CPUC Avoided Cost Workshop

1

CPUC Avoided Cost Workshop

T&D Avoided Cost

2

T&D Comment Areas

“Adopted T&D Costs, not a uniform methodology”

Are time and location differences worth the effort?

Need for $/kW-yr capacity cost Avoided cost only if DSM/CEE reductions are

reliable Generic T&D costs should not be used

3

Goals for the T&D Methodology

Disaggregate information by area and time to facilitate detailed analyses where appropriate

Use publicly available data, or information that can be easily provided by utilities

Transparent method Easily updated (not necessarily on an annual basis)

Working group deliberated on several alternative approaches for methodology

Conclusion was that results were similar across several approaches (Present Worth, DTIM, TIM)

4

PW Method is Based on Deferral of Investments

Year

Loa

d

Build Year

Capability

Area Loads

Loa

d

Change Year

Capability

InvestmentPV = $7.35M

Base Plan

Shifted Plan

Load Change

InvestmentPV = $7.93M

Load decrease delays investment need

PV value of deferral is PV(base plan) - PV(change plan)

5

PG&E Total T&D Marginal Costs

$0

$10

$20

$30

$40

$50

$60

$70

CE

NT

RA

L C

OA

ST

DE

AN

ZA

DIA

BLO

EA

ST

BA

Y

FR

ES

NO

KE

RN

LOS

PA

DR

ES

MIS

SIO

N

NO

RT

H B

AY

NO

RT

H C

OA

ST

NO

RT

H V

ALL

EY

PE

NIN

SU

LA

SA

CR

AM

EN

TO

SA

N F

RA

NC

ISC

O

SA

N J

OS

E

SIE

RR

A

ST

OC

KT

ON

YO

SE

MIT

E

Sys

tem

Ave

rage

Present Worth

TIM

DTIM

PG&E’s Electricity T&D: Comparison of Methods

Annualized MCs fell within a tight band using the DTIM, TIM or PW methods

MCs varied significantly by

planning division. Transmission MCs on

a system average basis only

1999$/ kW-yr

NOTE: Results exclude new business primary distribution marginal costs, which are borne by the customer and therefore not avoidable by the utility.

6

SCE Electric Distribution: Comparison of Methods

Southern California Total Distribution Marginal Costs

$0

$20

$40

$60

$80

$100

$120

Do

min

gu

ez

Hill

s

Fo

oth

ill

Ve

ntu

ra

Sa

nta

An

a

Ru

ral

Present Worth

DTIM

TIM

RCN*

2004$/ kW-yr

Under RCN, rural distribution has a high value because $14.5MM of investment for 20MW of new capacity

Transmission avoided costs exclude economic projects

SCE Transmission Marginal Costs

18.81

15.80 16.21

10.24

-

5

10

15

20

Present Worth

DTIM TIM RCN

2004

$/kW

-yr

7

SDG&E Electric T&D: Comparison of Methods

When E3 calculated distribution marginal costs using 2002-2007 data, the 3 methods yielded similar results

Transmission: Although SDG&E uses embedded costs, it has clarified which investments are demand-related

SDG&E Electric T&D Marginal Costs

0

10

20

30

40

50

60

70

80

Distribution Transmission

20

04

$/k

W-y

ea

r

PW

DTIM

TIM

8

SCE's T&D Costs by Area, Climate Zone: Blended

0

20

40

60

80

100

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

$/kW

-yr

Dominguez Hills (CZ 6,8)Foothills (CZ 10,16)Ventura (CZ 5,6,9)Santa Ana (CZ 6,8)2001 EE ValuesRural (CZ 14,15,16)System Average

Extrapolation of T&D Avoided Cost Estimates

Long RunTransition

PG&E: Retain cost differentials SCE: Blended Long Term Forecast Approach (shown below)

2001 EE Values

Rural

9

T&D Avoided Costs by Planning Division

SDG&E

$77.76

SCE

$36.00

$21.00

$5.00

PG&E

$70.00

$38.00

$5.00

10

Recommendation: Methods

“Use adopted T&D marginal costs, not a uniform methodology” PW, DTIM, and TIM all produce comparable results, so

using adopted values based on these methods would be fine --- provided that the adopted values retain the area differentiation

RCN can produce very different results, and should not be used.

“Are time and location differences worth the effort?” Location does reveal significant cost differences for

PG&E as well as SCE’s Rural area. Other SCE areas are less clear.

11

Allocation of T&D Based on Temperature by Climate Zone

Temperature Loads T&D Capacity Cost

Drives Drives

Load Information Missing or Difficult to Obtain in Many Areas

Temperature

Use temperature as a proxy for load, and as the basis for allocating costs to

hours of the year.

T&D Capacity Cost

12

Summer PeakLoad vs. Temperature

Fresno

Yellow8am to 10pm

Similar analysis done on 33 PG&E areas as part of CEC Title 24 development

13

Example Results for PG&E Stockton

15

913

1721

14

710

$0

$50

$100

$150

$200

$250

Hour

Month

Levelized Avoided Cost by Month and Hour ($/MWh)

$200.00 - $250.00

$150.00 - $200.00 $100.00 - $150.00

$50.00 - $100.00

$- - $50.00

14

PG&E: Allocation by TOU Period

Low avoided costs in Oakland climate zone allocated much more evenly across TOU periods due to temperate weather

PG&E Oakland Climate Zone

2004$/kW-yr TOU Period

Total Costs($000)

Total Load by TOU Period

(MWh)

T&D Marginal Costs (by TOU, PCAF)

($/kWh)Summer 24.50 1 On Peak 409,732 24,852,856 0.016

2 Shoulder 127,102 19,887,184 0.0063 Off Peak 130,397 58,181,254 0.002

Winter 24.50 4 On Peak - - 5 Shoulder 232,677 36,539,717 0.0066 Off Peak 361,739 53,490,832 0.007

PG&E Fresno Climate Zone

2004$/kW-yr TOU Period

Total Costs($000)

Total Load by TOU Period

(MWh)

T&D Marginal Costs (by TOU, PCAF)

($/kWh)Summer 34.31 1 On Peak 770,741 24,852,856 0.03101

2 Shoulder 9,001 19,887,184 0.000453 Off Peak 145,719 58,181,254 0.00250

Winter 34.31 4 On Peak - - 5 Shoulder 432,758 36,539,717 0.011846 Off Peak 390,232 53,490,832 0.00730

(From working group presentations)

Oakland on-peak is about half of Fresno.

15

Recommendation: Hourly Costs

Need for $/kW-yr capacity cost Not needed if hourly costs are used, but could

be more accurate when costs are averaged into TOU periods.

Caution must be exercised, however, to avoid overestimating capacity reductions.

Are time and location differences worth the effort? Time difference can be dramatic in comparing

coastal and inland areas.

16

Recommendation: Other T&D Issues

Avoided cost only if DSM/CEE reductions are reliable Two aspects to this issue:

“Do you expect the reduction when you need it?” This is addressed by the hourly costs (or hourly peak allocation factors)

“What happens if the reduction is not there when you need it?” This is probably not an issue for DSM because of the high diversity of DSM reductions. This can be an issue for other proceedings, in which case the avoided cost should be de-rated.

Generic T&D costs should not be used Do not see a need for individual studies for DSM

program evaluation.

17

Gas T&D Findings

18

Gas Results are Simpler Than Electric Results are by IOU, not sub-area Hourly allocations are not needed because of the

natural storage capability of gas pipe. Allocation of costs to winter peak months (November

through March) when peak loads occur Results are shown by customer class to reflect the

usage pattern differences of the classes. Results can also be provided at a system average level

19

Gas T&D Avoided Costs

Gas T&D annual average avoided costs can vary widely by utility, with PG&E having the highest costs

Results significantly above California 2001 energy efficiency values (about $0.03/therm)

20

No Gas Storage Costs in these Avoided Cost Estimates Need for inventory expansions to support future growth driven

by: Growth in core demand for firm withdrawal as core peak day

demand rises Varying attractiveness of storage depending on

summer/winter price differentials and the national storage market

To a considerable extent, storage and backbone transmission capacity are substitutes

The appropriateness of including storage will depend upon the extent to which peak capacity requirements are driving the need for the project.

21

Monthly Average Residential Throughput (%)

3%

5%

7%

9%

11%

13%

15%

17%

Jan

Fe

b

Ma

r

Ap

r

Ma

y

Jun

Jul

Au

g

Se

p

Oct

No

v

De

c

PG&E SoCal SDG&E

Gas T&D Costs Allocated to Winter Season

Aggregate throughput by utility by summer and winter seasons. Allocate marginal costs to winter months when peak loads occur

SummerWinter Winter

22

Winter Allocations by Utility