1

Proposed Input Assumptions to RTFCost-Effectiveness Determinations

February 2, 2010

Overview of Council Conservation Analysis Methodology

(1) Build Supply Curves

(2) Schedule

Availability

(3) Adjust NLO Supply Curve for Program Deployment

(4) Shape Savings by

Season & Hi/Lo

(5) Regional Portfolio

Model

(6) Strategy for Least-Cost & Least-Risk

(Cons Market Price Adder)

(7) Conservation

Build-Out over 750 futures

(8) Conservation

Targets & Action Plan

RTF >Translate Targets into Cost-Effective Portfolio of Measures and ProgramsRTF >Translate Targets into Cost-Effective Portfolio of Measures and Programs

3

Objectives of this “Translation”

• Establish a cost-effectiveness limit for conservation that:– Is high enough to secure sufficient resources to meet

near-term and long-term conservation targets– Is low enough to avoid acquiring conservation

resources that have a high risk of costing more than 110% of generating resources over their expected lives (i.e., accounts for risk)

– Is shaped like market prices so that savings are appropriately valued

4

Alternative Approaches

1. Use “110 % of Avoided Resource” cost to establish maximum levelized cost for conservation

Problems• Assumes perfect foresight• Not “shaped”

2. Use average amount of conservation acquired by RPM and supply curves to establish maximum levelized cost for conservation

Problems• Assumes all conservation is shaped the same• May hide underlying “pacing constraints”

3. Use “medium value” wholesale market price as “value of savings” to establish limit

Problems• Assumes perfect foresight• Wholesale market prices do not reflect full cost of new generation

5

Option 1 – Use 110% of Cost of New Generation – But Which One?

Assumptions :

Efficiency Cost = Average Cost of All Conservation in Draft 6th Power Plan Under $100 MWh

Transmission cost & losses to point of LSE wholesale delivery

2020 service - no federal investment or production tax credits

Baseload operation (CC - 85%CF, Nuclear 87.5% CF, SCPC 85%)

Medium NG and coal price forecast (6th Plan draft)

6th Plan draft mean value CO2 cost (escalating, $8 in 2012 to $47 in 2029).

$0

$50

$100

$150

$200

$250

$300

Energ

y Effic

iency

Geothe

rmal

Combin

ed C

ycle

Col. B

asin

Win

d

AB Wind

Advan

ced N

uclea

r

Super

critic

al C

oal (

No CSS)

IGCC (N

o CSS)

Recip

roca

ting

Engine

Woo

d Res

idue

(No C

HP)

MT

Win

d

WY W

ind

CSP Par

aboli

c Tro

ugh

Utility

Photo

volta

ic

Leve

lized

Life

cycl

e C

ost

(200

6$/M

Wh)

Emission (CO2) costTransmission & Losses

System IntegrationPlant costs

Even If We Pick A Resource, What is 110% of its Cost?

$0

$50

$100

$150

$200

$250

$300

$350

$400

0 2000 4000 6000 8000 10000 12000 14000 16000

MWa

20

06

$/M

Wh

Coal ConservationGasRenewablesNuclear

Generic coal, gas and nuclear units are shown at typical project sizes - more units could be built at comparable cost.

7

Option 2 - Match Median Conservation RPM “Build Out” – Lost Opportunity Resources

0

500

1000

1500

2000

2500

3000

3500

4000

2010

2012

2014

2016

2018

2020

2022

2024

2026

2028

Cu

mu

lati

ve R

eso

urc

eD

evel

op

men

t (M

Wa)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

3130 MWa by 2030

Lost-Opportunity Supply Technically Achievable Supply Curve

0

500

1000

1500

2000

2500

3000

3500

4000

<0 $20 $40 $60 $80 $100 $120 $140 $160 $180 $200TRC Levelized Cost (2006$/MWH)

Res

ourc

e P

oten

tial (

MW

a)

RPM Acquires 3130 MWa by 2030. This amount is technically available between $100 and $110/MWH

9

Option 2 - Match Median Conservation RPM “Build Out” – Non-Lost Opportunity Resources–

But Which One?

0

500

1000

1500

2000

2500

3000

3500

2010

2012

2014

2016

2018

2020

2022

2024

2026

2028

Cu

mu

lativ

e R

eso

urc

eD

eve

lop

me

nt

(MW

a)

0%10%20%30%40%50%60%70%80%90%100%

2830 MWa by 2030

2640 MWa by 2026

Non-Lost Opportunity Supply Technically Achievable Supply Curve

0

500

1000

1500

2000

2500

3000

3500

4000

<0 $20 $40 $60 $80 $100 $120 $140 $160 $180 $200TRC Levelized Cost (2006$/MWH)

Res

ourc

e P

oten

tial (

MW

a)

RPM Acquires 2640 MWa by 2026. This amount is technically available @ $90/MWH

11

Unfortunately, A Single Levelized Cost Does Not Reflect On and Off Peak Value

$0

$10

$20

$30

$40

$50

$60

$70

$80

January w/oCO2

August w/oCO2

January w/CO2 August w/CO2

For

ecas

t P

erio

d R

eal L

evel

ized

Cos

t (2

006$

/MW

H)

High Load HoursLow Load Hours

12

Option 3 – Use Wholesale Prices To Determine Cost-Effectiveness

$0

$10

$20

$30

$40

$50

$60

$70

$80

$90

$100

Ma

y-9

6

Ma

y-9

7

Ma

y-9

8

Ma

y-9

9

Ma

y-0

0

Ma

y-0

1

Ma

y-0

2

Ma

y-0

3

Ma

y-0

4

Ma

y-0

5

Ma

y-0

6

Ma

y-0

7

Ma

y-0

8

Ma

y-0

9

Mid

-C M

on

thly

Wh

ole

sale

Pri

ce

(20

06

$/M

WH

)

Which Wholesale Market Price Should Be Used?

$0

$20

$40

$60

$80

$100

$120

2005 2010 2015 2020 2025 2030

Who

lesa

le M

arke

t Pric

e (2

006$

/MW

H) Historical

6th Plan - Low w/Carbon6th - Medium w/Carbon6th Plan - High w/Carbon

Method for Selecting the “Appropriate” Market Price Forecast

• The “Right” Market Price forecast is one that results in the pace and amount of conservation identified as cost-effective by the Resource Portfolio Model (RPM) over 20-yrs

• Use either the Aurora Mid-C medium market price forecast with or without carbon control cost (or any other price forecast)

• Adjust Aurora price with “adders” to reflect carbon control price included in forecast until “mini-RPM” builds same level of conservation as full RPM

15

Proposed Calibration Goal – Match Average Conservation “Build Out” From RPM

0

1000

2000

3000

4000

5000

6000

7000

Sep-09 Sep-13 Sep-17 Sep-21 Sep-25Hydro-Year

Cu

mu

lati

ve

Sa

vin

gs

(M

Wa

) Lost OpportunityNon-Lost Opportunity

2640 MWa by 2026

3130 MWa by 2030

16

What’s the “mini-RPM”

• Identical to full Resource Portfolio Model (RPM) except all inputs are “deterministic”– Average price forecast for gas and electricity– Average load growth forecast– Expected value cost for resources, forced

outage rates, dispatch order– Average carbon control cost and timing

• Can be run with or without median carbon cost in market price

17

What’s The “Market Price Adder”

• Wholesale market prices no longer represent the full cost of new generation– State mandated RPS increase WECC system

wide energy surplus, which reduces value of wholesale market

• RPM uses 750 different wholesale market prices to establish the value and risk of acquiring new resources, including conservation

• Procost cost-effectiveness uses a single market price forecast which does not reflect uncertainty

18

Expected Value Market Price Forecast without Carbon Control Cost

$0

$10

$20

$30

$40

$50

$60

$70

$80

Jan-08 Jan-12 Jan-16 Jan-20 Jan-24 Jan-28

Who

lesa

le M

arke

t Pric

e(2

006$

/MW

H)

Annual Average

Monthly Average High Load HoursMonthly Average Low Load Hours

19

Expected Value of CO2 Cost Across Futures Modeled in RPM

0

10

20

30

40

50

60

Sep

-09

Sep

-11

Sep

-13

Sep

-15

Sep

-17

Sep

-19

Sep

-21

Sep

-23

Sep

-25

Sep

-27

Car

bon

Cos

t (2

006$

/ton

)

20

Expected Value Market Price Forecast with Expected Value CO2 Control Cost

$0

$10

$20

$30

$40

$50

$60

$70

$80

$90

$100

Jan-08 Jan-12 Jan-16 Jan-20 Jan-24 Jan-28

Who

lesa

le M

arke

t Pric

e(2

006$

/MW

H)

Annual Average

Monthly Average High Load HoursMonthly Average Low Load Hours

21

Input Options

• Use Aurora market price forecast, including median carbon cost.– Primary Advantage: Reflects impact of carbon cost on the resource

dispatch– Primary Disadvantage: Embeds specific carbon cost assumptions

in market price forecast (limits flexibility and reduces transparency)

• Use Aurora market price forecast, input carbon cost separately– Primary Advantage: Allow for explicit input of carbon cost

assumptions (increases flexibility and transparency)– Primary Disadvantage: Does not reflects impact of carbon cost on

the resource dispatch

22

What’s the “Calibration Adder”

• RPM uses 750 futures– Procost models only a single future

• RPM’s market prices and carbon cost are “volatile”– Procost’s prices are not “volatile”

• RPM futures are not “normally distributed,” hence the median values are not the average values

• RPM uses a single load shape for conservation– Procost models each measure’s load shape

23

Market Price Adders for Conservation Cost-Effectiveness for Average Conservation Shape

$0

$10

$20

$30

$40

$50

$60

$70

LOWithoutCarbon

LO WithCarbon

NLOWithoutCarbon

NLO WithCarbon

RPM AdderCalibration AdderTotal LO Adder

Proposed Final Input Assumption Choices

Resource Type

Carbon Cost Modeled

Pounds CO2 per kWh

RPM Adder ($/MWH)

Calibration Adder ($/MWH)

Total LO Adder ($MWH)

Lost Opportunity

Procost adds carbon

Fixed @ .99 lbs/kWh

$50 $18 $68

Non-Lost Opportunity

Procost adds carbon

Fixed @ .99 lbs/kWh

$35 $18 $53

Lost Opportunity

Aurora price w/carbon

Per Aurora Dispatch

$50 $8 $58

Non-Lost Opportunity

Aurora price w/carbon

Per Aurora Dispatch

$35 $8 $43

All of the above input produce identical TRC B/C Ratio’s for the weighted average conservation load shape and 13 year measure life.