Bill Bradbury Chair

Oregon

Jennifer Anders Vice Chair Montana

Henry Lorenzen

Oregon

W. Bill Booth Idaho

James A. Yost

Idaho

Pat Smith Montana

Tom Karier Washington

Phil Rockefeller

Washington

October 28, 2014

MEMORANDUM TO: Council members FROM: Tom Eckman, Power Division Director SUBJECT: Briefing on Power Planning Under Uncertainty BACKGROUND: Presenter: Tom Eckman Summary: Staff will present an overview of Council’s approach to resource planning,

focusing on how its analytical processes address the inherent uncertainty in major drivers of electricity demand, resource costs and risks. This is the second in a two-part series designed to provide Council members and interested stakeholders with background on the analytical methods used in plan development.

Relevance: The Council is engaged in the development of is 7th Power Plan. The

product of that development process is a plan that represents the Council’s collective assessment of resource development and other actions that will “assure the Pacific Northwest of an adequate, efficient, economical and reliable power supply.”

Workplan: 1D - Prepare for Seventh Power Plan and maintain analytical capability Background: In 1982, shortly after the Council was formed, Dr. Kai Lee, then a

professor at the University of Washington and who later served as Washington Council member, authored a paper entitled, The Path Along

the Ridge: Regional Planning in the Face of Uncertainty.1 In his paper Dr. Lee described how the Council’s planning process departed from traditional utility power planning.

In the Council’s power plans there is an explicit recognition that the future is uncertain and that risk management strategies to deal with that uncertainty are needed. For example, until the Council’s first plan, utility resource planning was based on a single forecast of the region’s most likely energy demand. Resources that took ten or 15 years to build were planned and constructed to that best guess; if the future turned out differently, the region faced the problem of either having under built or overbuilt resources. The cost of error on either side was enormous.

The Council explicitly recognizes that the future cannot be predicted accurately and that uncertainty is a fact of life in power planning. To accommodate this problem, the Council has developed plans to meet a broad range of potential growth in energy demand, setting a boundary of high and low load growth forecasts over the next 20 years. The Council’s plans have also identified flexible resources such as conservation and options that shorten the lead time of generating resources. This presentation will expand on the types of uncertainty the Council’s planning process must address as well as describe the analytical methods used by the Council to evaluate and identify resource strategies that can be used to mitigate risk at an acceptable cost.

More Info: See linked paper “An Overview of the Council’s Planning Methods.” From March 2011.

1 Lee, Kai N. The Path Along the Ridge: Regional Planning in the Face of Uncertainty. University of Washington Law Review, 1982-1983, Volume 58, pp 317 – 342. ( https://digital.law.washington.edu/dspace-law/bitstream/handle/1773.1/104/volume%2058.pdf?sequence=1)

Planning for Uncertainty An Introduction to the Council’s

Power Planning Process

Tom Eckman Director, Power Division

Northwest Power and Conservation Council November 5, 2014

The Resource Planner’s Problem

Don’t have too many resources Don’t have too

few resources Have “just the

right amount” of resources*

*Resources include energy, capacity, flexibility and other ancillary services needed for system reliability.

As A Utility’s Resource Mix Changes So Does Its Cost and Risk

0

2000

4000

6000

8000

10000

12000

Resources Loads

GWH/

yr

Firm Contracts/Resources

Market Purchases

Incr

easi

ng R

isk

Increasing Reserve Margin

Exposure to Market Volatility

Exposure to Load Volatility

Incr

easi

ng C

ost

The “Just Right” Resource Portfolio

3

Increasing Firm Contracts/Resources Increases Load Volatility Risk

0

2000

4000

6000

8000

10000

12000

Resources Loads

GWH/

yr

Firm Contracts/Resources/Loads

Market Purchases

Incr

easi

ng R

isk

Increasing Reserve Margin

Exposure to Market Volatility

Exposure to Load Volatility

Incr

easi

ng C

ost Over

Supply

4

The Region Has Experienced Overbuilding

3,000

8,000

13,000

18,000

23,000

28,000

1965 1970 1975 1980 1985

Tota

l PN

W E

lect

ric L

oads

(MW

a)

1960 Forecast 1965 Forecast 1970 Forecast 1975 Forecast 1980 Forecast Actual Loads

3,700 MWa Forecasting “Error” Only Five Years in the Future

Real World Example of the Cost of “Too Many Resources”

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

1938 1942 1946 1950 1954 1958 1962 1966 1970 1974 1978 1982 1986

BPA

Who

lesa

le P

ower

Rat

e (C

ents

/kW

h)

Year

Nominal$ Cents/kWh 2006$ Cents/kWh

Wholesale Prices (BPA Rates) REALLY Increase (418% in real dollars over 5 years)

6

PNW Retail Electric Rates 1938 - 1985

0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0

1938

19

40

1942

19

44

1946

19

48

1950

19

52

1954

19

56

1958

19

60

1962

19

64

1966

19

68

1970

19

72

1974

19

76

1978

19

80

1982

19

84

Reta

il Ra

te (C

ents

/kW

h)

Nominal$ Cents/kWh 2006$ Cents/kWh

Retail Electric Rates REALLY Increase in Response to Thermal Plant Costs

Decreasing Firm Contracts/Resources Increases Market Risk…

0

2000

4000

6000

8000

10000

12000

Resources Loads

GWH/

yr

Firm Contracts/Resources/Loads

Market Purchases

Incr

easi

ng R

isk

Increasing Reserve Margin

Exposure to Market Volatility

Exposure to Load Volatility

Incr

easi

ng C

ost Market

Purchases

8

The Region Has Also Experienced Underbuilding

$0

$100

$200

$300

$400

$500

$600

$700

$800

Mid

- C W

hole

sale

Ele

ctrici

ty P

rice

(2

006$

/MW

H)

During the mid-1990’s, Low Wholesale Market Prices Coupled With A Series of Above Average Water Years Led To An “Overexposure” to Market Prices

Real World Example of the Cost of “Too Few Resources”

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

1985 1990 1995 2000 2005 2010

NW

Ave

rage

Rev

enue

/kW

h (c

ents

)

PNW Retail Electric Rates 1985 - 2010

Nominal$ Cents/kWh 2006$ Cents/kWh

Retail Electric Rates Increase in Response to Over-Exposure to Short-Term Market

10

How The Council Addresses The

“Goldilocks” Problem?

First, We’ve Broken the Problem Into Six Simple Questions

1. When Will We Need Resources? 2. How Much Will We Need? 3. What Should We Build/Buy? 4. How Much Will It Cost? 5. What’s the Risk? 6. Who Can We Blame If We Get It Wrong?

(Answer – the Staff)

All Plans Require Assumptions About the Future

However, it is an occupational hazard of planners!

Perfect Foresight (i.e., prescience) is not possible.

13

Plan Must Address Three Major Sources of Uncertainty

1. Load Uncertainty 2. Resource Uncertainty

Output Cost Construction Lead Times

3. Wholesale Electricity Market Price Uncertainty

14

Load Uncertainty Is Particularly A Problem When Generating Resources Have Long Lead

Times and Large Sizes

0

5

10

15

20

0 200 400 600 800 1000

Reso

urce

Lead

Tim

e (y

ears

)

Resource Size (Megawatts)

Nuclear

Coal Gas-Fired CCCTs*

*Power Plant and Industrial Fuel Use Act of 1978 limited natural gas fired turbines run times to 1500 hours per year. (Repealed in 1987)

15

Generating Resource Long Lead Times Combined with Significant Load Uncertainty Created the Risk of Under and Over Building

3000

8000

13000

18000

23000

28000

1950 1955 1960 1965 1970 1975 1980 1985

Tota

l PN

W E

lect

rici

ty U

se (

aMW

)

1960 Forecast 1965 Forecast 1970 Forecast

1975 Forecast 1980 Forecast Actual Loads

~ 3700 MWa difference between NRF forecast and actual loads, only 5 yrs after forecast was done

Council Approach to Load Forecasting Assumes Lack of Perfect Foresight

(We Admit We Aren’t Prescient)

14,000

16,000

18,000

20,000

22,000

24,000

26,000

1980 1985 1990 1995 2000

Regi

onal

Loa

d (M

Wa)

Plan 1 – Load Forecast Range

1983 - Low 1983 Med-Low 1983 Med-High 1983 High

First Plan’s Response to Load Uncertainty

Rely on efficiency due to lower cost, short lead times and ability to match development to scale of load growth Develop “options” on thermal projects with

long lead times Get the siting and licensing out of the way so

that construction can commence when actual load growth requires development

18

Options Concept Was Designed to Address Load Uncertainty and Long Resource

Development Cycles

Source: 1983 Northwest Power and Conservation Plan

19

Historical Levels of Load Uncertainty Were Often Driven by Large Industrial Load

0

500

1000

1500

2000

2500

3000

3500

1960

19

62

1964

19

66

1968

19

70

1972

19

74

1976

19

78

1980

19

82

1984

19

86

1988

19

90

1992

19

94

1996

19

98

2000

20

02

2004

20

06

2008

20

10

Annu

al Lo

ad (M

Wa)

PNW Direct Service Industry Load

Now, Not So Much

20

As A Result, Load Uncertainty Still Exists, But Near Term Volatility Is Lower

-

5,000

10,000

15,000

20,000

25,000

1970 1975 1980 1985 1990 1995 2000 2005 2010

Regi

onal

Loa

d (M

Wa)

Increased Electricity Cost from West Coast “Energy Crisis” Resulting in Closure of PNW Aluminum Smelters and Other Industrial Facilities. Most of these smelters remain closed.

In Addition, Conservation and Shortened Lead Times and Smaller Sizes For Some Generating Resources Have Made

Them More Flexible

0

2

4

6

8

10

12

14

16

18

0 200 400 600 800 1000

Reso

urce

Lead

Tim

e (y

ears

)

Resource Size (Megawatts)

Nuclear

Coal

Conservation

Gas-Fired SCCT & CCCTs Wind

22

Now, Short Lead Time, Smaller Resources Are Available, Reducing the Risk and Cost

Mitigation Value of “Resource Optioning”

Source: 1983 Northwest Power and Conservation Plan

23

Plan Must Address Three Major Sources of Uncertainty

1. Load Uncertainty 2. Resource Uncertainty

Output Cost Construction Lead Times

3. Wholesale Electricity Market Price Uncertainty

24

Generating Resources Are Subject to Unanticipated Outages (i.e., “Forced

Outages”) Which Reduces Their Availability

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

8.0%

CCCT Coal Nuclear

Una

ntic

ipat

ed F

orce

d O

utag

e Ra

te

(%)

Generating Resource Forced Outage Rates by Resource Type

2007 2008 2009 2010 2011

25

Resource Variability Is Different Than Resource Uncertainty

But Both Are Important

26

Energy Efficiency Resource Uncertainty Stems from Delays in Deployment (i.e. Construction) Schedule

-

200

400

600

800

1,000

1,200

1,400

1,600

1984 1989 1994 1999

Cum

ulat

ive

Savi

ngs (

MW

a) Plan Cumulative Goals

Cumulative Achievements*

*Achievements reflect utility and NEEA savings only. Savings from codes and standards are included as baseline adjustments in each plan’s baseline load forecast

27

Although This Source of Uncertainty Appears to be Diminishing

Since the West Coast Energy Crisis Actual Program Achievements Have Exceed Council Plan Goals

0

50

100

150

200

250

300

1984 1988 1992 1996 2000 2004 2008 2012

Cum

ulat

ive

Savi

ngs (

MW

a) Plan Annual Goal

Actual Annual Achievement

*Achievements reflect utility and NEEA savings only. Savings from codes and standards are included as baseline adjustments in each plan’s baseline load forecast

28

$0

$50

$100

$150

$200

$250

$300

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

Real

Lev

elize

d Co

st (2

006$

/MW

h)

Lifetime Capacity Factor

Lifecycle Cost of Combined Cycle Gas Fired Combustion Turbine at Varying Gas Prices and Capacity Factors

$2.00/MMBtu $4.00/MMBtu $6.00/MMBtu $8.00/MMBtu

Resource Cost Uncertainty Is Driven by Input Fuel Prices and Utilization (i.e., “Capacity Factors”)

Minimum Capacity

Factor

Average Capacity

Factor Maximum Capacity

Factor

29

-

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0 Ja

n-91

Jan-

92

Jan-

93

Jan-

94

Jan-

95

Jan-

96

Jan-

97

Jan-

98

Jan-

99

Jan-

00

Jan-

01

Jan-

02

Jan-

03

Jan-

04

Jan-

05

Jan-

06

Jan-

07

Jan-

08

Jan-

09

Jan-

10

Jan-

11

Jan-

12

Jan-

13

Pric

e In

dex

Com

pare

d to

Ja

n-19

91 P

rice

Iron Ore Wheat Natural Gas (Henry HUB) Rubber Crude Oil - West Texas Uranium

Forecasting Natural Gas Prices Is Equivalent to Engaging in Commodity Trading

Which of these commodity price trends is natural gas?

30

Combined Cycle Generation Resource Capacity Factors Vary Significantly From Year-to-Year

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

100% 0 1 2 3 4 5 6 7 8 9 10

11

12

13

14

15

16

17

18

19

20

Annu

al C

apac

ity F

acto

r

Combined Cycle Resource Number

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

31

These Uncertainties Mean There’s No Single “Avoided Cost” for New Resources

$-

$50

$100

$150

$200

$250

$300

2020 2021 2022 2023 2024 2025 2026 2027 2028 2029

Leve

lized

Cos

t (2

012$

/MW

h)

Plant In-Service Year

Levelized Cost and In-Service Date of Combined Cycles Combustion Turbine Across A Sample of 6th Plan Futures

32

Plan Must Address Three Major Sources of Uncertainty

1. Load Uncertainty 2. Resource Uncertainty

Output Cost Construction Lead Times

3. Wholesale Electricity Market Price Uncertainty

33

Market Price Surprises – Electricity

$0

$100

$200

$300

$400

$500

$600

$700

$800

Mid

- C W

hole

sale

Ele

ctrici

ty P

rice

(2

006$

/MW

H)

$0 $10 $20 $30 $40 $50 $60 $70 $80 $90

$100

$0.00 $1.00 $2.00 $3.00 $4.00 $5.00 $6.00 $7.00 $8.00 $9.00 $10.00

Mid

C W

hole

sale

Ele

ctric

ity P

rice

($/

MW

h 20

12$)

Natural Gas Price ($/MMBtu 2012$)

Historic Wgt Ave Mid C Price Forecast Mid Case Lin. Fit to Forecast data

Wholesale Electricity Market Prices Are Strongly Correlated to Natural Gas Prices

35

So Natural Gas Market Price Surprises Also Affect Wholesale Electricity Prices

$0

$2

$4

$6

$8

$10

$12

$14 19

83

1984

19

85

1986

19

87

1988

19

89

1990

19

91

1992

19

93

1994

19

95

1996

19

97

1998

19

99

2000

20

01

2002

20

03

2004

20

05

2006

20

07

2008

20

09

2010

20

11

2012

$/M

MBT

U

36

With All These Uncertainties, How Does the Council Answer Those Simple Questions?

1. When Will We Need Resources? 2. How Much Will We Need? 3. What Should We Build/Buy? 4. How Much Will It Cost? 5. What’s the Risk?

The lowest cost, lowest risks resources first.

Resource Portfolio Analysis on One Slide

$0

$50

$100

$150

$200

$250

$300

$350

$400

0 2000 4000 6000 8000 10000 12000 14000 16000

2006

$/M

Wh

MWa

Coal Conservation Gas Renewable Nuclear

Generic coal, gas and nuclear units are shown at typical project sizes - more units

While the “All Resource Supply Curve” tells use what to acquire, it doesn’t tell us how much, when, or the costs and risks of acquisition!

^ Almost

38

The Answers to Those Questions Requires Planning for Uncertainty

Resource Strategies – actions and policies over which the decision maker has control that will affect the outcome of decisions

Futures – circumstances over which the decision maker has no control that will affect the outcome of decisions

Scenarios – Combinations of Resource Strategies and Futures used to “stress test” how well what we control performs in a world we don’t control

39

This Is What the Regional Portfolio Model Does

Electricity Demand Forecast

Regional Portfolio Model

Generating Resource Potential Assessment

Energy Efficiency “Supply Curves”

Load Forecast Range

(without efficiency)

Generating Resource

Cost & Availability

Energy Efficiency Resource Potential Assessment

Units & Baseline Unit Use

Council Reviews Cost and Risk of Alternative Resource Portfolios

Distributions of Key Drivers (e.g., Fuel prices, wholesale market prices)

Data to Create

Futures Council Adopts Plan’s Resource Portfolio Management Strategy and Action Plan

40

Council Follows the “Gump” Resource Strategy Testing Model

The Future’s Like A Box of Chocolates. You Never Know What You’re Gonna Get.

41

0% 2% 4% 6% 8%

10% 12% 14% 16%

0.8%

1.0%

1.3%

1.5%

1.8%

2.0%

2.3%

2.5%

2.8%

3.0%

3.3%

3.5%

3.8%

4.0%

Prob

abili

ty (

%)

Annual Load Growth

$0

$5

$10

$15

$20

$25

2010 2015 2020 2025

2006

$/M

MBtu

Natural Gas Prices

$0

$50

$100

$150

$200

$250

$300

2010 2015 2020 2025

2006

$/M

WH

Wholesale Market Electricity Price

$0

$20

$40

$60

$80

$100

$120

2010 2015 2020 2025

2006

$/To

n

Carbon Price

0

2

4

6

8

10

12

245 514 1598 2202 2560 3444 4934 6735 8945

Rea

l Lev

eliz

ed C

ost

(Cen

ts/k

Wh

- 20

00$)

Cumulative Supply (MW)

Resource Supply Curve

Regional Portfolio Model

0

10

20

30

40

50

60

70

$50 $75 $100 $125 $150 $175 $200 $225

Freq

uenc

y

NPV System Cost (billion2006$)

Portfolio ABCD

$155

$156

$156

$157

$157

$158

$158

$104 $104 $105 $105 $105 $105 $105 $106

NPV

Sys

tem

Ris

k (2

006$

billi

ons)

NPV System Cost (2006$billions)

Efficient Frontier

Council Portfolio Analysis Process “Test A Lot of Chocolates”

0

5000

10000

15000

20000

25000

Capa

city

(MW

)

Hydro System Output

42

The RPM Finds the Lowest Cost “Insurance” for the Same Risk Coverage

43

What We Learn From “Stress Testing” Alternative Resource Strategies Forms the Basis

of “The Plan”

Does the amount and pace of energy efficiency development change across “low cost” and “low risk” futures? How sensitive are resource strategies to

assumptions regarding future carbon risk/ prices? What resource strategies provide the

greatest “hedge” against electricity and gas price uncertainty?

44

Insights From Prior Plans Preferred Resource Characteristics

Resource Type Low Cost

Short Lead Time

Small Increment

No or Low Fuel Price Risk

Low Carbon Policy Risk

Energy Efficiency Wind Solar PV

Gas SCCT/CCCT Coal Nuclear

= Resource exhibits desired characteristic = Resource partially exhibits desired characteristics

45

Your Task Is to Ensure that the 7th Plan’s Resource Strategy’s Benefits Outweigh Its Risks

Benefits

Risks

46

Any Questions?