Vermont Long-Term energy and demand foreast

VERMONT LONG-TERM ENERGY AND DEMAND FOREASTOctober 8, 2013Eric Fox and Mike Russo

AGENDA» Recent Sales and Customer Trends» Preliminary State Sales and Demand Forecast» Building a No DSM Forecast » End-Use Saturation Projections Integrating the New Appliance

Saturation Survey» Estimating Solar Load Impacts

VELCO SYSTEM ENERGY 2003 TO 2013

While sales have been flat over the last ten yearsThere have been two distinct periods – before and after the recession

* 2013 – 7 mo actual /5 mo forecast

Year Energy (GWh) chg Wthr Nrm Energy chg2003 6,285 6,2092004 6,390 1.7% 6,395 3.0%2005 6,523 2.1% 6,426 0.5%2006 6,473 -0.8% 6,544 1.8%2007 6,536 1.0% 6,489 -0.8%2008 6,419 -1.8% 6,424 -1.0%2009 6,143 -4.3% 6,195 -3.6%2010 6,209 1.1% 6,207 0.2%2011 6,204 -0.1% 6,204 0.0%2012 6,175 -0.5% 6,192 -0.2%2013 6,252 1.3% 6,226 0.5%chg2003 - 2007 1.0% 1.1%2007 - 2013 -0.7% -0.7%2003 - 2013 0.0% 0.0%

VELCO SYSTEM PEAK DEMAND

*WNLF_Peak = WN Energy / WN Peak Load Factor

System peak shows a similar patternPositive growth in 2013

Year System Peak chg WN Peak chg WNLF_Peak chg2003 1,000.7 1,034.1 1,033.6 2004 984.9 -1.6% 1,070.8 3.5% 1,064.7 3.0%2005 1,073.6 9.0% 1,063.4 -0.7% 1,069.8 0.5%2006 1,126.5 4.9% 1,108.0 4.2% 1,089.5 1.8%2007 1,072.6 -4.8% 1,077.4 -2.8% 1,080.3 -0.8%2008 1,048.4 -2.3% 1,068.2 -0.9% 1,069.4 -1.0%2009 1,016.5 -3.0% 1,020.2 -4.5% 1,031.4 -3.6%2010 1,068.7 5.1% 1,040.5 2.0% 1,033.4 0.2%2011 1,062.2 -0.6% 1,057.2 1.6% 1,032.9 0.0%2012 1,007.0 -5.2% 999.2 -5.5% 1,030.9 -0.2%2013 1,060.7 5.3% 1,036.5 3.7% 1,036.6 0.5%chg2003 - 2007 1.9% 1.1% 1.1%2007 - 2013 -0.1% -0.6% -0.7%2003 - 2013 0.7% 0.1% 0.0%

PEAK-DAY TEMPERATURES

2013 was the hottest peak day in 11 years. Second hottest day based on a 3-day weighted THI

GMP YEAR TO DATE SALESJanuary through August

YTD 13 YTD 12 DifferenceResidential 1,067,772 1,063,989 0.4%Commercial 1,038,187 1,031,016 0.7%Industrial 502,694 498,188 0.9%Total 2,608,653 2,593,193 0.6%

YTD 13 YTD 12 DifferenceResidential 219,978 218,458 0.7%Commercial 38,145 37,798 0.9%

YTD 13 YTD 12 DifferenceResidential 4,854 4,870 -0.3%Commercial 27,217 27,277 -0.2%

YTD 13 YTD 12 DifferenceHouseholds 259,891 258,731 0.4%Employment 306,140 302,568 1.2%

Weather Normalized GMP Sales

Customers

Avgerage Use

Vermont Economic Growth

Beginning to see some sales and customer growth with improving economy

Sales & Customers

Weather Conditions

Economic Drivers

Electric Prices

Structural Changes

Customer Class and End-Use Energy

Forecast

System Hourly Load

Peak-Day Weather System Energy and Peak Forecast

System Profile

System Hourly Load Forecast

Solar Load and EVs

Demand Response

DSM

SYSTEM FORECAST APPROACH

Residential and Commercial Sales Forecasted using a SAE Modeling

Framework

End-Use CP Factors

SAE MODELING FRAMEWORK

KEY FORECAST DRIVERS» Economic activity and population growth

• Moody Analytics Vermont economic forecast» Weather conditions

• Twenty-year normal HDD and CDD» Price projections » End-Use Intensity Projections

• Saturation forecast – Vermont appliance saturation surveys• Baseline efficiency projections – 2013 AEO

» Intensity adjustments for state efficiency programs» Residential and commercial solar net metering system market

penetration» Electric vehicles» ???

PRELIMINARY VERMONT ENERGY AND DEMAND FORECAST» Revenue class sales and customer data through March 2013» Moody Analytics June 2013 economic forecast for Vermont» 2012 end-use intensity projections

• Reflects the 2012 VEIC efficiency savings projections» No solar load adjustment» No price impacts

Used as a basis for developing a No DSM Forecast

BASELINE END-USE INTENSITY TRENDS» End-use intensity trends reflect both saturation and average

stock efficiency improvements» Current intensity trends reflect KEMA state saturation survey, EIA

New England forecasts, and past BED appliance saturation surveys

0

200

400

600

800

1,000

1,200

1,400

1,600

1,800

2,000

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

kWh

Heating

Cooling

EWHeat

ECook

Ref1

Ref2

Frz

Dish

CWash

EDry

TV

Light

Misc

DSM ADJUSTED END-USE INTENSITIES» Develop baseline end-use sales forecast from SAE models

» Calculate adjusted end-use sales forecasts for DSM savings projections• Residential: assume 75% is embedded in the baseline forecast• Commercial: assume 50% is embedded in the baseline forecast

» Adjust end-use intensity forecasts to reflect DSM savings adjustments

DSM embedded assumptions based on calibration to actual sales

DSM ADJUSTED WATER HEATING INTENSITY

500

550

600

650

700

750

800

850

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

kWh

Baseline

Adjusted

TOTAL END-USE INTENSITY PROJECTIONS

7,000

7,200

7,400

7,600

7,800

8,000

8,200

8,400

2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022

kWh

Baseline

Adjusted

8.00

8.50

9.00

9.50

10.00

10.50

11.00

2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

kWh

per s

qftBaseline

Adjusted

Residential

Commercial

RESIDENTIAL AVERAGE USE COMPARISON

5,000

5,500

6,000

6,500

7,000

7,500

8,000

8,500

9,000

9,500

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

kWh

New England

Vermont (with DSM)

New England Vermont2003 - 13 -0.4% -0.9%2013 - 23 0.0% -0.9%

DEVELOPING THE “NO DSM” FORECAST

» Adjust end-use intensities upwards – add back in 100% of projected DSM savings• Assume no future DSM is embedded in the baseline forecast

1,000

1,200

1,400

1,600

1,800

2,000

2,200

2,400

2,600

GWh

1,000

1,200

1,400

1,600

1,800

2,000

2,200

2,400

GWh

Residential Commercial

No DSM With DSM2003 - 2013 -0.2% -0.2%2013 - 2032 1.0% -0.1%

No DSM With DSM2003 - 2013 0.3% 0.3%2013 - 2032 1.5% 0.5%

No DSMNo DSM

RESIDENTIAL AVERAGE USE

No DSM

With DSM

No DSM With DSM2003 - 2013 -1.0% -1.0%2013 - 2032 0.4% -0.9%

20,000

25,000

30,000

35,000

40,000

45,000

50,000

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

kWh

COMMERCIAL AVERAGE USE

No DSM

With DSM

No DSM With DSM2003 - 2013 -1.0% -1.0%2013 - 2032 0.5% -0.5%

VERMONT ENERGY REQUIREMENTS

4,000

4,500

5,000

5,500

6,000

6,500

7,000

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

GWh

No DSM

With DSM

No DSM With DSM2003 - 2013 -0.4% -0.4%2013 - 2032 1.0% 0.0%

500

600

700

800

900

1,000

1,100

1,200

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

MW

VERMONT PEAK DEMAND

No DSM

With DSM

No DSM With DSM2003 - 2013 0.4% 0.4%2013 - 2032 1.0% -0.1%

By 2023, DSM programs reduce peak demand by 119 MW (10%)

UPDATING END-USE SATURATION ESTIMATES

» Incorporate new appliance saturation survey information• NMR 2012 Residential Home Survey (Reported February

2013)

» Also have the KEMA 2005 survey, earlier BED survey, and the EIA estimates for New England

» We were hoping to connect the dots between the KEMA and NMR survey• Problem: The two survey points are not all that consistent

ELECTRIC HEAT

KEMANMR

Surveys show consistently lower shares than that implied by the billing data

0

200

400

600

800

1,000

1,200

1,400

1,600

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

kWh/

cust

omer

EH_AvgUse NEH_AvgUse

ELECTRIC HEAT CUSTOMER SHARE AND USE

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

16.0%

18.0%

20.0%

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

EH_CustShare

There appears to be significant electric heating

SECONDARY ELECTRIC HEAT

NMR

KEMA

Could be people with electric heat consider it a back-up source

CENTRAL AIR CONDITIONING

NMR

KEMA 2 out of 3 sources say CAC saturation is higher

ROOM AIR CONDITIONING

NMRKEMA

Reported RAC saturation is about the same butnot consistent with the BED saturation growth

ELECTRIC WATER HEATING

NMR

KEMA

KEMA and NMR diverge from New England water heating saturation trends

SECOND REFRIGERATORS

NMR

KEMA

NMR saturation is more consistent with New England

FREEZERS

NMR

KEMA

The two surveys imply an unlikely sharp decline in the number of freezers

CLOTHES WASHERS

NMRKEMA

DRYERS

NMR

KEMA

NMR survey closer to that reported by EIA

DISH WASHERS

NMR

KEMA

KEMA survey closer to that reported by EIA

PV MARKET PENETRATION» Though still relatively small net metering (particular residential

and commercial photovoltaic systems) has been growing quickly

Residential

Commercial

Share of customers with PV systems(GMP solar customer data)

Through 2012

FACTORS DRIVING NET METERING PENETRATI0N

» Federal Investment Incentives – Through 2017• 30% of system installed cost

» State Investment Incentives • Was $0.45 per watt now $0.25 per watt• On a 5 kW system incentive drops from $2,250 to $1,250

- But federal absolute dollar credit will offset some of this» High avoided electric rates

• average residential rate $0.16 per kWh» Generous buy-back rate

• Credit equal to $0.20 per kWh (retail rate plus $0.04 per kWh)

SOLAR LOAD IS HAPPENING WHERE STATE INCENTIVES ARE SIGNIFICANT

-

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

SHOULD SEE STRONG PV MARKET GROWTH

$-

$1.00

$2.00

$3.00

$4.00

$5.00

$6.00

$7.00

$8.00

$9.00

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022Assumes 10% annual decline in system cost

Simple Payback

Assumes continuation of federal tax credit

RESIDENTIAL SIMPLE PAYBACK MODELVariable Coefficient T-Stat P-ValueCONST 0.0008 1.886 6.66%Payback -0.00005 -1.673 10.22%LagShare 0.97447 36.929 0.00%

Adjusted Observations 44Deg. of Freedom for Error 40R-Squared 0.998Adjusted R-Squared 0.997Mean Abs. % Err. (MAPE) 2.30%Durbin-Watson Statistic 1.758

RESIDENTIAL PV MARKET PENETRATION FCST

Base Case

No Federal Tax Credit

No State or Federal Tax Credit

RESIDENTIAL AND SOLAR CUSTOMER FORECAST

Year ResCusts chg SolarCusts chg Shr Solar2012 308,967 2,023 0.7%2013 312,543 1.2% 2,816 39.2% 0.9%2014 314,006 0.5% 3,530 25.4% 1.1%2015 317,079 1.0% 4,193 18.8% 1.3%2016 320,383 1.0% 4,813 14.8% 1.5%2017 323,462 1.0% 5,386 11.9% 1.7%2018 326,240 0.9% 5,909 9.7% 1.8%2019 328,748 0.8% 6,385 8.1% 1.9%2020 331,226 0.8% 6,819 6.8% 2.1%2021 333,468 0.7% 7,214 5.8% 2.2%2022 335,488 0.6% 7,570 4.9% 2.3%

RESIDENTIAL GENERATION AND USEAverage kWh per month

Generated6,700 kWh per year

Billed Average Use6,680 kWh per year

Own Use5,200 kWh per year

RESIDENTIAL SOLAR LOAD FORECAST

Capacity estimated using a 0.17 load factor, assume average system size = 4.5 Mw

Year Generated (MWh) Used (GWh) Capacity (MW)2012 13,554 10,520 9.1 2013 18,867 14,643 12.7 2014 23,651 18,356 15.9 2015 28,093 21,804 18.9 2016 32,247 25,028 21.7 2017 36,086 28,007 24.2 2018 39,590 30,727 26.6 2019 42,780 33,202 28.7 2020 45,687 35,459 30.7 2021 48,334 37,513 32.5 2022 50,719 39,364 34.1