioritize and
treat their clients based upon a number ofcharacteristics. Several ofthe characteristics used to
prioritize clients are related to resource cost-effectiveness, but cost-effectiveness based specifically
upon the TRC or UCT test is not an explicit priority for the CAP agency."e8 There were no major
changes in electric low-income programs. Decoupling was not mentioned in analysis or
presentation for 2014.
For 2015, the approach to implementation of low-income weatherizationwas continued from
2014, with the same budget commitment. Decoupling was not mentioned in the analysis or
presentation of Annual Conservation Plans for 2014,2015,2016 or 2017. For 2017, Avista
notes openness to working towards a waiver for low-income electric customers like the waiver in
effect for low-income nafural gas customers.ee
Low]ncom. Elccbic Cons.ryation Achievsm.nt
==
Figtrre 6-4. Conservation Achievement - Low-Income Electric
es Avista Utilities Washington/Idaho 20 I 5 Demand-Side Management Business Plan, October 3 I , 20 14, P. I 0.
e6 Avista Utilities Washington/Idaho 2016 Demand-Side Management Business Plan, October 26,2015, Pp. 9-10.
e7 Avista Utilities Washington 2017 Electric Demand-Side Management Annual Conservation Plan, November 15,
2015, Pp.7-8.
e8 Avista Utilities Washington/Idaho 2014 Demand-Side Management Business Plan, November l, 2013, P. 19.
ee Avista Utilities Washington 2017 Electric Demand-Side Management Annual Conservation Plan, November 15,
2016, P. 10.
Exhibit No. 1
Page 6-12 Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 178 of 224
6
Nonresidential Electric Group
As shown in Figure 6-5, Nonresidential electric conservation achievement rises slightly in 2015
(as decoupling starts, but before decoupling has any effect on customer bills), jumps in 2016
(which has negligible bill effect from decoupling) and then rises further in20l7 (the first full
year subject to the decoupling adder each month). However, the reasons for these changes have
to do with I-937 planning and program-level realities, rather than decoupling.loo
Avista provides both prescriptive and site-specific programs (which may be proposed by the
customer). Two improvements were:
. Revisions to the site-specific program implementation processes to improve clarity and
promote the timely movement of projects through the pipeline.
. The establishment of two checklists (or "Top Sheets"), one prior to contracting and one
prior to the payment of the incentive, in order to ensure consistent documentation and
treatment of each project as it progresses through these processes towards completion.
Nonr!rldcnthl
II
F igure 6-5. llonres identia I E lectric Conservat ion Ac lt i eventent ( MWh)
There were also three changes to Washington Schedule 90, affecting electric programs:
. Shift the maximum energy simple payback for incentive eligibility from eight years to
thirteen years for lighting measures with independently verified lives of 40,000 hours or
more (e.g. LED lighting).o Increase the maximum incentive from 50% of customer incremental cost to 70Yo of
customer incremental cost for (1) typical lighting measures (those with lives under
40,000 hours) with energy simple paybacks under three years and (2) all other measures
with energy simple paybacks less than five years.o Clarification regarding how incentive caps apply to prescriptive measure applications.
Otherwise, non-residential electric programs for 2014 continued as in the prior years and
marketing continued to be based primarily on an account manager approach. There were no
major changes in electric non-residential programs. Decoupling was not mentioned in analysis or
presentation for 2014.
r00 Avista Utilities Washington/Idaho2014-2017 Demand-Side Management Business Plans.
Exhibit No. 1
Page 6-13 Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 179 ot 224
6
For 2015, the20l4 program was continued with some technical adjustments. Decoupling was
not mentioned in the analysis or presentation for 2015,101 2016,102 or 2017.103
Residential Naturul Gas Group
Leading up to the planning study and following direction from the Commission, Avista was
continuing the Washington natural gas portfolio under a gross Utility Cost Test (UCT) metric
rather than the previously applied net TRC. This was the first time that the Company employed
the UCT test as the primary metric for optimizing portfolio performance.roa This switch to the
UCT has its source in the fall in the commodity cost of natural gas due to extensive fracking in
the USl0s. Successful technological improvements in fracking have caused avoided cost to fall
dramatically. This change has also meaningfully lowered the cost-effectiveness of much natural
gas DSM,106 with some carryover to electric DSM. Since residential natural gas programs were
resumed using the UCT test, these programs were evolved to meet the UCT test and continued.
As shown in Figure 6-6, residential natural gas Conservation Achievement dropped from 2014 to
2015, then rose in2016 and rose again in 2017 (the first year in which customers experienced the
decoupling bill adder each month). However, the reasons for these changes have to do with
program realities, rather than with decoupling. Decoupling was not mentioned in analysis or
presentation for 2014. Residential natural gas programs continued from2014 through 2015,
2016 and2Ol7,107 again with no mention of decoupling in either analysis or presentation.
Figure 6-6. Residential Natural Gas Conservation Achievement
rot Avista Utilities Washington/Idaho 2015 Demand-Side Management Business Plan, October 31,2014,Pp. 12-14.
r02 Avista Utilities Washington/Idaho 2016 Demand-Side Management Business Plan, October 26,2015,Pp. I l-12.
r03 Avista Utilities Washington 2017 Electric Demand-Side Management Annual Conservation Plan, November 15,
2016,Pp.7-8.
roa Avista Utilities Washington/Idaho 2014 Demand-Side Management Business Plan, November l, 2013, P. 4.
r05 Estimate ofthe percentage ofregional US natural gas that is fracked range from about 50% to70o/o, depending on
region. The cost reduction caused by fracking is estimated to be from about $180 to $430 per residential customer
per year. This is the equivalent of a very substantial customer discount program and it applies to all customers, not
only low-income households.
106 It also translated into lower avoided cost for electricity from natural gas generation, but generally electric
measures remained cost-effective.
r07 Avista Utilities Washington 2017 Gas Demand-Side Management Annual Conservation Plan, November 15,
2016,Pp.5-7.
Exhibit No. 1
Page 6-14 Case Nos. AVU-E-19-0_ and AVU-G-I9-0_
P. Ehrbar, Avista
Page 18O ol 224
E
0
Low-Income Natural Gas Group
The low-income programs are special since though they are referenced to cost-effectiveness, it is
understood that low-income customers are not able to receive weatherization services unless the
cost is fully paid by the utility or other transfer such as federal and state funding and voluntary
contributions. Low-income weatherization is substantially supplemented by state and federal
funding. As with electric low-income weatherization, "CAP agencies individually prioritize and
treat their clients based upon a number of characteristics. Several ofthe characteristics used to
prioritize clients are related to resource cost-effectiveness, but cost-effectiveness based specifically
upon the TRC or UCT test is not an explicit priority for the CAP agency."r08 Federal and state policy
substantially guides low-income weatherization.
As shown in Figure 6-7, low-income Conservation Achievement dipped from20l4 to 2015, then
increased dramatically in2016 and dropped to below the pre-decoupling 2014 level in2017 .
These changes were not driven by decoupling. Decoupling was not mentioned in analysis or
presentation in2014,2015,2016 or 2017 DSM Annual Conservation Report & Cost
Effectiveness Analyses. In2077, natural gas low-income progmms operated using a waiver
system for natural gas measures that permits full-funding of those measures.l0e'110
Low lncomc Natulal Gas Cons!ryadon
F
Figure 6-7. Low-lncome Nalut'ol Gos Conservqtion Achievement
For low-income customers, "[t]he list of measures offered is derived from the Department of
Commerce's Weatherization Manual. To guide the agency toward projects that are most
beneficial for the Company's energy efficiency efforts, in most cases an "Approved" list of
measures is provided that allows for fulI reimbursement of those that in most cases have a Total
Resource Cost (TRC) of 1 or better. For efficiency measures with a TRC less than 1, a "Rebate"
r08 Avista Utilities Washinglon/Idaho 2014 Demand-Side Management Business Plan, November l, 2013, P. 19.
roe Avista Utilities Washington 2017 Electric Demand-Side Management Annual Conservation Plan, November 15,
2016, P. 10.
r10 Avista Utilities Washington 2017 Gas Demand-Side Management Annual Conservation Plan, November 15,
2016, P. 8.
Exhibit No. 1
Page 6-15 P. Ehrbar. Avista
Page 181 of 224
O
that is equal to the Company's avoided cost of energy is provided as the reimbursement to the
Agency."lll
Nonresidentiul Nqtural Gas Group
Nonresidential natural gas Conservation Achievement (Figure 6-8) rises dramatically from 2014
to 2015 (as decoupling starts, but before decoupling has any effect on customer bills. Then
achievement drops dramatically from 2015 to 2016 (which has negligible bill effect from
decoupling). Achievement then from 2016 to 2077, reaching to a point just above the 2014 (pre-
decoupling) level in20l7 (the first fulIyear with the decoupling adder).
Nonraridantial Natural Gaa Concctudion
Yar
Figure 6-8. Nonresidential Natural Gas Conservalion Achievement
Schedule 190 (natural gas efficiency, Washington only) was modified as follows:r12o Decrease the incentives of each of the incentive tiers by approximately 1/3rd due to the
decrease in avoided costs.o Eliminate the maximum energy simple payback of thirteen years for incentive eligibility. Clarification regarding how incentive caps apply to prescriptive measure applications.
The revisions to the Washington Schedule 190 tariff were part of a larger interim planning
process designed to optimize the natural gas DSM portfolio for improved performance against a
gross UCT cost-effectiveness metric. Decoupling was not mentioned in the analysis or
presentation in 2014, 2015, 201 6 or 2017 .tt3
Summary - Impact on Conservation Achievement
In this section of the evaluation, we have shown that decoupling was an important factor
facilitating Conservation Achievement, but not a driver of Conservation Achievement. On the
rrr Avista Utilities Washington 2017 Gas Demand-Side Management Annual Conservation Plan, November 15,
2016, P. 8.
r 12 Avista Utilities Washington/Idaho 20 l4 Demand-Side Management Business Plan, Novemb er l, 2013, P . 23.
r13 Avista Washington 2014 Annual Conservation Report (ACR) & Cost-Effectiveness Analysis, May 29,2015;
Avista Washington 2015 Annual Conservation Report & Cost Effectiveness Analysis, May 31, 2016; Avista
Washington 2016 DSM Annual Conservation Report & Cost-Effectiveness Analysis, June l, 2017; Avista Utilities
Washington 2017 Gas Demand-Side Management Annual Conservation Plan, November 15,2016, Pp. 8-9.
Exhibit No. 1
Page 6-16 Case Nos. AVU-E-19-0_ and AVU-G-19-o_
P. Ehrbar, Avista
Page 182 of 224
6
electric side the l-937 ten-year plan was the primary driver. On the natural gas side, Commission
direction towards use of the gross UCT test was a primary driver (in maintaining or expanding
programs that were not cost-effective using the net TRC test). On both the electric and natural
gas sides, the size of the signal from decoupling was too small to be of meaningful impact on
Conservation Achievement, and, in any case, the signal is neutral.
Considered subjectively, these decoupling signals were even smaller because so many other
programmatic and policy efforts occurred at the same time. Also, the price signals were mixed
as to sign (plus or minus). It comes down to the fact that decoupling is known to be a way to
remove the "throughput" barrier to energy conservation, but not as a stimulus to energy
conservation. The removal of a barrier does not in itself provide a "pull" towards energy
efficiency. Based on this analysis, we conclude that there is no evidence that decoupling has any
meaningful impact as a driver for energy Conservation Achievement. However, in the presence
of a strong driver likel-937 or a strong driver like Commission direction to use the gross UCT
test, it provides revenue stability and more timely revenue recovery and so is a part of a
"package" in that it eliminates the "throughput" incentive. Decoupling comes in when a
program is exceeding its planningtarget, sometimes by a large amount. Where a non-decoupled
utility will turn away energy conservation customers, having reached its budget cap, Avista has
demonstrated that a decoupled utility can keep on servicing to acquire all cost-effective energy
conservation.r14 This is also the perspective of the Regulatory Assistance Project (Figure 6-91.tts
Decoupling eliminates a strong disincentive to invest in energy efficiency. By itself, however,
decoupling does not provide the utility with a positive incentive to invest in energy efficiency
or other customer-sited resources, but it does remove the utility's natural antagonism to such
resources due to their adverse impact on short-run profits.
Figure 6-9. Regulatory Assistance Pro.ject on Decoupling
We should note as a qualification that our conclusions are based on analysis of fourteen months
of application of the decoupling adjustments (Schedules 75 and 175) on customer bills, for the
last two months of calendar 2016 and calendar year 2017. It is possible that long-run impacts
might be different. There is also a lagged impact on decoupling revenue from conservation
achievements that leads to higher decoupling revenue collected from the rate group achieving the
savings. Essentially what current program participants in a rate group do not pay toward fixed
costs through volumetric charges is collected from everyone else in the rate group through future
decoupling revenues. Conservation savings cumulate until a rate case resets the test year
Ita Another benefit of decoupling is illustrated in comparison to the altemative of assigning all variable costs to
variable charges and all fixed costs to hxed charges. This alternative would require alarge, non-bypassable fixed
fee each month and result in a low volumetric charge. This would create difficult economics for low and moderate-
income customers and very effrcient customers. It would raise strong barriers to the dollar value of conservation to
customers when it comes to the "please pay" amount on customer bills. Again, however, this is an instance of
decoupling removing barriers to energy conservation. It is not a case of decoupling acting as a driver to stimulate
energy conservation.
rrs Regulatory Assistance Project, Revenue Regulation and Decoupling, A Guide to Theory and Application.
Second Printing, November 2016 (https:i/u,u'u'.rapon line.orgiu'p-content/uploads/20 I (r/ I I i rap-re vcnue-regu lation-
decouplin g-euide-second-printin g-20 I 6-rrov enrber.pdf).
Exhibit No. 1
P. Ehrbar, Avista
Page 183 of 224
Page 6-17
0
incorporating recent program savings into the new base. This is true regardless of the prevailing
decoupling rate at the time of conservation savings. Electric to natural gas conversions result,
with a lag, in higher elechic decoupling revenue to recover fixed electric system costs that
conversion participants are no longer paying and lower natural gas decoupling revenue to refund
the over collection of natural gas system fixed cost by the same conversion participants.
Exhibit No. 1
P. Ehrbar, Avista
Page 184 of 224
Page 6-18 UaSe l\OS. AVU-tr- lY-U_ ano AVU-( - lV-U_
Section 7.sis of Possible Adverse Im acts
Decoupling is a purposive reform designed "...to ensure that utilities have a reasonable
opportunity to earn the same revenues that they would under conventional regulation,
independent of changes in sales volume."l16 An optimal decoupling mechanism would achieve
revenue neutrality while removing the inherent management and organizational drive to increase
energy sales ("the throughput incentive").
Sometimes, purposive programs have unintended side effects. Here we focus on possible
adverse impacts caused by or associated with decoupling (Figure 7-1).
Task 7: Analysis of Possible Adverse Impacts
Identification of any conclusive evidence to suggest that the Mechanisms adversely
impacted customer service, distorted price signals for customers resulting in lower
participation in conservation programs, or eroded Avista's incentive to control costs
and improve efficiency and/or Washington required service quality measures.
Figure 7-1. Identify Adverse Impacts
Are there Adverse Effects?
Both formal learning and lessons of experience teach us that any rationally designed and
purposive program may develop unanticipated side effects.llT No matter how skilled the
development, or the degree of integrity and insight from which a program springs, or the ability
of policy reform to achieve intended results in actual practice, any reform may have
unanticipated and unintended consequences.ll8 The high-level question in this section of the
evaluation is to determine if there is any conclusive evidence to suggest that the Mechanisms
adversely impacted Avista's customer service, created price signals that lowered participation in
tt6 Lazar, Jim, "Examples of Good, Bad, and Ugly Decoupling Mechanisms", presentation to NARUC Symposium:
Aligning Regulatory Incentives with Demand-Side Resources. San Francisco, Califomia Augrst2,2006
(https://pubs.naruc.orgipub.cfin?id:4AC7A83 F-235.1-D7 l.l-5 li0--1C68971 7l iC B).
rr7 Although the recognition of unintended/unanticipated consequences is currently attributed to Merton, Merton
himself notes a deep historic chain of prior writers: "In some one of its numerous forms, the problem of the
unanticipated consequences ofpurposive action has been treated by virtually every substantial contributor to the
history of social thought." See: Merton, Robert K, "The Unanticipated Consequences of Purposive Social Action,"
American Sociological Review,Yol. l, No. 6 (December., 1936), pp. 894-904. Beyond this, by observation,
intelligent animals experience unanticipated consequences, so it is quite likely that, being a phenomenon observed in
animals, experiential recognition of unintended consequences is older than human history. This observation of the
historically deep experience of unanticipated consequences fits with the Darwinian model for both biological and
social evolution.ll8 Following Donald Campbell, the terms "program" and "reform" are used interchangeably: a new approach or
program, such as decoupling - a policy reform effected in govemance and institutional practice is both a program
and a reform.
Exhibit No. 1
6
Page 7-l P. Ehrbar, Avista
Page 185 ot 224
o
conservation programs, or eroded Avista's incentive to control costs and improve efficiency
and/or Washington required service qualrty measures.lle
Following the research questions for this evaluation, we focus on three sub-areas:
o Did decoupling impact Avista's service qualrty, on the Washington required service quality
measures?o Were there decoupling price signals that resulted in lower participation in conservation
programs?o Did decoupling erode Avista's incentive to control costs and improve efficiency?
Customer Service snd Service Quality Indices (SQD
Avista implements the State of Washington required Service Quality Indices (SQI) and reliability
measures.l20 The existence of this series of yearly reports permits examination of customer
service metrics to see if service goals have been met since the beginning of decoupling in 201 5
and/or since the first impact of decoupling on energy bills in November 2016.
First, we examine Avista Service Quality Indices following decoupling to see if service goals
were met, keeping in mind that calendar 2017 is the only year fully within the "after decoupling"
time window from a customer perspective. As shown in the tables for 2015, 2016 and2017
service goals were achieved each year. There were no negative fficts on these SQI indicators.
We may also note that there were no positive effects on the SQI indicators. For example,
"Percent of customers satisfied with our Contact Center services, based on survey results" was
about 960/o for 201 5, 93o/o for 2016 and 94o/o for 201 6, so within a band of 3o/o. The complex
nature of the formation of indicator values in terms of context (for example, weather) and human
behavior suggests that as a methodological rule, key performance indicators (KPIs) not be over-
interpreted. We expect yearly results on each KPI to dance around from year to year within a
reasonably judgmentally assessed neutral bandwidth without the size or direction of differences
conveying meaning. A sense for defining a "neutral band" is developed from practical
experience.
Conceptually this "neutral band" is made up of movements in indicators that result from a very
large mix of small influences from a large range of factors including both proximate and remote
influences. In addition, many of the active factors are likely random. So, performance tables
like Table 7-l through Table 7-7 usually cannot be used to analyze these small differences
(positive or negative).
Though not useful for assessing small differences, KPIs provide a powerful tool so that
regulators can monitor a utility's performance. The primary use of the KPIs is to make
achievement of regulatory goals explicit. This is shown, using check boxes in the final columns
rre Sometimes side effects may be anticipated by some parties while the preponderance of parties involved in
shaping, managing and implementing a program may not see a side effect, except retrospectively. In such a case we
might say, retrospectively, that the effect was "hidden in plain sight".
r20 The Washington required Service Quality Indices are provided by Avista in response to H. Gil Peach &
Associates LLC Data Request No. 52.
Exhibit No. 1
Page 7-2 Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 186 of 224
6
of Table 7-1 through Table 7-3. Where there has been a regulatory reform such as decoupling, a
secondary use of KPIs is in review to determine if there has been a correlated systematic
structure of change in KPI results (either a directionally consistent string of positive or negative
results by year (regardless of size) or a directionally consistent string of large positive or
negative results by year). While for decoupling the primary question concerns possible adverse
effects, results might be positive as well as negative.
If either a directionally consistent string of small changes or a directionally consistent string of
large changes is found, then the question shifts from correlation to possible causation. For
example, in Washington it would not be unusual to find that severe weather events or severe
weather patterns is the primary cause for change in KPI results. Also, we have sometimes found
that when customer contact or services are outsourced, change can be due to performance of a
particular service vendor or replacement by a different service vendor.
We find no directionally consistent string of either small or large changes in this analysis. There
are no meaningful patterns evident in these tables of this section of the study (Section 7).
Performance is high and consistently high. There are no meaningful negative or positive effects
on any of the Section 7 KPIs.
Table 7-1. 2015 Indicotors of Customer Service Quality DR 52
Customer Service Measures Benchmark
2015
Performance Achieved
Percent of customers satisfied with our Contact
Center services, based on survey results At least 90%96.t%,/
Percent of customers satisfied with field
services, based on survey results At least 90%96.8%,/
Number of complaints to the WUTC per 1,000
customers, per year Less than 0.40 0.tl ,/
Percent of calls answered live within 60
seconds by our Contact Center At least 80%80.70 *,/
Average time from customer call to arrival of
field technicians in response to electric system
emergencies, per year
No more than 80
minutes 44 Minutes ,/
Average time from customer call to arrival of
field technicians in response to natural gas
system emergencies, per year
No more than 55
minutes 51 Minutes ,/
* Results for 201 5 on percent of calls answered live within 60 seconds by the Avista Contact Center include all
calls received for the year, including the nearly 56,000 calls answered during the November Wind Storm event
from November 1 7 through November 27 , 2015.
Exhibit No. 1
Page 7-3 wAVU-G-'r9-0_
P. Ehrlcar, Avista
Page 187 ot 224
Customer Service Measures Benchmark
2016
Performance Achieved
Percent of customers satisfied with our Contact
Center services, based on survey results At least 90%92.7%,/
Percent of customers satisfied with field
services, based on survey results At least 90%94.7%,/
Number of complaints to the WUTC per 1,000
customers, per year Less than 0.40 0.25 ,/
Percent of calls answered live within 60 seconds
by our Contact Center At least 80%8r.7%,/
Average time from customer call to arrival of
field technicians in response to electric system
emergencies, per year
No more than 80
minutes 39.3 Minutes ,/
Average time flom customer call to arrival of
field technicians in response to nafural gas
system emergencies, per year
No more than 55
minutes 48.4 Minutes ,/
6
Table 7-2. 2016 Indicators of Customer Service Quality - DR 52
Table 7-3. 2017 Indicators of Customer Service Quality DR 52
Customer Service Measures Benchmark
2017
Performance Achieved
Percent of customers satisfied with our Contact
Center services, based on survey results At least 90%93.6%,/
Percent of customers satisfied with field services,
based on survey results At least 90%95.2%,/
Number of complaints to the WUTC per 1,000
customers, per year Less than 0.40 0.16 ,/
Percent of calls answered live within 60 seconds
by our Contact Center At least 80%81.5%,/
Average time from customer call to arrival of field
technicians in response to electric system
emergencies, per year
No more than
80 minutes 39.9 Minutes ,/
Average time from customer call to arrival of field
technicians in response to natural gas system
emergencies, per year
No more than
55 minutes 50.29 Minutes ,/
Next, as shown in Table 7-4, for customer service measures that were collected both before and
after decoupling, there is no change in the perceived level of customer service by customers.
Given the very small fluctuations in year-to-year, these results are stable from2012 through
2017. There were no negative effects on these "before and after" SQI indicators.
Exhibit No. 'l
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 188 of 224
Page 7-4
Table 7-4. Customer Service Indicatorsfor Before and After Decoupling- DR 52
Customer Service Measure 2012 2013 2014 20ls 2016 2017
Percent Satisfied with Contact Center Services 93.1%94.1o/o 94.9%96-1Y.92.7%93.6%
Percent Satisfied with Field Services 93.3%95.2%94.4%96.8%94.7%95.2%
Percent Calls Answered in 60 Seconds 83.7%82.8%82.9%80.7%8t.7%81.5%
Note: Percent Satisfied includes customers who were either "satisfied" or "very satisfied" with their service.
Note: Results for 2015 on percent of calls answered live within 60 seconds by the Avista Contact Center include all
calls received for the year, including the nearly 56,000 calls answered during the November Wind Storm event from
November I 7 through November 27, 201 5.
For electrical reliability (Table 7-5) two measures are reported. The System Average
Intemrption Frequency Index (SAIFI) and the System Average Intemrption Duration Index
(SAIDD. SAIFI indicates the frequency of long-term (greater than five minutes) service
intemrptions. Reliability improves as SAIFI becomes smaller. The System Average Intemrption
Duration lndex (SAIDD measures the duration of long-term (greater than five minutes) service
intemrptions. Reliability improves as SAIDI becomes smaller. As shown in Table 7-5, values of
SAIFI and SAIDI change from year to year. The highest values for both occur in20l7, the first
full post decoupling year. However, this fluctuation does not provide conclusive evidence of a
meaningful change. One would need to see a pattern (beginning with the values of the2017
indicators) that continues for more years before drawing a systematic conclusion (negative or
positive).121 For electric reliability, there is no conclusive evidence of an adverse effect.
121 Also, one would need to see if there is an explanation for the fluctuation in sources other than decoupling, such as
weather. Avista, in response to Data Request 080, fills out the contextual background needed to more fully
understand fluctuation in SAIFI and SAIDI (emphasis in italic added): "As noted on pages 53-57 of Avista's
Customer Service Quality and Electric System Reliability report for 2017, approximately twolhirds of the utility's
system performance each year is subject to randomforces such as weather patterns and storms, or other random
events such as an outage caused by a car striking a pole, whichfactors are generally beyond the control ofthe
utility. Consequently, there is a natural variation in results (both up and down) from year to year, due largely to the
interaction of these randomfactors. The "direction" of the annual results and the magnitude of the variation
generally reJlects the combination of the frequency and magnitude ofweather-related events, the contribution of
olher randomly-occurringfactors, as well as the effect of standardized adjustments made to the yearly results based
on "major event days" (please see footnote 47 onpage 54 of the above-mentioned Service Quality and Reliability
report for 2017). As an illustration of these principles in action, Avista's SAIFI score for 2016 was the lowest value
recorded since our 2005 baseline year, while the 2017 result was the fifth highest recorded over the same period.
Likewise, the annual score for SAIDI in 2016 was the third lowest measured since 2005, while the value for 2011
was the second highest measured over the same period of time. Generally, the results for 2017 reJlect the greater
storm activity we experienced compared with 2016, combined with the relatively small downward adjustment in the
numbers based on minimal major events in 2017." We accept this explanation for this evaluation.
Exhibit No. 1
Page 7-5 Case Nos. AVU-E-19-0 and AVU-G-19-0
P. Ehrbar, Avista
Page 189 of 224
6
Electric Service Reliabilitv Measure 2012 2013 2014 2015 2016 2017
SAIFI System Average Intemrption Frequency Index 1.14 1.05 l.l I 1.05 0.86 t.20
SAIDI System Average Intemrption Duration Index 138 138 139 163 133 183
Note: The System Average Interruption Frequency Index or "SAIFI" is the average number of sustained intemrptions (outages) per customer
for the year.
Note: The System Average Intemrption Duration Index or "SAIDI" is the average duration of sustained intemrptions (outages) per customer
for the year (measured in minutes).
Table 7-5. Indicators of Electric Service Reliability DR 52
Begiruring January 1,2016, Avista introduced a new set of indicators, which can also be
considered a very visible tool to motivate staff with the Customer Service Guarantee to
Washington customers.l22 There are seven specific performance guarantees. Missing the goal
for performance on a guarantee will result in a payment of fifty dollars ($50) as a credit on the
customer bill.l23 As shown in Table 7-6 and Table 7-7, Avista's performance on these new
indicators is very good, with an error rate of about five out of a thousand (0.0053) for 2016 and
of about two out of a thousand (0.0023) in20l7.
Taken together, these service quality results show no adverse impacts of decoupling on service
quality. There are only two measurement years for these results and the values are so small
relative to the number of customers that weather and small influences and random factors are
likely to predominate in generating results. Several years of measurement or the occurrence of
large effects in results would be needed to demonstrate correlation and then call for a search for
causation. With the data that exists, there is no indication of adverse effect of decoupling on
customer service.
122 See: Response to Data Request 081 and: https:inruu'.rn)'avista.com/about-us/contact-us/customer-st'rr icc--
guarantees.
123 Subject to conditions. There is no payment if a customer cancels or misses an appointment or if the Company
reschedules an appointment with at least 24-hours' notice; or, if there is a major weather event that impacts a large
number of customers or lasts for a longer period of time, such as a major snow, ice, or wind storm; or, if there is an
action or default by someone other an Avista employee or outside of Avista's control; or, if construction is required
before service can be energized, evidence that all required govemment inspections have been satisfied has not been
received by Avista, required payments to Avista have not been received, or service has been disconnected for non-
payment or there has been theff/diversion of electric service; or, when power is interrupted for less than five
minutes, power is interrupted because of work on a meter, or the safety of the public or of Avista employees or the
imminent failure of Avista equipment was a factor causing the intemrption in service.
Exhibit No. 1
0
Page 7-6 P. Ehrbar, Avista
Page 19O ol 224
6
Table 7-6. 2016 Customer Service Guarantee - DR 52
Table 7-7. 2017 Customer Service Guarantee - DR 52
Customer Service Guarantee Successful Missed $ Paid
Keeping Our Electric and Natural Gas Service Appointments
scheduled with our customers 1,477 l0 $s00
Restore service within 24 hours of a customer reporting an
outage (excluding major storm events)26,344 I $50
Turn on power within a business day of receiving the request 3,380 J $ 150
Provide a cost estimate for new electric or nafural gas service
within l0 business days of receiving the request 5,024 0 $0
Investigate and respond to a billing inquiry within l0 business
days ifunable to answer a question on first contact 1,760 0 $0
Investigate a reported meter problem or conduct a meter test
and report the results within 20 business days 309 2 $100
Noti$, customers at least 24 hours in advance of a planned
power outage lasting longer than 5 minutes 30,336 349 $ 17,450
Totals 68,630 365 $18,250
Customer Service Guarantee Successful Missed $ Paid
Keeping Our Electric and Natural Gas Service Appointments
scheduled with our customers 1,584 ll $ss0
Restore service within 24 hours of a customer reporting an
outage (excluding maior storm events)30,669 23 s 1,150
Turn on power within a business day of receiving the request 9,551 0 $0
Provide a cost estimate for new electric or natural gas service
within l0 business days of receiving the request 3,929 0 $0
Investigate and respond to a billing inquiry within l0
business days ifunable to answer a question on first contact 1,623 0 $0
Investigate a reported meter problem or conduct a meter test
and reoort the results within 20 business davs 1,082 I $s0
Notify customers at least 24 hours in advance of a planned
power outage lasting longer than 5 minutes 17,079 115 $5,7s0
Totals 65,523 150 $7,500
Exhibit No. 'l
P. Ehrbar, Avista
Page 191 ol 224
Page 7-7
Price Signals and Conservation ation
Decoupling does not change the overall amount of fixed cost to be recovered. It changes the
timing of recovery and reduces volatility by recovering fixed cost not already recovered from
volumetric charges. These amounts are recovered in small yearly increments.l2a Determination of
the revenue requirement associated with fixed cost is a step in the process of developing a cost of
service analysis. Cost of service analysis is a separate form of analysis that occurs independent of
the form of recovery. The decoupling mechanism recovers fixed cost outside of volumetric rates
annually and balances any under-recovery or over-recovery annually. In the absence of
decoupling, the utility would either over or under recover its fixed costs.
With or without decoupling, once established as a revenue requirement, the established fixed cost
is allocated to customer groups. Projected recovery involves construction of planning targets
(projections based on experience). In decoupling, fixed costs are recovered in the volumetric
charge (if energy usage matches planned energy usage); or if there is under-recovery, are set to be
recovered through an adjustment in volumetric rates in the following rate year, subject to certain
control tools, including the three-percent(3%) cap. Similarly, any over-recovery is refunded
through a reduction in volumetric rates in the following rate year. The decoupling allocation of
fixed costs for a customer group is based on the group's actual energy use in relation to the group's
projected energy use.
Historically (and contrary to what might be expected from the term "fixed" cost), many fixed costs
are recovered in volumetric revenue (cost per unit of energy). In Avista's decoupling, two separate
time windows are used: a measurement time window, during which the data for decoupling
adjustment for the next implementation time window is collected; and a rate yeor, arl
implementation time window in which the resulting rate adjustment is applied. In Avista's
decoupling, the measurement time windows are calendar years. When, during a measurement
window calendar year, a $oup decreases energy usage so that the average usage for the group is
below the planning projection for that group for that year, the decoupling adjustment automatically
makes up the lost revenue in the next rate year l2-month implementation window by requiring an
increase in the group's volumetric cost per unit (cost per kWh or cost per therm). Conversely, if in
a measurement time window calendar-year the average usage for a group exceeds the planning
projection, the mechanism will require a reduction in unit cost for the next l2-month
implementation time window (rate year).
Given the decoupling price signols observed, did decoupling price signals influence energ/
conservation effort?
Calendar 20152 The answer is "no" for 2015. While the first measurement window was
calendar 2015, no decoupling amounts were billed to customers during 2015.
124 The more frequent yearly rate effect with decoupling should sum to the (theoretical) less frequent aggregated rate
recovery impact (without decoupling) over a set ofrate cases.
Exhibit No. '1
Page 7-8 Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 192 ot 224
O
o
Calendar 20162 The answer is "no" for 2016 because the signal was too small to influence
changes in energy conservation. Any changes in energy conservation effort in20l6 would be
due to other factors.
In fact, the rate impact of decoupling for the electric decoupled groups was negligible in 2016
(Table 7-8). The first l2-month implementation time window (rate year) ran from November
2016 through October 2017. As shown in the table, no decoupling amounts were billed from
January through October 2016 so there could have been no influence for most of the year. Price
signals were present only in November and December.
Since energy bills are sent using billing cycles (allocated throughout the days of a month) the
price signal phased in across the month of November. The first price signal fully experienced by
decoupled customers occurred in December 2016.
Further, response to a very small price signal usually occurs with a lag. If a response were
beginning to be developed, it would not be detectable until2017. Also, except for special cases,
from experience December is not a likely month for focus on energy conservation projects.l2s
Private life, vacation time, the holidays and the weather tend to envelop people in December.
Institutional efforts tend to slow down, to return to vigor in January.
Table 7-8. Electric Decoupling Signal as Percentage of Average Billfor Calendar 2016
Group Jan-Oct Nov Dec Total (2016)
El: Residential 0%t.t%2.8%0.4Yo
E2A: General Services 0%0.4%-1.2%-0.1%
E2B : Larse General Services 0%-0.6%-1.5%-0.2%
E2C: Pumping 0%-0.4%-13%-0.1%
Table 7-9. Natural Gas Decoupling Signal as Percentage of Average Billfor Calendar 2016
Group Jan-Oct Nov Dec Total (2016)
Gl: Residential lYo 1.2%3.3%0.60h
G2A: General Services 0%1.3%3.0%0.5%
G2B: Large General Services 0%1.4%1.3%-t4.7%
Similarly, the rate impact of decoupling for the natural gas decoupled groups was also negligible
in2016 (Table 7-9). As with decoupled electric service, natural gas service provided no
decoupling price signals until November 2016. As with decoupled electric service, the signal for
decoupled natural gas service was phased in over the days of November due to billing cycles. As
shown in the table, price signals for G1: Residential and for G2A General Services are
negligible, so any changes in conservation effort in2016 would be due to factors other than the
price signal from decoupling. For G2B: Large General Services, there is an anomaly in the data
due to a base problem that occurred in December (and continued through January 2017), so data
from Table 7-9 cannot be used.
r2s An exception is auto plants which typically take advantage of holiday expectations to shut down for a week in
December to implement physical changes in the plant.
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 193 ol 224
Page 7-9
6
Calendar 2017: The answer is also "no" for calendar 2017. Calendar 2017 is the first full year
of customer experience with the decoupling price signals. But, for both electric and natural gas,
the size of both monthly and yearly signals is small (Table 7-10 and Table 7-11). Likely, these
changes would not be noticed. If small changes were to be noticed (positive or negative),
drawing of conclusions or taking actions that might affect conservation would likely occur with a
lag. If were to be an effect, it would not be expected in the first quarter of 2017; and likely not
until the fourth quarter of 2017 or after.
As a customer strategy, it remains true that participation in conservation programs can
substantially lower energy bills. Almost always, this will much more than offset a number of
small rate increases over a number of years. A small rate increase or decrease does not have a
signal strength to outbalance the cost advantage of using fewer units of energy. And, of course,
the price signal from fixed cost will occur anyway, with or without decoupling. Only the timing
would be different.
For 2016, the 3Yo cap came into play for the El: Residential electric group, so there was a limit
on the decoupling adder for 2016 and a deferral carryover to 2017 . However, there was no
deferral carryover from the 20t7 rate year to the 20 I 8 rate year. For natural gas, the 3oh cap
came into play for the Gl: Residential electric group in20l6, creating a deferral carried over into
2017. For this group, there was also a cap for 2017 and a deferral carryover into 2018.
However, the carryover into 2018 was small. Sustained or snowballing deferral can have an
impact on GAAP accounting, which requires that revenues must be recovered within two
years.l26 Avista refers to decoupling deferrals that go unreported in revenue due to GAAP
accounting rules as contra-decoupling deferrals. Contra-decoupling deferrals were recorded for
natural gas in both 2015 and20l6. What happens next depends on the weather. Through 2017,
decoupling is operating as expected (as plarured) and is not presenting price signals that would
adversely affect conservation.
In summary, analysis of price signals and conservation shows no adverse effect from Avista's
decoupling on energy conservation.
126 In the Response to Data Request 064, Avista indicates ways in which the mechanism could be improved: "GAAP
reporting rules do not allow for recognition of revenues from a mechanism like decoupling in excess of the amount
expected to be recovered within 24 months of the end of the deferral period." One solution would be moving to a
July I effective date for implementation of rate changes. Another would be "to make the mechanism more
symmetrical so that in rebate years some benefit could be withheld to offset future surcharges. Please see the
Company's response to Decoupling_DR_058 regarding the higher likelihood of surcharges than rebates due to
continued energy efficiency implementation." We support the proposal for a July I effective date and exploration of
seeking more symmetry.
Exhibit No. 1
Case Nos. AVU-E-19-O_ and AVU-G-I9-O_
P. Ehrbar, Avista
Page 194 ot 224
Page 7-10
ll(u.tb-Nbi'<NF;<bLD soLl (! 0)UO-E-E or
ETFa,
tf
6i
Fit-ll5I
o2oa(!
[J
IF\
q)
bo
tr
oz
=IExtu
F\
a\
L.\
\)
(_)
\
AJMa\\)
rri
o'
q)
b,')s
a,UL
a,)
c)
!
b.
4
bO
$
oUq)
!
V1!
\
n-
e
--:
It-
a,)
-a
F\
N
\
€
q)
-$\)
L
aq)M
!
q)
\
o'
qJ
bnB
UU\
^9\{!
S
b.q
oo
a-
Uq)a
U
Uq)
L{
If.\
q,)
.a
N
?c
t< 6t
F-
co
o\\o
co
o\t'-
co
I0)
\o
o\
\o
\o9(-'
zz o\
$
\oo\n$
o\
$
(J o\00
c.l
^\
c.t
\o
c1
co
oa
\oo\
6i
6\oo
N
\o
co
c.)
a0 o\
c.l
.oo\\c.l
\onc.t
\o6\
oi
o\I-r
N
N
c.)
\o
v')ol
\oo\q(\
o\c.t
co
ta
6\
a.)co
o\c.i
ca
6\N
co
o\
ca
\onc.t
c!
tA
\oo\c.l
co
\o
c.l
co
o\dl
c.t
0),\ca
ca
6\cl
co
o\$
c.)
\o6\crlci
o\c'.)
co
6\\ca
o
a0)&
o
aoot()a
d!0)
o
c.l
qoot(.)a
E
o
q)
oDI!cdrl
triNr'1
"Gtr 6f
\o
co
-o^\
I
s..1
I
s\
I
CJ0)o\\o
<.
5\co +c.l
Iz
o\
ca
o\\o
I
o\Fr
I
o\
I
I o\00
a.l
\o
I
o\
I
o\r-
I
ov)
6\@
N
o\ol
I
\o
I
\o
00
I
bo \o
0q
N
scl
I
\o9
I
o\oo
I
o\0q
N
o\N
I
\o
u.)
I
\o
@
I
o\@
c..l
o\N
I
6\
I
r-
I
cla
o\09
N
-oo\
I
o\
I
\o
\o
I
a 6\oo
N
o\N
I
\o
I
\o5\,{'.)
I
=a
o\@
c.l
c.l
I
\o
9
I
\o
I
otr
o\oo
N
\o
N
I
\o9
I
o\
o\oq
N
o\c.l
I
o\@
I
o
(€
()
()&
-iEl
a0)(,)
!oa
CB!0)
o
tNr!
oo
0)a
!o
()
()sc!J
diNr!
ul
o
Uc{trl
@
Cost Control and Operational Elftciency
We find no indication of any adverse effect of decoupling on the utility's incentive to control costs.
Avista's perspective is that "[t]he adoption of decoupling has not resulted in a change of efforts by
the Company to operate efficiently, rather the Company has, prior to decoupling, and with
decoupling, strived to be as efficient as possible while at the same time providing safe and reliable
service for our customers."127 Further, the Company points out that "[t]he decoupling mechanisms
provide recovery of fixed costs, on a revenue per customer basis, that were approved by the
Commission in a prior general rate case for recovery. To the extent those fixed costs increase, or
escalate, over time, the mechanisms do not provide for recovery of the change in costs above the
approved level already embedded in the allowed revenue per customer. The Company continues
to bear the risk of changes in costs between general rate cases, and therefore must (and has)
manage the business in a prudent manner."l28
By removing the focus on sales, decoupling may permit utility executive management to focus
more effectively on other goals. Because cost recovery proceeds in a decoupled utility following a
target revenue requirement that has already been projected in a commission proceeding, costs have
been anticipated. A focus on cost control can function within this already established revenue
requiremenl to improve eamings. This does not mean that current cost-control projects derive
directly from decoupling. Avista has continually developed cost-control projects prior to
decoupling. However, with decoupling, Avista cannot increase profits by increasing sales but can
only positively improve profits by improving cost control and operational fficiency. The nature of
this relationship under decoupling has been described by the Regulatory Assistance Project (Figure
7-2).
Decoupling does not guarantee utilities a level of earnings, only an assurance of a
level of revenue. If the utility reduces costs, it increases eamings, just as it would
under traditional regulation. Also, because the utility cannot increase profits by
increasing sales, improved operational efficiency is the only means by which it
can boost profits.
Source: The Regulatory Assistance Project, Revenue Regulation & Decoupling: A Guide
to Theory and Application. Montpelier, Vermont: Regulatory Assistance Project, June
2011,P.4s.
Figure 7-2. Increasing Earnings in a Decoupled Utiliry- (RAP)
r27 Response to DR 063
r28 Response to DR 063
Exhibit No. 1
6
Page 7-12 Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 196 ot 224
6
The Company has provided examples of ways that it is lowering operational expenses to benefit
customers:129
Careful evaluation of each component of overall compensation.
We note that utilities typically re-evaluate each element of overall compensation yearly or every
few years. This cost-control tool is likely the same focus that would be implemented with or
without decoupling. Whether or not deriving specifically or in part from decoupling in the
current context, this is an approach to reducing operational expenses.
A current hiring restriction which requires approval of the hiring manager, as well as the
President of Avista, the CFO, the CEO and the Sr. VP for Human Resources for all
replacement or new hire positions.
This step is not a standard cost-control tool and may or may not be related to the influence of
decoupling. It is unusual for a utility to implement this level of review for all replacement or
new hire positions, although utilities may find it prudent to implement controls from time to time
or (alternatively) to open up for new hiring in certain areas or for certain scarce special skills
from time to time. Whether or not deriving specifically or in part from decoupling in the current
context, this is an approach to lower operational expenses.
However, from an independent outside perspective, a potential problem we notice is that staffing
cuts might be a little too deep. We see senior people with great command, knowledge and years
of experience in their assigned areas; we see some staff assigned to understudy senior staff to
provide for a system of succession and backup.l30 We do not see the new hires in general
training or expected staffing depth for intermediate analysts or assistant analysts that would be
typical staffing for a utility in the past. This helps in short term cost control, but we would like
to see more staffing depth to insure hard won experience and tacit knowledge is not lost should
one or two senior staff decide to retire.l3l We have a sense that staffing is a bit thin compared
with other utility clients with whom we recently have been engaged for projects. What works as
a short-run cost savings may not work as well long-term and may have long-term unintended
consequences.
r2e Response to DR 063.
130 In the response to DR 055, Avista notes that DSM staffing has been essentially stable from 2012tlrough2017,
though organization has been rationalized: "The number ofenergy efficiency staffhas remained relatively stable
over the years, but the strucfure has changed over time. Some years the staff levels have increased or decreased,
including part timers, to meet the needs of programs and support staff. Starting in 2010 the structure included
program managers, engineers, and account executives (for commercial customers) that reported to a Director along
with a small group of EM&V and analytics staff that reported to a different Director. In June 2014 there was a
reorganization and the program managers, engineers, and EM&V/analytics staff all started reporting to the same
new Director. The account executives, of which only a portion of their time is for energy efficiency for commercial
customers, continue to report to a different Director who oversees a range of customer services. From time to time
the program managers have shifted programs around to better meet the needs of the programs and the inclusion of
new programs as well in response to the discontinuation of some programs." Our concern is limited to Rate &
Regulatory staffing and DSM staffing - we did not look at other areas of the Company.
l3l In some ways, utilities are like university research labs - it may take one to five years of application to
sufficiently learn a functional area.
Exhibit No. 1
UASC NOS. AVU-E-]9.U ANd AVU-U.I9.U
P. Ehrbar, Avista
Page197 o1224
Page 7-l 3
0
Effective January lr2014, Avista no longer contributes toward medical insurance
premiums for the retiree medical plan.
Beginning January 112020, a new calculation method will shift more expenses to retirees.
To reduce the number of medical office visits, the Company is providing web and phone
based 2417 telemedicine and there is an on-site clinic.
Beginning in20l7, the Company has offered a High Deductible Health Plan along with the
current self-insured plan.
Medical costs are an area that requires constant vigilance for cost-control. Medical cost-control
steps (no longer contributing to premiums for the retiree medical plan, shifting more expenses to
retirees, introducing a telemedicine option and offering a High Deductible Health Plan option)
are all ways to reduce Company medical costs.
Since escalation of medical costs has been a very visible and long-term social problem in the
United States, it is likely that the medical area would have been similarly addressed with or
without decoupling. Whether or not deriving specifically or in part from decoupling, these steps
lower operational expenses.
Effective January lr20l4 the defined benefit pension plan was closed to all non-union
employees hired or re-hired after January lr2014. This transfers risk to employees. The
Company also now offers a lump sum payout to non-union employees, further reducing
risk to the Company.
Utilities typically subscribe to high qualrty market surveys that provide industry benchmarks for
employee salaries and benefits and then adjust salaries and benefits where possible to
approximate these national benchmarks. This is one of the reasons why utility pay and benefit
packages are generally better than those offered in most sectors of the national economy or in
local communities.
We note the general trend across business sectors towards the replacement of defined benefit
pensions by 401K plans. Although comparatively slow to develop in the utility industry, this is
now also a utility industry trend, and so would be indicated by a relevant market study.
However, benchmarking and market matching, while a very useful indicator approach may not
be a fully adequate criterion in this area: additional criteria might be relevant and provide an
alternative perspective. In the short-run, most employees will be in the defined benefit
retirement system so there should be no short-run downside. In the intermediate and long-term,
transferring retirement risk for employee families from the Company to the individual employees
may have unintended effects.
From the end of WWII through the early 1970s, the United States experienced relatively high
economic stability and shared economic growth. Since then, from a working person's
perspective, not so much. This is in part because productivity gains have not transferred to
workers while costs have increased so that the economy is much more fragile than surface
appearance would suggest.
Exhibit No. 1
Page 7-14 P. Ehrbar, Avista
Page 198 ol 224
6
Since most of our analysis is based on looking backwards in time to evaluate how things have
worked up to the present, we need to also make the jump to facing forward. If we envision the
general economic situation in the United States as it belatedly and finally tries to come to grip
with climate change and finds the situation so far advanced that adaptation has become
extremely difficult, we get a very different picture than if we look back to the era that ended in
about 1972. There is no guarantee of economic stability and there appears to be an increasing
risk of political instability, so economics might be working within a different and reduced
context. We have the sense that it is not unlikely that there will be growing percentages of
customers in need of assistance, and that utilities may be needed as anchors for good jobs if there
is a general economic recession ahead.
Other possible concerns are the thin profit margin for producers of fracked natural gas and the
steep decline curve for fractured gas vs. conventional gas wells132; as well as the push towards
exporting natural gas which would likely raise prices in the United States as a firm export market
is established. However, we understand that Company projections of both price and supply
indicate reliable supply at reasonable prices into the future.
One of the characteristics that makes utilities strong and able as organizations has been career
commitment, which likely changes when defined benefit pensions end. Individual employees,
like other nano-investors are largely at the mercy of the market. Non-professional, non-insider
investors are typically hurt during cyclical market downfurns and in the unusual or extreme
events that exceed the "design basis" for normal projected market refurns (extreme events like
9lll or the so called "Great Recession" from which wages have not recovered). Climate change
affects global availability of food, changes living conditions on most of the planet, increasingly
acidifies the oceans and causes great migrations and problems of immigration.
In these changes, small investors, such as employees, likely do better in the long-run with an
institutional guarantee between them and the downside effects of markets which, over a lifetime,
tend to show patterns of stable growth punctuated by severe market events. In addition, with
market fluctuations due to climate change and shortages, markets are not likely to be reliable for
r32 Fracked natural gas currently makes up roughly 70%o ofnatrral gas in the US and producers are having trouble
making a profit due to both over-investment based on speculative financing and the sharp depletion curve for
fracked natural gas compared with conventional natural gas. Fracked natural gas is a low-cost solution, but is
economically fragile even without taking in to account local physical environmental damage to air quality, water
supplies and land, as well as health effects and global climate deterioration due to fugitive methane release
associated with fracking. We note in this connection that the cunent administration is facilitating methane release to
the atmosphere and so is accelerating climate problems. On the positive side, the discovery of rock fracturing
technology and the rapid expansion and further development of fracturing technology has become equivalent to a
very large subsidy that benefits low-income and all other natural gas customers. However, as has been typical of the
natural gas supply curve in the past, eventually the supply curve will turn down. At the same time, climate is
warming will create a declining need for heating. For this critique, please see Mclean, Bethany, "The Next
Financial Crisis Lurks Underground," New York Times, September 1, 2018
(https://u'rvrv.n)'tinres.conr/2018/09/0 l/opinionithe-next-tl.rancial-crisis-lurks-undersround.htnrl). Also see:
Mclean, Bethany, Squdi America: The Truth qbout Fracking and how It's Changing the World. New York,
New York: Columbia Global Reports, 2018.
Exhibit No. 1
P. Ehrbar, Avista
Page 199 of 224
Page 7-1 5
6
the average investor. During this time, it might be valuable for utilities to restore defined benefit
pensions to enable them to be an anchor in their communities and regions.
The Company is introducing more automation for ISiIT and is working towards providing
longer contracts to venders in return for discounts.
From experience, the Information Services/Information Technologies areas have long been
somewhat independent of utility organizational cultures. Utilities are very reliant on data and
computer systems, yet these systems tend to be operated somewhat by their own internal logics
which can sometimes present unexpected yet necessary new costs. Working towards discounts
from venders in these areas is a useful approach to cost-control. Whether or not deriving
specifically or in part from decoupling, this step lowers operational expenses.
We also make the following observations:
o In our interactions with management and staff we found no indications of any
lack of attention to cost control and operational efficiency. We believe that the
company maintains a careful and prudent approach to controlling costs and we
found no indication of any form of dysfunction or fractionalization within the
organization.. We found dedication to high performance, individual and group achievement of
strong technical proficiency and a sense of personal and business commitment to
public service.
o We found no indication of any cynicism, apathy or disaffection during the formal
workday or in informal discussions with management and staff. Staff holds each
other, corporately, to high standards.
o As noted previously, in the discussion of service quality, the service quality
indicators (SQI) are good, which is an indirect indication of operational
efficiency.
One additional aspect of operational efficiency is the relation of rate of return compared with
utility cost of capital. This is not specifically a decoupling question, but it arises in decoupling.
The concern is that if rate of return is consistently higher than utility cost of capital there could
be an advantage in "gold plating" activities subject to the rate of return. As shown inTable 7-12
this relationship does not hold for Avista and so, no adverse effect of this type exists in Avista's
decoupling.l33
t33DR066AttachmentA. TheAverclr-Johnsoneffectistheacademicnanreforwlrat,inindustryjargon,isusually
referred to as "gold plating" or "high-grading". This is a theoretical "moral hazard" of regulated cotnpanies to
engage in excessive arlrounts of capital accumulatior.r in order to expand the volurne of their profits. If companies'
prot'its to capital ratio is regulated at a ceftain percentage then, depending on the gap there nray be a strong incentive
fbr con.rpanies to over-invest in order to increase profits overall. Investment is then optinrized not for operational
efficiency, but for adr.ninistratively supported protit maximization. We do not see this happening with Avista
decoupling. See: Averch, Harvey; Johnson, Leland L. (1962). "Behavior ofthe Firm Under Regulatory
Constraint". American Economic Review. 52 (5): 1052-1069.
Exhibit No. 1
gase Nos. AVU-ts-l9-U_ and AVU-U-]9-U_
P. Ehrbar, Avista
Page 200 of 224
Page 7-16
6
Table 7-12. Rate of Return vs. Cost of Capital - DR 066, Revised, Attachment A
We see no current adverse impact on cost control and operational efficiency
Summary - Task 7 (Adverse Impacts)
We find no conclusive evidence of current adverse impact of decoupling on cost control,
operational efficiency, price signals or service quality. We have expressed two concerns for the
intermediate to long-term for two cost-control approaches: making hiring reviews more extensive
and so possibly creating some short-staffing problems over time; and moving away from defined
benefit pensions. We address these two concems in the Recommendations section.
Exhibit No. 'l
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page201 ot224
Washington Electric
2012 2013 2014 2015 2016 2017
Normalized Rate of Retum 7.16%7.57%7.92%7.33%7.33%7.34%
Authorized Rate of Return 7.91%7.64%1.64%7.32%7.29%7.29%
Normalized Return on Equity 8.10%9.90%r0.60%9.40%9.40%9.40%
Authorized Retum on Equity 10.20%9.80%9.80%blackbox 9.50%9.s0%
Washington Natural Gas
2012 2013 2014 2015 2016 2017
Normalized Rate of Retum s.44%6.23%5.79o/o 6.14%796%7.84%
Authorized Rate of Retum 7.9t%7.640/o 7.64%7.32%7.29%7.29%
Normalized Retum on Equity 5.20%7.20%6.40%7.00o/o t0.70%l0.40Yo
Authorized Retum on Equity r0.20%9.80o/o 9.80%blackbox 9.50%9.50%
Notes: The Authorized Rate ofRetum for 2015 has been corrected as per discussion in the presentatior/review meeting. The
number in the original table was 7.64; the corrected entry is 7.32. The term "blackbox" means the information is not available
because it is sealed by a settlement agreement.
Page 7-17
6
Exhibit No. 1
Case Nbs. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 202 ol 224
Page 7-18
O
Section 8. Low-Income A endix
The Avista Decoupling Evaluation RFP No. R-41321 provided two related Attachments to the
Scope of Work: Attachment G - An Estimate of the Number of Households in Poverty Served by
Avista Utilities in Washington Statel3a and Attachment H - The Self-Sufficiency Standard for
Washington State 2014.13s Attachment G provides an estimate of how many Avista customers
are below the Federal Poverty Level in counties served by Avista. Attachment H estimates the
level of income required by households to achieve self-sufficiency without public assistance. We
reviewed these two documents and correlated findings with the low-income energy assistance
information that we reviewed for Task 3. This Appendix summarizes findings.
Attachment G - Estimate of the Number of Households in Poverty
This study provides estimates of the number of Avista low-income customers in the State of
Washington. The estimates are based primarily on Census Tract data, particularly the American
Community Survey which provides counts of household at different poverty levels for each
census track. Within each tract, the study provides an estimated count of households with income
at or below five multiples of the Federal Poverty Level (FPL): 50oA,l25oh,l50Yo,l85o/o and
200%.
Table 8-l combines information from Attachment G with information provided in DR's related
to Task 3 and compares the number of Avista low-income customers served by one or more
energy assistance programs to the number of households estimated to be at or below 150% of the
pp1.t:0
The sources and descriptions of data for each of the columns in Table 8-l are presented below.137
o Columns (1,2) An Estimate of the Number of Households in Poverty Served by Avista
Utilities in Washington State. These are Census 5-year rolling estimates for the period 2009-
20t3.
o Column (3) An Estimate of the Number of Households in Poverty Served by Avista Utilities
in Washington State, is based on an estimate of the number of households at or below 150%
of the FPL as reported in Attachment G.
o Column (4) DR 47 A, is the average number of bill assistance grants from all funding
sources provided to Avista customers annually during the period 2012-2017.138
o Column (5) DR 49 A, is the average number of Avista Weatherization rebates annually
during the period 2012-2017.
t3a An Estimate of the Number of Households in Poverty Served by Avista Utilities in Washington State, Brian
Kennedy, MS and D. Patrick Jones, Ph.D., Institute for Public Policy and Economic Analysis, May 2015.
t35 The Sey-Sfficiency Standardfor Washington State 20l4,Diana M. Pearce, PhD, Center for Women's Welfare
and the School of Social Work at the University of Washington, Revised August 2015.
136 One hundred and fifty percent (150%) of the Federal Poverty Level (FPL) is the national LIHEAP eligibility
standard used in most states to determine eligibility for energy assistance.
r37 Responses to DR's: 047 Attach. A,036 Attach. A
r38 The data from Attachment G covered the period 2009-2013. Based on the data available in evaluation DRs for
columns (4) and (5) we used the average number of customers served over the 2012-2017 period. While the data
does not match chronologically, using averages helps to eliminate yearly variations.
Exhibit No. 'l
P. Ehrbar, Avista
Page 203 ol 224
Page 8-I
6
o Column (6): [Column (4) + Column (5)]/Column (3), an estimate of the percentage of
LIHEAP Eligible Households served by energy assistance and Avista Weatherization.r3e
Table 8-1. 150% of Poverty or Less - Receiving Bill Assistance or Avista Weatherization
(l)(2\(3)(4\(s)(6)
Countv
Estimated
Households
Avista
Residential
Customers
Estimated
Households
Eligible for
LIHEAP
Avista
Customers
Receiving Bill
Assistance
Avista Customers
Receiving
Weatherization
Assistance
7o of LIHEAP
Eligible Customers
Receiving Energy
Assistance
Adams 5,747 4,s40 t,692 399 8 24Yo
Asotin 9,052 9,294 2,264 848 32 39%
Ferry t,669 1,630 667 189 I 28o/o
Franklin 2,683 167 6l 0%
Grant 1.163 10 J 0%
Klamath NA NA NA I NA
Klickitat 3.656 763 263 2l I 9%
Lincoln 4.463 3.462 866 252 )29%
Shoshone NA NA NA 1 NA
Skamania 764 320 82 6 1 8%
Spokane 186,259 169,287 43,6t3 13,044 182 30%
Stevens 17,569 19,972 6,tt3 t,754 t7 29%
Whitman 16,630 17,437 7,322 1,040 15 14%
Total 249,657 226,882 62,946 17,553 260 2804
This analysis finds that on average approximately 28Yo of the estimated LIHEAP eligible
households (150% of Poverty or less) receive some type of energy assistance from one or more
of the following programs: LIRAP, LIHEAP, Project Share, MISC or Avista Low-income
Weatherization. The percentage of estimated LIHEAP eligible customers receiving assistance in
each county ranged from 8% to 38o/o.
Attachment H - The Self-Sufficiency Standard for Washington State 2014
This reportrao presents and analyzes the Self-Sufficiency Standard for Washington State in2014.
This measure describes how much income families of various sizes and composition need to
make ends meet without public or private assistance in each county of Washington State. The
Self-Sufficiency Standard is a measure of income adequacy based on the costs of basic needs for
working families: housing, child care, food, health care, transportation, and miscellaneous items,
as well as the cost of taxes and the impact of tax credits. The Standard is intended to provide a
more detailed, up-to-date, accurate, and comprehensive measure of economic well-being than the
Federal Poverty Level.
13e It should be noted that Avista customers receive weatherization assistance from other programs such as the US
Department of Energy Weatherization Assistance Program, which were not documented in this evaluation, since
these services are not tracked by Avista. See Avista Response to Data Request No. 029(l).
140 Pearce, Diana M., op cit.
Exhibit No. 1
Page 8-2 P. Ehrbar, Avista
Page 2O4 ol 224
6
We reviewed Attachment H and extracted the Self-Sufficiency Standard for the same 11 counties
analyzed for Attachment G above. Table 8-2 provides a summary of the percentage of the FPL
that a family would need to earn to achieve Self-Sufficiency in each of the 11 counties. This
percentage varies from a low of 17l%o to a high of 235o/o of FPL to achieve Self-Suffrciency,
depending on location and household composition.
Table 8-2. Self-Stfficiency Standard Expressed as a Percentoge of PoverQ
One Adult
One Preschooler
One Adult
One Preschooler
One School-Aee
Two Adults
One Preschooler
One School-Ase
County
Self-Sufficiencv Standard
Annual
Percentage of
Federal Poverty
Level (FPL)
Annual
Percentage of
Federal Poverty
Level (FPL)
Annual
Percentage of
Federal Poverty
Level (FPL)
Adams s30.449 l94yo s37,601 t90%$45.29s 190%
Asotin $29,993 t9t%$34,81 5 1760/,$42,s49 178%
Ferry $30,919 197%$43,738 221%$s0,680 212%
Franklin $35,210 2240 $46,078 233%$52,936 222Yo
Grant $32,229 205%$38,810 196%$46,6s3 196%
Klickitat $31,915 203%$44,088 223%$s0.998 214%
Lincoln $28,991 184%$33.80s 17 loA $41.563 174%
Skamania $33, I 87 2tt%$40.340 204%$47,776 200%
Spokane $36,023 229Yo $46,453 235%$53, l 36 223%
Stevens $34,009 216%$44,912 2270 $s l,80s 217%
Whitman $38,420 244%$48,209 244%$5s,ss2 233%
The variation of Washington's Self-Sufficiency Standard by county for each of three family
types is illustrated in Figure 8-1. While there is meaningful variation across both family types
and counties, results cluster somewhat above 200% of FPL. We can, conservatively, use 200%
of the FPL to estimate need. In a more rigorous approach, we would need to take both family
type and county directly into account, but since 200% is above the 150%o of FPL or lower
percentages used for some Avista low-income programs we can reasonably use 200% for
practical purposes. Attachment G provides an estimate of the number of Avista customers at or
below 200% of poverty as illustrated in Table 8-3.
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 205 of 224
Page 8-3
6
Self-Sufficiency as Percentage of FPL (2014)
(One Adult with One Preschooler)
300x
2W"
2W
15096
l@%
5@r
096
224% 22g% 744x
IIIIIIIIIII
C ,no *"d ."."s ".- C "*.""""".,-"o" ,rr....ud
Self-Sufficiency as Percentage of FPL (2014)
(One Adult, One Preschooler, One School-Age Childl
3@96
2@t6
150?6
1m96
5096
096
227j4. 233% 22316 235% 744%
IIIIIIIIIII
."o *$ .'d ..""" """ C i.odrC,,--" .,.r..i|"..."
Self-Sufficiency as Percentage of FPL (2014)
(Two Adults, One Preschooler, One School.Age Child)
2fl%ttzx 222% 214% 223% 29.4 233%
ilIilililI
2@16 l9o%
15096
1m96
s@6
o%
C *S -'d ."""s "". C "..."""" "-""o
.,."..5.*d
Figure 8-1. Variation of Self-Stfficiency Standard across Washington Counties
Exhibit No. 1
Case Nos. AVU-E:1g.0rnilAVU:Gl EFo-
P. Ehrbar, Avista
Page 206 of 224
Page 8-4
6
In Attachment G, using calculations based on the American Community Survey, Kennedy and
Jones estimate that, on average, thirty-seven and one-half percent (37.5%), of Avista Customers
are at or below 200% of FPL (Table 8-31.tat
During the period 2012-2017, bill assistance or Avista Weatherization services were provided to
17,813 customers per year.ra2 Based on the Self-Sufficiency Standard model this service record
comprises about twenty-one percent (21%) of the 85,159 Avista customers whose incomes are at
or below the Self-Sufficiency Standard, when approximated at200o/o of the FPL.
Table 8-3. Results at 20094 Poverv based on American Communitv Survev
Countv
American Community
Survey Estimated
Households
Total Avista
Customers
(Households)
Estimated Avista
Customers:
200yoFPL
Estimated Share of
Avista Customers:
2000/0FPL
Adams 5.747 4.540 2.310 s0.90%
Asotin 9.052 9.294 3.488 37.s0%
Ferry 1.669 1.630 813 49.90%
Franklin 2.683 167 85 51.10%
Grant 1.163 10 5 49.80%
Klickitat 3.6s6 763 376 49.20%
Lincoln 4,463 3.462 1,242 3s.90%
Skamania 764 320 100 31.30%
Spokane 186,259 169.287 59,532 35.20%
Stevens 17,569 19.972 8,412 42.10%
Whitman 16,630 17,437 8,796 50.40%
Total 249,657 226,882 85, I 59 37.50%
Making Sense of Federal Poverty Level vs. Income Insufficiency
Pearce compares several "benchmarks of income", including the Self-Sufficiency Wage, Welfare
(TANF, SNAP & WIC), the Federal Poverty Level, the full+ime minimum wage for Washington
and the Department of Housing and Urban Development Income Limits for three levels of low-
income (the top level is the highest income eligible for federal housing assistance: 80% of area
median income; in addition, there is a Low-income Limit and a Very Low-income Limit). Each
of these is a separate indicator that a household is in a situation of income difficulty.la3
Of these benchmarks, the most used in the United States is a multiple of the federal poverty level
(FPL), yet this is also one of the most challenged indicators. The fact that almost no agency uses
the FPL, but, instead, agencies use a multiple of the FPL for program eligibility suggests that
problems with the FPL are universally recognized. The FPL was created using 1950s data in the
early 1960s. It assumes a stereotypical 1950s family with a single wage earner and a full+ime
unwaged person at home to do the work of raising children, housework, and meal preparation.
In the 1950s, one wage eamer could typically support a family, unlike today when it usually
takes two fulltime workers to earn slightly more than one worker earned in the 1950s, accounting
tat An Estimate of the Number of Households in Poverty Served by Avista Utilities in Washington State, Brian
Kennedy, MS and D. Patrick Jones, Ph.D., Institute for Public Policy and Economic Analysis, May 201 5, page 7 .
r42 This is the sum of totals for columns 4 and 5 in Table 8-1.
ra3 Pearce, Diana M, op cit., Pp.28-29.
Exhibit No. 1
P. Ehrbar, Avista
Page207 of224
Page 8-5
6
for inflation.raa In low-income families, typically older children also do part-time work to bring
in money for the household and (for some) volunteer for the armed services when they become
of age in order to be able to send money back to their parents and keep their family viable. Also,
as pointed out by Pearce, the official FPL was based on a single indicator (the cost of the lowest
level of food that could sustain a family), which was then multiplied by the number three. Each
year, this highly flawed indicatorras is adjusted for inflation using one of the Bureau of Labor
Statistics (BLS) consumer price indexes (CPIs). This type of adjustment is itself flawed because
the BLS CPI seriously underestimates inflation over a period of years. The outcome is a
severely underestimated benchmark sequentially adjusted each year by a flawed multiplier, so it
is often argued that the FPL is severely flawed. Indeed, the Census Bureau itself states, "the
official poverty measure should be interpreted as a statistical yardstick rather than as a complete
description of what people and families need to live."146
In contrast, the US Department of Housing and Urban Development benchmark of 80% of area
median income automatically adjusts each year as incomes change,raT though it is sensitive only
to the median of the income dishibution and not sensitive to the increasingly severe income
inequality that we experience.
The most well-grounded method is the Self-Suffrciency Standard benchmark used by Pearce and
developed jointly by Wider Opportunities for Women and the Ford Foundation. This method is
the current version ofthe household budget approach in use by social workers for the past one-
hundred years. It is updated every few years by changes to the costs of items required by
households for a lower-moderate level of living and is based on family size and the ages of
persons in the household. Table 8-4 illustrates the specific items that comprise the Washington
Self-Sufficiency Standard for Spokane County in20l4.ras Pearce has calculated a specific Self-
Sufficiency Standard for each county in Washington State. These studies are repeated
approximately every three years.
If we were to use the Poverty Guidelines (only) for Spokane County in 2001, one-hundred and
fifty percent (150%) of poverty for a single adult is $12,885. In2017, it is $18,090. This is an
increase of about 140%between 2001 and2017 (Table 8-5). If we were to use the Self-
Sufficiency Standard (only), for Spokane County in 2001, the standard for a single adult is
$14,910. For 2017, it is $18,972. This is an increase of about l27yo, yet there is another factor
raa Though disposable income is less for today's two-income families than it was for counterpart single-income
families in the 1950s.
ta5 Highly flawed since based on a single indicator and because the diet selected is no longer available and since the
food items required several hours of work to make the food edible. It was a good effort for the time; there was no
official poverty indicator before this.
la6 Carmen DeNavas-Walt, Bemadette Proctor, and Jessica C. Smith, "Income, Poverty, and Health Insurance
Coverage in the U.S.: 2012," U.S. Census Bureau, Current Population Reports, Series P60-245, Washington, D.C.
(U.S. Government Printing Office), http://www.census.gov/prod/2013pubs/p60-245.pdf (accessed June 24,2014).
ta7 Due to a long-term shortage of public housing, although the upper eligibility limit is 80o/o of area median income,
most apartments that become available are assigned to households with lower incomes.
148 Pearce, Diana M, op cit., P. 103.
Exhibit No. 1
Page 8-6 P. Ehrbar, Avista
Page 208 of 224
6
to take in to account: the amounts for both 2001 and 2017 are higher for the Self-Sufficiency
Standard than for the Poverty Guidelines.
While there is not much difference for a single adult, the real strength of the Self-Sufficiency
Standard is shown in the remaining columns of these tables. The Self-Sufficiency Standard takes
in to account, not only family size, but also ages of household members and it is based on actual
cost of essential items for a specificyear. The size of the gap between these two methods is
about ten percent (10%) for the single adult in 2017,frfty-six percent (56%) for a household with
one adult and one preschooler, and about fifty-two percent (52%) for a household with two
adults, one preschooler and one school-age child. As has been noted by Pearce, the relative
failure of CPI measured inflation is demonstrated in the method's inability to capture the actual
differences measured in the Self Sufficiency Standard approach.lae
The Washington Self Sufficiency Standard is based on the family budget method and is updated
every three years to capture data on changes to the costs of items required by households,
characterized by family structure and the age of household members. The Standard is based on
achieving a lower-moderate level of living and is calculated at the county level. In contrast,
federal poverty guidelines, though based on the number of members of a household, are not
based on family structure and not age adjusted or based on county-level costs. The CPI tends to
lack adequate information while the Self Sufficiency Standard does not.
Table 8-4. Monthly Costs included in tlte Self-Sufficienct,Standat"d - Spokane 2014
r4e Pearce, Diana M., Attachment H - The Self-Sufiiciency Standard for Washington State, 2014, op cit., P . 27
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 209 of 224
MONTHLY COSTS Adult
Adult +
Preschooler
Adult +
Infant
Preschooler
Adult +
Preschooler
School-ase
Adult +
School-age
Teenaser
2 Adults +
!nfant
2 Adults +
Preschooler
School-ase
2 Adults +
Infant
Preschooler
School-ase
Housing $57 1 s773 $773 $773 $773 $773 $773 $ 1,l0s
Child Care $o $692 91,492 $r.224 $532 $800 $r.224 $2.024
Food s24s $371 $487 $s60 $641 $s93 $768 $8s0
Transportation $2s7 s266 s266 $266 $266 $507 $s07 $s07
Health Care $l r3 $392 $405 $41 0 $439 s4s l $467 $479
Miscellaneous $l l9 $249 $342 $323 s266 $3 l2 $374 $497
Taxes $ 189 $43s $654 $59 I s365 $sl3 $61 s $93s
Eamed Income Tax Credit G)$0 ($33)$0 s0 ($r 57)$0 $0 $0
Child Care
Tax Credit G)$0 ($60)($ I oo)(s1oo)(s63)($s0)($ I 00)($ l 00)
Child Tax Credit (-)$0 ($83)($ l 67)($167)(s l 67)($83)(s I 67)($2s0)
SELF.SUFFICIENCY WAGE
HOURLY $8.49 $ 17.06 $23.s9 $22.0s $16.49 $10.84
per adult
sl2.67
oer adult
$1 7.1 8
oer adult
MONTHLY $ l,494 $3,002 $4, l s2 $3,88 l $2,903 s3,8 l 6 $4,461 $6,047
ANNUAL $17,923 $36,023 $49,82s $46,s73 $34,830 $4s,796 $s3,s32 $72,s64
EMERGENCY SAVINGS
(Monthlv Contribution)$36 s8l s 109 $l0s $9s $50 s6l s79
Page 8-7
o
Table 8-5. 150% Poverty Guidelines (2001 vs. 2017)
Independent of County (2001 vs. 2017)
1507o Poverty Guidelines (Onlv)
Year Sinele Adult
One Adult with
One Preschooler
Two Adults with
One Preschooler and
One School-Ase Child
2001 $l 2,88s $17,41s $26,47s
2017 $r 8,090 $24,360 $36,900
Percent Change t40%l40o/o 139%
Table 8-6. Self-Sfficiency Standard Spokane County (2001 vs. 2017)
Spokane County (2001 vs. 2017)
Self-Sufficiency Standard (Onlv)
Year Sinele Adult
One Adult with
Preschooler
Two Adults with
One Preschooler and
One School-Ase Cbild
2001 $ 14.93 0 $25.094 $39.428
2017 $ I 8.972 $38. l 03 $56.010
Percent Change 127%ts2%t42%
Table 8-7. 150% of FPL vs. Self-Sfficiency Standard, Spokane County, 2001
Spokane County (2001)
1507o Poverty Guidelines vs. Self-SuIIiciency Standard
Calculation Method Sinele Adult
One Adult with
Preschooler
Two Adults with
One Preschooler and
One School-Aee Child
I50% FPL $ 1 2.885 sl7.4l s $26,475
Self-Sufficiency Standard $14,930 $25.094 $39,428
Percent Difference tt6%144%149%
Table 8-8. I 50% of FPL vs. Self-Sfficiency Standard, Spokane County, 2017
Spokane County (2017)
1507o Povertv Guidelines vs. Self-SuIIiciency Standard
Calculation Method Single Adult
One Adult with
Preschooler
Two Adults with
One Preschooler and
One School-Ase Child
I 50% FPL $ 1 8.090 $24.360 $36.900
Self-Suffi ciency Standard $18,972 $3 8,1 03 $s6,010
Percent Difference 105Yo 156%152%
A useful analysis of what happened to the CPI is provided by ShadowStats (Figure 8-2). In this
figure, the top line (blue) is the ShadowStats CPI and the bottom line (red) is the BLS CPI. Note
that the two measures are nearly identical until about 1983 at which point they begin to diverge.
The two curves continue with very similar shapes, except for the growing spread of vertical
distance between comparable points on each curve. The Shadowstats CPI continues the original
method of the BLS CPI (and the method for a price index as described in older economic
textbooks). Changes in the original BLS CPI method were introduced gradually under both
Republican and Democrat administrations. These changes have academic explanations yet tend
to move the indexed inflation down, having the effect of making things look better than they
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 21O of 224
Page 8-8
6
are.l50 They function to lower social security increases, wage increases indexed to the BLS CPI
and other government program expenditures tied to the CPI. The latest BLS irurovation is
movement towards a "chained CPI" which used geometric rather than arithmetic means. This
will also make the CPI register weaker inflation than that known to the population through lived
experience.
Figure 8-2. Historical Divergence of BLS CPI (Courtesy of ShadowStats.com)
Level of Rigor
These differences in methods have several implications in the estimation of the number of low-
income customers. Table 8-1 suggests that about twenty-eight percent (28%) of Avista's
residential service population is low-income, based on the one-hundred and fifty percent (150%)
of poverty level criterion, as in most states. Table 8-3 shows that if a two-hundred percent
(200%) of poverty criterion is chosen, the result is about thirty-seven percent (37.5%) ot
residential customers. The Self Sufficiency Standard approach tends to center on two-hundred
percent (200%) though it varies with family type and by county. In Table 8-2, values range from
l74o/o to 233% depending on county and family type.
The result in the number (and percentage) of low-income households in Avista's service territory
depends on the method of analysis selected. Selection of method depends on a choice of level of
rigor. Most utilities simply go with a percentage like one-hundred and fifty percent (150%) of
poverty because it is simple. It is administratively convenient since the appropriate poverty
150 (http://wrvw.shadorvstats.com/alternate data/inflation-charts) ShadowStats charts must be published without
modification in any way and must contain, under the chart, "Courtesy of ShadowStats.com" See also: Boring,
Perrianne, "If You Want to Know the Real Rate of Inflation, Don't Bother with the CPI", Forbes, February 3,2014
(https://wwrv. tbrbes.conr/sites/perianneboring/20 I 4/02/03/if--) ou-want-to-know-the-real-rate-of-inflation-dont-
bother-with-the-cpi/#47059396200b). For an opposing perspective, see Greenlees, John S. and Robert B.
McClelland, "Addressing Misconceptions about the Consumer Price Index." Monthly Labor Review, August 2008,
Pp. 3- I 9 (https://rvww.bl s.eov/opub/mlr/2008/08/aft I f u ll.pd0.
Exhibit No. 1
Page 8-9 Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page211 of224
Consumerlnflation - Ofricial vs ShadowStats (1980-Based) Atternate
Year to Year Change. Through June 2018 (BLS, SGS)
- sGS Atternste cPt, ,1980-Eased - cPt-u
t9t? tttS {ttt t9tt tgtf itg, 2c00 2003 ?00t 2009 20t2 20t5 20tt
Pubtishef Juty 1?- 20lg -liar/owStirts.tont
r5%
10%-
5%-
6
numbers and program guidelines are published each year in the Federal Register and a multiple
of Poverty can be easily implemented.
A middle level of rigor would look more closely at the variations from a textbook approach in
calculating the BLS CPI and choose, instead, the ShadowStats CPI (which is proprietary but easy
to access by crossing a paywall). Or, by melding the BLS CPI and the Shadowstats CPI using a
simple ratio following a study of both methods. This approach would offer the same
administrative convenience as a low rigor approach but would be more accurate.
A high level of rigor would use neither the official definition of Poverty based on the original
flawed analysis and flawed updates produced by the government using the BLS CPI (as modihed
away from original BLS practice and textbook method many times). A high level of rigor would
begin with the existing work on the Self Sufficiency Standard, calculated and updated for
Washington approximately every three years by the Center for Women's Health at the University
of Washington School of Social Work. This is the most truthful and realistic method. However,
it would require calculation by county and it would be tailored to family structure by ages of
household members and not only to family size. Strictly, it would have to be administratively
applied at a county level, and provision of different levels of eligibility by county could be an
administrative concern. The problem is not just optics, but, for example, households located near
county borders or other possible needs for exceptions. However, if this high-rigor method were
used for analysis, an administrative simplification could be employed for program
administration.
The implication of this analysis is that more households need help than are indicated by the
Poverty Guidelines as adjusted by the BLS CPI. We recommend using the using the Self
Sufficiency Standard. However, we are aware that rigor in analysis might need to be
accompanied by simplification to meet the needs of program administration.
At the same time, in evolving the strucfure and scope of payment assistance and weatherization
assistance, the cost to customers providing the assistance must be considered and balanced.
Customers just above the cutoff for eligibility are in essentially the same financial bind as
customers eligible for assistance, so attention could be focused on "feathering out" assistance at
the top of the eligibility range, or to exempting from tariffs that support assistance to low-income
customers those customers who are in income groups just above the eligibility range.
Exhibit No. 1
Page 8-10 uase Nos. AVU-E-]9-U_ and AVU-G-]9-U_
P. Ehrbar, Avista
Page 212 of 224
6
Understanding Low-Income within the Overall Allocation of Income
If we consider the allocation of income for Washington, the income donut shown in Figure 8-3
provides an image that is easy to remember. This is the income donut for 1990, computed from
census data.151 For comparison, the income donut for 2000 was computedl52 and is shown in
Figure 8-4.
If we compare the two donuts, we see that income for the upper twenty percent (20%) ot
households by income moved up by eight percent (8%) from 1990 to 2000. The bottom twenty
percent of households dropped from five percent (5%) to four percent @%\ The lower middle
dropped one percent (l%), the middle five percent (5%) and the upper middle dropped trvo
percent (2%).
8.1t0fr 209t, 5%
Lo*r Mirdlc. t046
ToD 20%. 13%
Mirdb 20% l8%
Upp.r Miidlc. 25%
lIoD2fr
oud& 2ft
O&no6 ft*
Figure 8-3. lnconte Donutfor l4/ashington State (Censtts 1990)
From the end of WWII through the early 1970s, the United States increasingly took on many
characteristics of an economic democracy as income shares increased throughout most of society
and shares to upper income groups dropped; for example, the upper one percent (1%) lost
income share during this era. From about 1970 or l972,the process reversed, and income flow
has concentrated more and more toward the very top of the distribution of income to households.
Within the upper flve percent (5%) this flow to the top repeats very strongly; within the upper
lYo the pattern again repeats but more intensely.
The two income donuts shown only indicate a little of this change. However, income inequality
is increasing dramatically. As suggested by the two figures presented, income share is taken
from the boffom through the upper middle and transferred into the top quintile. However, within
r5r Source: Columns I and2 from Table P080, Household Income in 1989, Census 1990 Summary Tape File 3 -
Sample Data.
r52 Source: Columns I and2 from Table P52, Household Income in 1999, Census 2000 Summary File 3 - Sample
Data.
Exhibit No. 1
P. Ehrbar, Avista
Page 213 ol 224
Page 8-1 I
0
the top quintile the same pattern of extraction and allocation occurs with income moving from
the lower parts of the top quintile to the upper one percent (1%).
Fropzo*-l!!%,
tmdde eot6]Lr3% l
Uppcr Uaddlc
23%
tTop zo%
aUpp.rt lddl.otiddl. 20%oLoHrf,lddl.
oBotom 20%
Figure 8-4. Income Donutfor l4/ashington State (Census 2000)
This pattern of income allocation creates a dilemma for providing support for low-income
households, since income share is being taken from those households that would normally have
been able to support some form of low-income assistance in the past. This is a dilemma for
funding low-income weatherization and payment assistance and should be taken in to account in
informing development of a low-income rate. Balance is very important.
Exhibit No. 1
P. Ehrbar, Avista
Page 214 ot 224
Page 8-12
Lowr Maddle
Eottm 20PA
/t%
Section 9. Weather A dix
6
Everyone knows the weather is changing. The NW Climate Hubls3 has issued a drought forecast
(Figure 9-l) as of July 31,2018, beginning in August 2018. The forecast includes a map of
potential wildland fire areas (Figure 9-2). While these projections become a quickly dated and
one-time forecast, they report on an underlying change in the weather. The projections are
consistent with rapid (in geologic time) climate warming. Nearly every year now, there is more
warrn weather, including warm evenings. The trees from California up through British
Columbia (and over to Colorado and Utah) are stressed and thousands are dying. The "new
normal" is a warming trend with statistical fluctuation. The "new normal" also is a process
(flow) variable - it is not static, but moving. It is getting warmer and warmer and there is no
apparent end to the warming on a typical human scale of time.
A combination of high temperatures, low humidity, and dry to record-dry conditions has
increased fire danger.
. Wildfires continue to threaten lives, property, crops, rangeland, and forests.
. Drier-than-normal conditions are expected to continue across most of the region, which will
perpetuate fire danger. CURRENT CONDITIONS
. OR and WA have been experiencing dry weather. Combined with high temperatures, this led to
the designation of moderate drought in the Olympic Peninsula, abnormal dryness in parts of
eastem WA, and the introduction of severe drought across the Cascades and into the Willamette
Valley last week. Southem ID and the panhandle are abnormally dry with some areas of
moderate drought.
. According to the Northwest River Forecast Center, monthly precipitation through July 30,
2018, is below 50% of normal. Over the last 90 days, precipitation totals for parts of western OR
and WA were the lowest they've been in at least 40 years.
Figure 9-1. Drought Conditions
rs3 https://www.drought.gov/drought/sites/drought.gov.drought/files/StatusUpdate PNW July3 I Final.pdf.
Exhibit No. 1
P. Ehrbar, Avista
Page 215 of 224
Page 9-l
e
Significant Wildland Fire Potential Outlook
August
q
'tE,j:
Slg.t0crnt mldLnd FIE Pobnd.l!u*xom -ffi"-l*lmr _*ffid*&..BeI lt66d _ shbsffi m@
M.|l,@Sler tll
Figure 9-2. Wildland Fire Potential Outlook
Within this context of changing weather, the first thing to note in the two figures below (Figure
9-3 and Figure 9-4) is the increasing prevalence of warm years with fewer heating degree days
and more cooling degree days. The orange bars denote years that are warner than normal.
Although there is statistical variation, the orange bars are mostly strongff than the blue bars and
are increasingly frequent. Occasional years with more heating degree days occur, but years with
more heating degree days are becoming scarcer. The bars each represent the difference in
heating or cooling degree days to a base of 65' Fahrenheit, calculated using a rolling thirty-year
average (normal) weather. I 5a
Figure 9-3. Pattern of Heating Degree Days (Spokane)
r5a Beginning in 1947 values are from the Spokane airport (GEG) weather station. Values in Figure 9-3 and Figure
9-4 run from I 976 through 2017 (a range of 42 calendar years).
Exhibit No. 1
L;ase Nos. AVU-ts-]9-U_ ano AVU-U-]V-U_
P. Ehrbar, Avista
Page 216 of 224
\
Blue bars denote colder than normal
Orange bars denote warmer than normal
ltl,l,,ll-'tl't'll,rl ;p
l,l
u
Ei :r.1, r
ig -.' tl l 'll'"
$ ,,0
-10@
-1250 @N@oO<d6l6Ots @6 O<ddl 4ONAOoidd{6qts @6Ordd!5OFh5h58BB88ttBtB 8Se8 8SS88888888888885565E556iiii NdNddNdNdNNdddNNNd
Page 9-2
6
Sequences of Warm and Cold Years
Looking at Figure 9-3 or Figure 9-4,the frequency of cold years is decreasing, but also warrn
years tend to run in series and their values are becoming stronger, while cool years run in short
blocks of one or two years and their strength is becoming weaker (as indicated by the length of
the bars).
For decoupling designs, this pattern is important. In the abstract, we might think of a deferral
mechanism as easily balancing over two years if the pattern of years is alternately warm and
cold. But since warm years are occurring in runs and the runs are appearing longer for warm
years (as well as warn years becoming stronger), this factor should be considered in decoupling
design in relation to defeating any "snowballing" effect, especially for natural gas rate groups. If
the pattern holds, we can expect declining need for heating in Winter. Avista's decoupling
design is special in that it allows for ratchetting decoupling rates to amortize higher levels of
deferral balances (it works on incremental changes); a good design feature. A practical
implication of this ratcheffing will be decoupling rates that may look high as a percent of total
revenue (exceed the three percent (3Yo) cap, since the mechanism works incrementally each
year), until the rates reset following a normal or colder than normal year or in the next rate
case. I ss
Figure 9-4. Pattern of Cooling Degree Days (Spokane)
lss In this paragraph, we use "normal" in the "old normal" sense of a 30-year moving average rather than in the sense
of the recent flow of the "new normal" which might be based on fifteen years or most recent seven years, for
example. The "old normal" is a flow variable, as is the "new normal". From a mathematical perspective, the rate of
flow increases substantially in a smaller set of most recent years. The mathematics reflects physical change.
Exhibit No. 1
uase Nos. AVU-ts-l9-U_ and AVU-G-I9-U_
P. Ehrbar, Avista
Page 217 ot 224
1,ilh,|ll l'l l-r r r -r r r I Illl-,lrll"
350
3m
250
6' zmo_oPB1!oEO!E'*-!:; so'=o:B
50Eg *
oI '1oo
-150
-200
-250 6tsFOOi4O!49FO6qddill6QNFOOHRdl6@NS6Odd6!6ONh5hBESSEEBSSBBS888BEESS88tttttt88t65653556ddNddNdddRddNddddN
Blue bars denote colder than normal
Orange bars denote warmer than normal
Page 9-3
Zero Heating Degree Days
Using data from the Spokane airport weather station (GEG), we can project the approximate year
when there will be zero heating degree days (HDD). The practical implication of an indicator
that tends towards zero HDD is that the need to turn on heat for buildings tends towards zero. In
a simple regression of HDD on year, beginning in 1947 (when Spokane's weather station was
moved to the airport), it is easy to see that HDD is declining over time (Figure 9-5). Using the
parameter estimates from Table 9-1, we get a constant of 20,890 and a slope of -7.120. Using
the standard equation of:
Y:mx*b,
Or, in this application:
HDD: (-7.120XYEAR) + 20,890
g: (-7.129)YEAR + 20,890
(7 .129)(YEAR : 20,890
YEAR : (20,890)/ (7,129)
YEAR:2934
Solving for the case in which HDD : 0, we get the year 2934
2934 -2018:9t6
Or, about 916 years from now.
Table 9-1. Model Summary and Parameter Estimates
Model Summary and Parameter Estimates
Dep,lndentVariahle: HDD65
Equation R Square F
Model Summary
dfl df:sis
Parameter Estimates
Conslant b1
Linear 101 7.750 1 6S .007 208s0.153 -7.120
The independent variahle is Year,
This is a very conservative estimate, since we use airport data rather than a carefully developed
climate model. Also, since the strength of the climate warming has shown itself only since about
the year 2000, datafrom 1947 (the year our data series begins) is likely not relevant. In fact,
even the "old normal" method would employ 30 years of data, rather thanTl years.
Exhibit No. 1
o
Page 9-4 Case No-TVU-E:f9:0_ a nt AVU-G-1 9-0_
P. Ehrbar, Avista
Page 218 ol 224
6
oo
o
o
ooooo $
o
o
ooooooo& ooo
o
oooo
oo oooo
o
o
oo Ogo
o oo
o o
I
oo
c o
oo
oo
o
HDD65
1
Year
Figure 9-5. Regression of HDD on Year
If we reduce the years in the analysis to the 18 most culrent and re-run the analysis using the
airport data beginning in 2000, the year in which HDD is zero is 2175 (or 157 years from now).
lf we re-run using only the 8 most recent years beginning in 2010, the year in which HDD is zero
is 2104 (86 years from now). Figure 9-6 and Figure 9-7 show these relationships for different
numbers of analysis years, reaching back from the most current data which is for calendar 2017.
Year in which HDD = Zero is Rcachcd
Number of Most Recent Weathcr Years included in the Analysis
Figure 9-6. Year in which HDD : Zero is Reached, Using dffirent Nttntbers of Analysis Years
Exhibit No. 1
o
!Eo
(,
Case Nos. AVU-E-19-0_ and AVU-G-19-o-
P. Ehrbar, Avista
Page 219 ol 224
Page 9-5
6
ooNt
ooI
5
Iz
,o
.EE
E
d.
Number of Most Recent Weather Years included in thc Analysis
Figure 9-7. Years from 2018 until Zero HDD, using dffirent Numbers of Analysis Years
We need to note that these are only the results of standard regression analysis and not science.
Climate scientists tend to be very careful and conservative and do not like to project for more
than about 100 years since the error bands around their results increase with time and there may
be points of inflection and dialectical oppositions that are not yet well understood.ls6 However,
we are in a constantly moving new nornal and these estimates are an attempt developing useful
indicators rather than science.lsT The range of 916 to 86 years is a large range (note that we
have not provided error bands). Yet a very big thing is happening, irreversible on a typical
human scale. And, the reason for looking at most recent years is connected to physical
phenomena with an increasing flow rates. So, how one interprets these numbers and these
calculations depends on one's sense ofphysics.
While science must be quite conservative almost all of the time, persons with business sense and
those with responsibility for public administration must be more practical so as to be aware in
advance of things "hidden in plain sight." We suggest these calculations be considered as
indicators, each with a different number of data points (calendar years of weather information
from past years). Each of the indicators can be calculated each year so as to form a data series
156 An example of dialectical tension is that physical constants such as the estimate of 100 years for carbon (as a
generic for greenhouse gas) to reach a sink (or 20 years for fugitive methane) are unlikely to hold as sinks become
overloaded. Vegetation as a source ofcarbon sequestering is expected to reverse at some point and become a carbon
source (for example, from forest fires as trees and grass are increasingly stressed). Another tension is the
expectation that primary ocean currents may change. Another is that air rivers have changed and are continuing to
change, altering the behavior of hurricanes and rain storms. Another is the loss of snow cover which shifts wide
areas from reflection to absorption. Dialectical analysis is required to take these kinds of factors into account.
157 Why doesn't science give us more certain answers to our weather questions? Because it is young and
underdeveloped. If we date modern science somewhat conservatively from the date of founding of the Royal
Society of London for Irnproving Natural Knowledge in 1660, that is only 358 years ago, essentially a blink of the
eye. To help with understanding tirne, the Long Now foundation advocates thinking in I 0,000-year blocks and
would write the founding year as 01660, r.vhile this report is submitted in 02018. If or.re thinks in a 10,000-year
block,thensciencein020l8isessentiallynewandprimitive. However,asysternofrnovingindicatorsmaybe
relevant for organizational decision n.raking.
Exhibit No. 1
uase Nos. AVU-E-I9-U_ and AVU-U-I9-U_
P. Ehrbar, Avista
Page 22O ol 224
Page 9-6
6
constructed as a moving average in the same way that the traditional 3O-year "normal" is
calculated.
It is not good enough to revert to the 30-year normal. Clearly, the curves in Figure 9-6 and
Figure 9-7 show fluctuation and this should be considered; but they also show an increasing
tendency to bring the zero HDD year rapidly closer in time. For practical decisions, the
decision-maker might maintain and review each indicator and act on those that appear most
relevant to the purpose at hand. This analysis suggests that the 30-year normal is no longer a
useful indicator. It is not a good indicator of the moving new normal.
We suggest, for now, running 30 years, 20 years, 15 years and 10 years and developing the
curves for these indicators and then carrying the indicators into the future. We suggest that the
2}-year indicator is the right one to rely on right now, that the 3O-year indicator is not a good fit
right now due to systematic changes in the weather (climate warming), and the l5-year and 10-
year indicator will be more sensitive but also less stable than the 2}-year indicator. The 30-year
and the 2}-year indicators will, of course, get better over time assuming the climate tum is the
"new normal" and more and more warm years replace the cooler years at the beginning of each
moving average. Figure 9-8 shows that the 2}-year,l5-year, and l0-year averages are quicker to
register the decline in HDD than the 30-year measure, though as the downward trend in HDD
continues, the curves are converging."t
ThirV Yca. Aycragc vt. Otfi ar Avcragls
Hc.ting Dcgr.! Oayr from 1976 FoMard
o
o
E
C.l.nd.r_Yaaa
Wedhsr d:tion hes be6n locA.d at GEG since 1947
0a.3eledion insures.lldd. i3 from 19{7 fomd.
Figure 9-8. Thirty Yeor Average vs. Other Averages for HDD
An implication for Demand-Side Management is that the effect of going to a2}-year moving
average will be to create stronger cost-effectiveness results for cooling measures and somewhat
weaker cost-effectiveness results for heating measures.
1s8 See also: Drury, Matt and Mallorie Gattie-Garza, "Climate Change and its Effect on Weather Data". Pp, 9-I to
9-l I in Proceedings of the 2016 American Councilfor an Energt Efiicient Economy Summer Study on Energt
Efiiciency in Buildings. Washington, DC: ACEEE,20l6. Drury and Gattie-Garza suggest applying simple
regression analysis to project HDD and CDD over the life of a DSM project rather than use backward looking
weather normalization averages. Projections based on regression models may be more useful than weather
normalization by means of backwards-looking moving averages.
Exhibit No. 1
uase Nos. AVU-tr- tv-u_ ano /\VU-u- tv-u_
P. Ehrbar, Avista
Page 221 ot 224
Page 9-7
6
Exhibit No. 1
Case No-TVU:FI g-0_ and AVU-G-1 9-0_
P. Ehrbar, Avista
Page 222 of 224
Page 9-8
Section 10. Recommendations
(l) The decoupling mechanisms have worked as expected to stabilize revenue without
impacting utility operations and energy efficiency programs. We also found no
evidence of adverse impacts to any customer groups. We recommend the electric and
natural gas mechanisms be continued and certain modifications be considered.
(2) If practical for Avista, move the decoupling tariff effective date up from November lst
to July 1st to substantially increase the likelihood that reported revenue will be collected
within two years, as required by the Securities and Exchange Commission.
(3) Avista might consider adjusting the low-income "carve out" each year for inflation to
keep its value more stable between rate cases.
(4) We have a sense that staffing is a bit thin compared with other utility clients with whom
we recently have been engaged for projects. What works as a short-run cost savings
may not work as well long-term. We recommend consideration of some additional
hiring of some additional staff in Rates and in DSM (not short-term supplementary or
temporary arrangements).
(5) We notice that as a cost savings measure, Avista has moved from a defined benefit
pension system to a system that puts employees at individual risk in developing funding
for retirement. We agree this will represent cost-savings in the short term. Although
such change is currently viewed as normal in the industry, reflecting the market in this
case may not be useful long-term. Thinking of the five most recent "crashes" including
the recent "Great Recession", Avista might want to consider a plan that would enable
some form of pension that places institutional strength between employees as individual
"nano-investors" and market forces.
(6)Continue to work towards a possible low-income rate. Households in need of income to
meet the expectations of American households prior to the income allocation reversal
that began in the early 1970s, are likely about one-half of residential households (or at
least37.5%o, as shown in the low-income appendix). A low-income rate would provide
an additional tool to maintain service for all customers.
(7)In the low-income area, consider either moving to a higher level of rigor in evaluation
and program administration by using the Self-Sufficiency standard; or use the 200o/o of
the Federal Poverty Level as the program guideline for need for program payment
assistance and weathe rization services.
(8) Consider a redefinition of normal weather that moves away from the 30-year moving
average to a20-year moving average, and also maintain a moving average indicator for
15 years and l0 years to see how that behaves empirically, since "normal" has become a
flow variable and it is rapidly getting wanner as a secular trend.
Exhibit No. 1
o
Page l0-1 Case Nos. AVU-E-19-0_ and AVU-G-19-0_
P. Ehrbar, Avista
Page 223 of 224
Avista Decoupling Evaluation
H. Gil Peach & Associates, LLC
16232 NW Oakhills Drive
Beaverton, Oregon 97 006
(s03) 64s-0716
[e-mail]
www.peachandassoci ates. net
Exhibit No. 1
Case Nos. AVU-E-19-0_ and AVU-G-19-0-
P. Ehrbar, Avista
Page 224 ol 224